COMPUTATIONAL FILTERING OF METHYLATED SEQUENCE DATA FOR PREDICTIVE MODELING

Abstract

Computational techniques are disclosed for using methylation profiles to classify the medication condition of a person. Initial sequence data is obtained containing sequences of an initial set of nucleic acids from a biological sample of a person. The initial sequence data is filtered to generate filtered sequence data that describes sequences of a filtered subset of nucleic acids from the biological sample. A methylation profile is determined for the filtered subset of nucleic acids from the biological sample. The methylation profile can be processed to determine a likelihood that the person has the specified medical condition. The system outputs an indication of the likelihood that the person has the specified medical condition.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims the benefit of priority to U.S. Application Ser. No. 62/832,157, filed on Apr. 10, 2019, U.S. Application Ser. No. 62/882,215, filed on Aug. 2, 2019, U.S. Application Ser. No. 62/928,156, filed on Oct. 30, 2019, U.S. Application Ser. No. 63/007,204, filed on Apr. 8, 2020, U.S. Application Ser. No. 63/007,208 filed on Apr. 8, 2020, and U.S. Application Ser. No. 63/007,218, filed on Apr. 8, 2020. The disclosures of each of these applications are considered part of the disclosure of the present document, and are each incorporated by reference in their entireties.

GOVERNMENT FUNDING STATEMENT

This invention was made with government support under grant number HD068578 awarded by the National Institutes of Health. The government has certain rights in the invention.

BACKGROUND

1. Technical Field

This document relates to systems and methods for classifying a condition of a mammal (e.g., a human) using predictive models (e.g., machine-learning models) that process methylation patterns of DNA obtained from a biological sample. Certain implementations of the techniques described herein employ computational methods to perform filtering operations on an input data set that can improve efficiency of the prediction or classification task while increasing model sensitivity.

2. Background Information

Machine-learning involves the analysis of data samples to identify features and patterns in the data samples that can be employed to perform tasks such as classification and prediction without explicit instructions. Some machine-learning techniques determine a model for performing the desired task, such as a neural network, a regression model, a decision tree, a support vector machine, and a naïve Bayes machine.

Methylation patterns in a mammal's DNA have been correlated with certain medical conditions or phenotypes of the mammal. However, data sets describing DNA sequences and/or methylation patterns are typically quite large and can be computationally expensive to process.

SUMMARY

This document describes systems, methods, devices, and other techniques for training and applying models to classify a medical condition of a person or other mammal based on methylation patterns occurring in DNA sequences of the person. In some aspects, the disclosed techniques employ a filtering operation to enrich a subset of sequences represented in an initial data set based on methylation characteristics and/or copy number characteristics of the nucleic acids. The filtering operation can, in some embodiments, achieve advantages including reducing the size of the input data set (e.g., a methylation profile) provided to the classifier (e.g., a machine-learning model), decreasing the time and computational expense required to process the input data set, and/or improving the sensitivity of the model to patterns that have the highest predictive power in differentiating persons with normal from abnormal medical conditions.

In some aspects, the disclosed techniques can involve identifying a set of reference CpG sites of abnormal individuals. Computing systems can estimate either restricted reference component methylomes or mixture methylomes that are independent linear combinations of certain reference component methylomes. The proportions of these components at the reference CpG sites for the tested biological samples can further be estimated, and the system can then predict the methylation level of the tested biological samples at a target set of CpG sites under the hypothesis that the sample is from a normal individual. The predicted methylation levels can then be compared against the observed methylation levels, and a classification for the individual as either exhibiting a normal or abnormal condition with respect to the specified medical condition can then be determined.

Further implementations of the disclosed subject matter include methods performed by a computing system. The system can obtain initial sequence data that describes sequences of an initial set of nucleic acids from a biological sample of a person, the initial set of nucleic acids including nucleic acids originating from multiple different tissues of the person. The system can filter the initial sequence data to generate filtered sequence data that describes sequences of a filtered subset of nucleic acids from the biological sample. Filtering can include (i) selecting target nucleic acids from the initial set of nucleic acids based on at least one of a methylation characteristic or a copy number characteristic of the target nucleic acids and (ii) enriching the target nucleic acids in the filtered subset. A methylation profile can be determined for the filtered subset of nucleic acids from the biological sample. The system processes the methylation profile for the filtered subset of nucleic acids to determine a likelihood that the person has a specified medical condition, and an indication of the likelihood that the person has the specified medical condition can then be provided as an output of the system.

These and other implementations can further include one or more of the following features.

The system can identify a pre-defined set of genomic regions (i.e., genomic loci). Selecting target nucleic acids from the initial set of nucleic acids can include comparing nucleic acid sequences from the initial set of nucleic acids to sequences from the pre-defined set of genomic regions. Enriching the target nucleic acids in the filtered subset can include discarding nucleic acid sequences from the initial sequence data that are not among the sequences from the pre-defined set of genomic regions, while retaining nucleic acid sequences from the initial sequence data that are among the sequences from the pre-defined set of genomic regions. At least a first subset of the pre-defined set of genomic regions can be defined based on those regions in the first subset exhibiting a minimum level of stability with respect to at least one of the methylation characteristic or the copy number characteristic in a population of individuals.

The biological sample can be plasma, and the initial set of nucleic acids can include cell-free DNA in the plasma.

The method can further include actions of identifying a set of restricted reference component methylomes in the initial set or filtered subset of nucleic acids; identifying a set of reference component methylomes; determining a proportion of the reference component methylomes at a reference set of CpG sites in the initial set or filtered subset of nucleic acids; generating predictions of methylation levels at a target set of CpG sites in the initial set or filtered subset of nucleic acids; comparing the predictions of methylation levels at the target set of CpG sites to observed methylation levels; and determining whether the person likely has or does not have the specified medical condition based on the comparison.

The biological sample can be a stool sample.

The biological sample can be cerebrospinal fluid.

The initial set of nucleic acids can be treated to facilitate detection of methylated sites before sequencing.

The specified medical condition can be ovarian cancer, endometriosis, necrotizing enterocolitis, fetal aneuploidy, preeclampsia, or a brain condition.

The methylation profile for the filtered subset of nucleic acids can indicate, for each of a set of multiple genomic loci, a methylation level of the locus. The genomic loci can be a CpG site, CpG island, differentially methylated region (DMR), promoter region, enhancer region, or CpG island shore.

Determining the likelihood that the person has the specified medical condition can include determining a probability that the person has the specified medical condition.

Determining the likelihood that the person has the specified medical condition can include generating a binary indication that the person either likely has the specified medical condition or likely does not have the specified medical condition.

Processing the methylation profile can include providing data representing the methylation profile as input to a machine-learning model, and obtaining the likelihood, or a value from which the likelihood is derived, as an output of the machine-learning model.

The machine-learning model can include at least one of a classifier, an artificial neural network, a support vector machine, a decision tree, or a regression model.

The machine-learning model can define reference or predicted methylation profiles against which the methylation profile for the filtered subset are compared to determine the likelihood that the person has the specified medical condition.

The determined likelihood that the person has the specified medical condition can be used by a medical provider to assess whether to perform additional diagnostic testing on the person.

The determined likelihood that the person has the specified medical condition can be used by a medical provider to at least one of diagnose the person or treat the person for the specified medical condition.

Outputting the indication of the likelihood that the person has the specified medical condition can include at least one of presenting the indication on an electronic display, audibly playing the indication through a speaker, storing the indication in a memory of a computing system for subsequent retrieval, or transmitting the indication in an electronic message to one or more users.

Enriching the target nucleic acids in the filtered subset can include generating the filtered subset so that a fraction of the target nucleic acids that occur in the filtered subset is greater than a fraction of the target nucleic acids that occur in the initial set of nucleic acids.

The filtered subset can consist exclusively of the target nucleic acids. Alternatively, the filtered subset can include both the target nucleic acids and non-targeted nucleic acids.

Some implementations include yet another method performed by a computing system. The method can include actions of obtaining initial sequence data that describes sequences of an initial set of nucleic acids from a biological sample of a person, the initial set of nucleic acids including nucleic acids originating from a multiple different tissues of the person; filtering, by the computing system, the initial sequence data to identify a first subset of sequences from the initial sequence data that correspond to a first pre-defined set of genomic regions; filtering, by the computing system, the initial sequence data to identify a second subset of sequences from the initial sequence data that correspond to a second pre-defined set of genomic regions; processing, by the computing system, data that includes an observed methylation profile of the first subset of sequences to generate a predicted methylation profile of the second subset of sequences; comparing, by the computing system, an observed methylation profile of the second subset of sequences to the predicted methylation profile of the second subset of sequences to determine whether the person has a specified medical condition, wherein the person is deemed to have the specified medical condition if a difference between the observed methylation profile of the second subset of sequences and the predicted methylation profile of the second subset of sequences meets a minimum difference criterion; and outputting, by the computing system, an indication of whether the person was determined to have the specified medical condition.

These and other implementations can further include one or more of the following features.

The first pre-defined set of genomic regions can be regions that exhibit a minimum level of stability with respect to at least one of a methylation characteristic or a copy number characteristic in a population of individuals. The second pre-defined set of genomic regions can be regions that exhibit a minimum difference with respect to at least one of the methylation characteristic or the copy number characteristic between a first sub-population of individuals who have the specified medical condition and a second sub-population of individuals who do not have the specified medical condition.

The first pre-defined set of genomic regions can be a first reference set of genomic regions, and the second pre-defined set of genomic regions can be a first target set of genomic regions. The actions can further include selecting the first reference set of genomic regions as the first pre-defined set of genomic regions from a database that includes multiple reference sets of genomic regions, wherein different ones of the multiple reference sets of genomic regions correspond to different medical conditions; and selecting the first target set of genomic regions as the second pre-defined set of genomic regions from the database, wherein the database further includes a multiple target sets of genomic regions, wherein different ones of the multiple target sets of genomic regions correspond to different medical conditions.

The specified medical condition is preeclampsia, endometriosis, ovarian cancer, necrotizing enterocolitis, or a brain condition.

Some implementations include yet another method performed by a computing system. The method can include actions of obtaining, by a computing system, initial sequence data that describes sequences of an initial set of nucleic acids from a biological sample of a person, the initial set of nucleic acids including nucleic acids originating from multiple different tissues of the person; filtering, by the computing system, the initial sequence data to identify a target subset of sequences from the initial sequence data that correspond to a pre-defined set of genomic regions; comparing, by the computing system, an observed methylation profile of the target subset of sequences to a pre-defined methylation profile to determine whether the person has a specified medical condition, wherein the person is deemed to have the specified medical condition if a difference between the observed methylation profile of the target subset of sequences and the pre-defined methylation profile meets a minimum difference criterion; and outputting, by the computing system, an indication of whether the person was determined to have the specified medical condition.

Additional aspects of the disclosed subject matter includes a computing system having one or more processors and one or more computer-readable media having instructions stored thereon that, when executed by the one or more processors, cause the one or more processors to perform actions of any of the methods/processes disclosed herein. Further aspects include one or more computer-readable media having instructions stored thereon that, when executed by one or more processors, cause the one or more processors to perform actions of any of the methods/processes disclosed herein.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Although methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, suitable methods and materials are described below.

Other features and advantages of the invention will be apparent from the following detailed description, and from the claims.

BRIEF DESCRIPTION OF DRAWINGS

FIG. 1 depicts a computing environment adapted for processing of sequence data to classify a patient as likely or not likely exhibiting a specified medical condition.

FIG. 2 is a functional illustration of a process for classifying a person as either exhibiting a specified medical condition or not based on methylation patterns in nucleic acids recovered from biological samples.

FIG. 3 is a depiction of a filtering operation that results in increasing the fraction of sequences for target nucleic acids in a set of sequence data.

FIG. 4 illustrates a set of methylation patterns on multiple fragments of cell-free DNA of an individual, along with an example of a bisulfite treatment process.

FIG. 5 is a flowchart of an example process for training and using a machine-learning model to determine whether a specified medical condition of a person or other mammal is or is not normal.

FIG. 6 is a flowchart of an example process for processing sequence data for nucleic acids from a biological sample to determine a methylation profile of the sample and classify a person with respect to the specified medical condition.

FIG. 7 is a flowchart of an example process for training a machine-learning model to generate, from a methylation profile of a patient, a classification result indicating whether a specified medical condition of the patient is or is not normal.

FIG. 8 depicts a block diagram of computing devices that can be used in some embodiments to implement aspects of the techniques disclosed herein.

FIGS. 9A-9G displays the distribution of CpG DNA methylation in plasma samples obtained from pregnant (blue) and non-pregnant (red) women. (FIG. 9A) All CpG sites; (FIG. 9B) CpG sites in promoters; (FIG. 9C) CpG sites in introns; (FIG. 9D) CpG sites in exons; (FIG. 9E) CpG sites in CpG island shores; (FIG. 9F) CpG sites in CpG islands; and (FIG. 9G) CpG sites in enhancers. x axis: % DNA Methylation; y axis: Density.

FIG. 10 provides a hierarchical clustering of plasma methylation sequencing data obtained from pregnant (PL) and non-pregnant (EN) women.

FIG. 11 provides chromosome-specific plots of spatial DNA methylation distributions between DNA methylation in plasma samples obtained between 10-13 weeks from pregnant women (blue) versus non-pregnant (red) women.

FIGS. 12A-12F shows spatial distribution of methylation levels in defined genomic elements on chromosome 19 in cfDNA from pregnant (blue) and non-pregnant female plasma (red). (FIG. 12A) Promoters; (FIG. 12B) Exons; (FIG. 12C) Introns; (FIG. 12D) CpG Island Shores; (FIG. 12E) Enhancers; and (FIG. 12F) CpG Islands.

FIGS. 13A-13G displays the distribution of DNA methylation in chorionic villus sampling (CVS) samples (red) and maternal leukocyte samples (blue) obtained from pregnant women between 10-13 weeks gestation. (FIG. 13A) All sites; (FIG. 13B) Sites in promoters; (FIG. 13C) Sites in introns; (FIG. 13D) Sites in exons; (FIG. 13E) Sites in CpG island shores; (FIG. 13F) Sites in CpG islands; and (FIG. 13G) Sites in enhancers.

FIG. 14 displays the genomic element-specific distribution of CpG DNA methylation in CVS samples (red) and maternal leukocyte samples (blue) obtained from pregnant women between 10-13 weeks gestation.

FIG. 15A-15B provides a heatmap of differentially methylated CpG sites in maternal plasma collected between 10-13 weeks gestation from pregnant women who later developed severe preeclampsia (SPE)<32 versus normal controls (FIG. 15A) or mild preeclampsia (MPE) versus normal controls (FIG. 15B).

FIG. 16 shows the DNA methylation patterns in plasma samples obtained at 10-13 weeks gestation from women who went on to develop severe preeclampsia (red) or mild preeclampsia (blue) compared to normal controls (black). Methylation fraction displayed on the y-axis and the genomic coordinates for chromosome 1 are displayed on the x-axis.

FIG. 17 shows spatial distribution of methylation levels on chromosome 19 in cfDNA samples from pregnant (blue) and non-pregnant female plasma.

FIGS. 18A-18X shows the influence of tissue-specific DNA methylation changes between chorionic villus (CV; red) and maternal leukocytes (MBC; blue) on DNA methylation in plasma from pregnant (dashed red) and non-pregnant (dashed blue) women at specific loci. FIGS. 18A and 18B=RUNX3; FIGS. 18C and 18D=SOX8; FIGS. 18E and 18F=GLI2; FIGS. 18G and 18H=ADGRA1; FIGS. 18I and 18J=ARHGEF16; FIGS. 18K and 18L=MAF; FIGS. 18M and 18N=TBX15; FIGS. 18O and 18P=TFAP2B; FIGS. 18Q and 18R=VGLL4; FIGS. 18S and 18T=FOXP1; FIGS. 18U and 18V=HHEX; FIGS. 18W and 18X=BCAT1.

FIG. 19 shows CV tissue is broadly hypomethylated relative to maternal leukocytes based upon unbiased genome-wide microarray analysis (Chu et al., PLoS One. 2011; 6:e14723).

FIGS. 20A-20B show the presence of PMDs in CV tissue (FIG. 20A) relative to maternal leukocytes (FIG. 20B) is illustrated on a genome-wide level. 1B. Beta represents DNA methylation rate (0-1.0 scale) (Chu et al., PLoS One. 2011; 6:e14723).

FIGS. 21A-21C show that, for each of FIG. 21A, FIG. 21B, and FIG. 21C, Top panel=moving average of the hypomethylation levels of CpG sites levels in CV (solid line) and MBC (dashed line). Middle panel=moving average of the difference in the hypomethylation levels of CpG sites between CV and MBC. The short dense vertical lines above the X axis (appearing as a solid horizontal line) in both the top and bottom panels represent locations of the MspI sites in each chromosome. Bottom: Histogram of the EST and mRNAs aligned to the chromosome generated using NCBI genome Map Viewer. (FIG. 21A) Chromosome 13; (FIG. 21B) Chromosome 18; (FIG. 21C) Chromosome 21.

FIG. 22 shows representative region of chromosome 1 (red box) showing statistically significant differences in CpG methylation rate between CVS and maternal leukocytes. Sites that are hypomethylated in CV relative to MBC are shown in red below the x axis whereas sites hypomethylated in maternal leukocytes relative to CV are shown in blue above the x axis.

FIG. 23 shows the relationship between genomic context of CpG location and differential methylation in chorionic villus.

FIGS. 24A-24D show pre- (FIGS. 24A and 24C) and post- (FIGS. 24B and 24D) LCM of neonatal gut epithelial cells. Representative (n=2) samples.

FIG. 25 shows hierarchical clustering of DNA methylation patterns determined by WGBS following LCM of neonatal gut epithelial cells.

FIG. 26A-26C show plots of fragment size against read count for plasma (EN1, FIG. 26A) and CSF samples (LP372, FIG. 26B; LP374, FIG. 26C), confirming that cell cfDNA fragments from CSF are significantly shorter in length than those from plasma.

FIG. 27 shows plot of fragment size distribution for cfDNA from CSF (LP372; red and LP374; blue) and plasma (EN1; green). CSF cfDNA exhibits a periodicity of fragment density peaks differing in size by approximately 10 bp.

FIG. 28A shows relative distribution of low methylation (<20%), intermediate methylation (20-80%) and high methylation (HM) groups in cfDNA from plasma and CSF.

FIG. 28B shows distribution of DNA methylation in CSF and plasma samples adjusted for specific categories of genomic element (intergenic, intron, exon or promoter).

FIG. 28C shows distribution of DNA methylation in CSF and plasma samples adjusted for specific categories of genomic element (CGI and CGI shore).

FIG. 29 provides sliding windows analysis identifying genomic regions (250 bp) whose CpG methylation characteristics differ significantly between cfDNA from CSF and plasma.

FIG. 30 provides the 100-most differentially methylated loci identified in which cfDNA from CSF is methylated at a lower rate than that of plasma.

FIG. 31 provides the 100-most differentially methylated loci identified in which cfDNA of CSF is methylated at a higher rate than that of plasma.

FIG. 32 provides differentially methylated regions that overlap with genes whose expressions are high in the brain and low in whole blood.

FIG. 33 shows the results of a method of detecting fetal aneuploidy in accordance with a CM model, using all CpG sites.

FIG. 34 shows the results of a method of detecting fetal aneuploidy in accordance with a CM model, using CpG sites differentially methylated between CVS and MBC.

FIG. 35 shows the results of a method of detecting fetal aneuploidy in accordance with a CnP model, using all CpG sites.

FIG. 36 shows the results of a method of detecting fetal aneuploidy in accordance with a CnP model, using CpG sites differentially methylated between CVS and MBC.

FIG. 37 shows the results of a method of detecting fetal aneuploidy in accordance with a PnP model, using all CpG sites.

FIG. 38 shows the results of a method of detecting fetal aneuploidy in accordance with a PnP model, using CpG sites differentially methylated between CVS and MBC.

FIG. 39. Hierarchical clustering of CpG sites identified by bisulfite sequencing of plasma DNA from women who had or had not previously had a hysterectomy.

FIG. 40. Pre- (Panels A and C) and post-(Panels B and D) LCM of neonatal gut epithelial cells. Representative samples are shown (n=2).

FIG. 41. Unsupervised clustering of whole genome sequencing data from laser captured NEC (green) and control (blue) colonic epithelium.

FIG. 42A. Hierarchical clustering of CpG sites identified by whole-genome bisulfite sequencing (WGBS) that were differentially methylated between laser captured gut epithelium from NEC colon (NEC) and control (Ctrl) samples.

FIG. 42B. Hierarchical clustering of CpG sites identified by WGBS that were differentially methylated between laser captured gut epithelium from NEC ileum (NEC) and control (Ctrl) samples.

FIG. 43A. Classical multi-dimensional scaling of targeted genome-wide bisulfite sequencing data from histological sections of NEC (blue) and control (green) colonic epithelium.

FIG. 43B. Classical multi-dimensional scaling of targeted genome-wide bisulfite sequencing data from histological sections of NEC (blue) and control (green) ileal epithelium.

FIG. 44. Hierarchical clustering of CpG sites identified by targeted whole-genome bisulfite sequencing that were differentially methylated between whole blood samples from NEC-affected neonates (NEC) and control individuals (Ctrl).

FIG. 45. Hierarchical clustering of CpG sites identified by targeted whole-genome bisulfite sequencing that were differentially methylated between stool samples from NEC-affected neonates (NEC) and control individuals (Ctrl).

FIG. 46. Differences and overlap between NEC-specific differentially methylated CpG sites identified in blood and stool. CpGs hypermethylated in NEC versus control (designated “up”) are labeled in red. CpGs hypomethylated in NEC versus control (designated “down”) are labeled in blue.

FIG. 47 depicts density plots showing the distribution of DNA methylation across the genome in distinct genomic elements including (A) all sites, (B) introns, (C) exons, (D) enhancers (E) CpG islands, (F) promoters, (G) CpG island shores. Data obtained from the plasma cfDNA of ovarian cancer patients (red) and normal controls (blue) are shown. In each case, the blue line is the higher line at the second peak. X axis=density. Y axis=% DNA methylation.

DETAILED DESCRIPTION

FIG. 1 depicts a computing environment 100 adapted for processing of sequence data to classify a patient as likely or not likely exhibiting a specified medical condition. The environment 100 is configured to train machine-learning model(s) corresponding to one or more medical conditions, and to apply the trained machine-learning models in classification tasks that facilitate screening, diagnosis, and/or treatment of medical conditions of a patient. In some examples, the computing environment 100 allows for improved efficiency and accuracy in screening for conditions such as pre-eclampsia, ovarian cancer, endometriosis, necrotizing enterocolitis, fetal aneuploidy, and/or certain brain or nervous system disorders, through analysis of sequence data derived from biological samples obtained from a patient via minimally invasive techniques (e.g., plasma or stool samples). The computing environment 100 can further employ filtering engines that filter an initial batch of sequencing data to reduce the universe of nucleic acids analyzed when classifying a given medical condition, thereby improving sensitivity of the classifier, while reducing model size and reducing the number of operations required to generate a classification result.

Environment 100 includes a sample analyzer 104 that processes and sequences biological samples 150 from a person. In some implementations, the system is configured to perform “liquid biopsies” on liquid-based biological samples 150 from a person, e.g., plasma samples, stool samples, saliva samples, cerebrospinal fluid samples, urine samples, or cervical swab samples. The plasma (or other biological samples) may include cell-free DNA from the person (and, in some cases, cell-free DNA from a fetus if the person is a pregnant female). Cell-free DNA can originate from various tissues in a person's body. For example, cell-free DNA in a biological sample 150 can include fragments of DNA that were released from cells as a result of processes such as active secretion, necrosis, apoptosis, or a combination of these. The level of cell-free DNA in a sample 150 originating from certain tissue(s) can be correlated with a given medical condition (e.g., disease). Moreover, the methylation patterns of cell-free DNA originating from certain tissue(s), and which occur in the biological sample 150, can be correlated with a medical condition (e.g., disease) of the person. To analyze the level of cell-free DNA or other nucleic acids associated with a specified medical condition, and to analyze the methylation patterns of these nucleic acids, the sample analyzer 104 may process the biological sample 150 to generate initial sequence data for the fragments of extracellular DNA and/or other nucleic acids in the sample 150.

The initial sequence data describes sequences of nucleotides occurring in nucleic acids from the sample 150, e.g., sequences of bases along fragments of cell-free DNA. Typically, the initial sequence data includes sequence descriptions for both targeted nucleic acids and background (non-targeted) nucleic acids in the biological sample 150. The targeted nucleic acids are those that are deemed significant to detection of a specified medical condition, while the background nucleic acids are deemed insignificant or less significant to detection of the specified medical condition. The mixture of targeted and background nucleic acids may vary for different medical conditions. For example, some nucleic acids may be classified in the target set for detection of ovarian cancer, while those same acids may be classified in the background set for detection of necrotizing enterocolitis. In some implementations, the set of target nucleic acids that are deemed significant to a particular medical condition are those originating from particular tissue(s) affected by a specified medical condition. For instance, the target nucleic acids associated with endometriosis may include or consist of cell-free DNA from the endometrium or uterus. Likewise, the target nucleic acids associated with necrotizing enterocolitis may be based on intestinal tissue. Often, the fraction of targeted nucleic acids in the biological sample is small in relation to the fraction of background nucleic acids, and the fraction of sequences reflected in the initial sequence data for targeted nucleic acids may also be small in relation to sequences of background nucleic acids. As a result, the initial sequence data may contain substantial levels of “noise” relative to the signals present in the sequences for targeted nucleic acids, which can degrade the performance of models in predicting whether or not a patient likely has a specified medical condition.

The sample analyzer 104 is configured to sequence nucleic acids in the biological sample 150 using any suitable technique, including polymerase chain reaction (PCR)-based methods such as droplet digital PCR, or next-generation sequencing (NGS). In some implementations, nucleic acids from the biological sample 150 undergo bisulfite conversion before sequencing in order to facilitate subsequent detection of methylated sites. Bisfulite treatment has the effect of converting unmethylated cytosines (C) to uracil (U), which in turn are converted to thymine (T) in the course of DNA amplification. In contrast, bisulfite treatment does not affect methylated cytosines (C). As a result, bisulfite conversion enables differentiation of methylated from non-methylated cytosines, which appear as different bases in the sequencing data. Methylation arrays, methylation-specific PCR, enrichment, and/or additional methods may also or alternatively be applied.

Sample analyzer 104 can output initial sequencing data for a biological sample 150 for receipt by a user's computing system 102. The user's computing system 102 can comprise one or more computers in one or more locations. System 102 can be, for example, a desktop computer, notebook computer, tablet computer, or smartphone. System 102 includes a network interface for communicating over one or more networks 106 such as a local area network (LAN), a wireless LAN (WLAN), the Internet, or a combination of these. System 102 may include peripherals such as a keyboard, pointing device, and display screen to enable user interaction with the system 102. In some implementations, the user may coordinate activities at the system 102 for obtaining a classification result related to a specified medical condition based on sequencing data from biological sample 150. For example, upon obtaining initial sequencing data from sample analyzer 104, the user may instruct system 102 to send a classification request 152 to classification system 120. Classification system 120 can comprise one or more computers in one or more locations. System 120 may be located remotely from the user's system 102, or may be located at the same premises on system 102. In some implementations, the capabilities of systems 102, 120, 110, or any two of these, are consolidated in a single, integrated system. In general, classification system 120 processes a classification request 152 and returns a classification result 158 that indicates a predicted likelihood that the person who provided biological sample 150 either has or does not have a specified medical condition (e.g., preeclampsia, necrotizing enterocolitis, endometriosis, ovarian cancer, fetal aneuploidy, an abnormal brain condition, or others).

In more detail, classification system 120 can include a package selector 122, filtering engine 124, methylation profiler 126, and model evaluator 128. Each of these components 122, 124, 126, and 128 may be implemented on one or more computers using a combination of software, hardware, or firmware. Package selector 122 is operable to select, from a library of medical-condition packages 140a-n, a particular package 140 that corresponds to the medical condition specified in classification request 152. Each package 140a-n defines information or instructions usable by classification system 120 to generate a classification result as to a different medical condition. In some implementations, the package 140 for a given medical condition can include one or more sequence filters 142, a loci list 144, and a machine-learning (ML) model 146, each of which is specific to the corresponding medical condition. Classification system 120 can load a retrieved package 140 to facilitate generation of a classification result 158 responsive to request 152. In some implementations, package selector 122 can select individual components of a package 140 as needed, such as filters 142, loci list 144, or ML model 146, to the exclusion of the others.

Sequence filters 142 provide information that enable a filtering engine 124 of the classification system 120 to filter initial sequence data by retaining sequences for target nucleic acids corresponding to the specified medical condition and discarding sequences for background (non-targeted) nucleic acids for the specified medical condition. In some implementations, filters 142 provides a whitelist that identifies target nucleic acids that should be retained in a filtering operation, such as cell-free DNA fragments that originate uniquely from a particular tissue associated with the specified medical condition. In some implementations, filters 142 provide a blacklist that identifies background nucleic acids that should be discarded in a filtering operation, such as cell-free DNA fragments that are not uniquely originated from a particular tissue associated with the specified medical conditions. In some implementations, filters 142 provide a whitelist that identifies target nucleic acids that should be retained (while other nucleic acids not specified in the whitelist are discarded) in a filtering operation, such as cell-free DNA fragments that are uniquely originated from a particular tissue associated with the specified medical condition. The whitelist, blacklist, or other information in filters 142 can define a set of genomic regions or sequences of nucleic acids against which sequences in the initial set of nucleic acids from the classification request 152 are compared to assess whether to retain or discard for further processing. For example, all nucleic acids in a plasma sample can be sequenced and the filtering can remove certain sequences identified in the blacklist. Alternatively, all nucleic acids in the sample can be sequenced and the filtering can discard everything but the sequences identified in a whitelist.

In some implementations, the filtering engine 124 is configured to perform a filtering operation that involves (i) selecting target nucleic acids from the initial set of nucleic acids based a methylation characteristic (e.g., methylation status), a copy number characteristic of the target nucleic acids, or both, and (ii) enriching the target nucleic acids, e.g., to increase a fraction of the target nucleic acids after filtering relative to their fraction in a pre-filtered set. In some examples, filters 142 identify a set of genomic regions in a whitelist. Selecting target nucleic acids from the initial set of nucleic acids can include comparing sequences of nucleic acids from the initial set of nucleic acids to sequences in the pre-defined set of genomic regions. Enriching the target nucleic acids in the filtered subset can include discarding sequences of nucleic acids from the initial set of nucleic acids that do not appear in the pre-defined set of regions, while retaining nucleic acids that do appear in the pre-defined set of regions. The pre-defined set of genomic regions can identify target nucleic acids having at least a minimum level of stability (e.g., minimum threshold stability) with respect to the methylation characteristic and the copy number characteristic, wherein the identified target nucleic acids originate from multiple different tissues. Additionally or alternatively, the pre-defined set of genomic regions can identify target nucleic acids originating from a subset of the multiple different tissues for which at least one of the methylation characteristic or the copy number characteristic differs by at least a minimum amount (e.g., minimum threshold) between individuals who have the specified medical condition and individuals who do not have the specified medical condition. In other examples, filters 142 identify a pre-defined set of genomic regions or sequences in a blacklist/exclude list, and corresponding operations can apply to select and enrich target nucleic acids except that the target nucleic acids are identified by discarding nucleic acids within the blacklist/exclude list.

Loci list 144 identifies the set of genomic loci whose methylation statuses are processed to generate a classification result with respect to a particular medical condition. The methylation profiler 126 within classification system 120 can use the loci list 144 to construct a methylation profile for a set of nucleic acids. For example, the loci list 144 may identify specific nucleotides, CpG sites, CpG islands, differentially methylated regions, promoter regions, enhancer regions, and/or CPG island shores whose methylation statuses can be processed to inform a classification with respect to the corresponding medical condition. In some implementations, the loci list 144 identifies genomic loci that occur only within the set of target nucleic acids for the medical condition. The loci list 144 can provide that the methylation statuses of all CpG sites within the target nucleic acids should be processed to generate a classification result. Alternatively, the loci list 144 can provide that the methylation statuses of only a subset of CpG sites within the target nucleic acids should be processed to generate a classification result. The subset of CpG sites (or other genomic loci) can be deemed the most statistically significant, or those that have the highest predictive power, for accurately classifying whether a patient has or does not have a specified medical condition. In some implementations, the loci list 144 identifies genomic loci that occur anywhere in the genome, regardless of whether the loci occur in target or background nucleic acid. The genomic loci in this embodiment may be processed without a separate filtering step that discards all or some of the background nucleic acid sequences, and the loci may have been identified as the most statistically significant, or those having the highest predictive power across the genome for accurately classifying whether a patient has or does not have a specified medical condition. The methylation profiler 126 can analyze the initial set of sequence data from the classification request 152 or the filtered set of sequence data from filtering engine 124 to determine the methylation status of all or some of the loci identified in list 144. The methylation status can be expressed in a number of ways, such as a binary value indicating whether the methylation level at a locus is above or below a pre-defined threshold, or a normalized value within a pre-defined range of values indicating a relative methylation level at the locus across multiple DNA fragments encompassing the locus.

As used interchangeably herein, “methylation state,” “methylation profile,” “methylation status,” and “methylation level” refer to the presence, absence, percentage, and/or quantity of methylation at a particular nucleotide, or nucleotides, within a DNA region, e.g., a genomic locus. The methylation status of a particular DNA sequence (e.g., a genomic locus) can indicate the methylation state of every nucleotide in the sequence, indicate the methylation state of any of the nucleotides (e.g., cytosines) in the sequence, can indicate the methylation state of a subset of the nucleotides (e.g., of cytosines), can indicate the percentage or fraction of methylated cytosines at any particular stretch of nucleotides within the sequence or can indicate the average rate of methylation of all the cytosines (or a subset of the cytosines) present in a nucleic acid.

As used herein, a “methylated nucleic acid molecule” refers to a nucleic acid molecule that contains one or more methylated nucleotides that is/are methylated.

As used herein, a “CpG site” or “methylation site” is a nucleotide within a nucleic acid that is susceptible to methylation either by natural occurring events in vivo or by an event instituted to chemically methylate the nucleotide in vitro. A “CpG island,” as used herein, describes a segment of a nucleic acid, e.g., DNA sequence, that have a high frequency of CpG dinucleotide repeats. A “CpG island shore,” as used herein, refers to methylation hotspots that are present a short distance, e.g., less than 2 kb, from CpG islands.

Machine-learning model 146 is a model that correlates methylation patterns from a methylation profile to likelihoods that a person has or does not have a specified medical condition. The machine-learning model 146 can be loaded by model evaluator 128 in classification system 120. Further details on training and applying machine-learning models are described with respect to FIGS. 5-7. For example, model evaluator 128 can provide the methylation profile from methylation profiler 126 as input to the machine-learning model 146, and can evaluate the machine-learning model 146 based on the methylation profile to generate a classification result 158 indicating whether the patient likely has or does not have the specified medical condition. The machine-learning model 146 can be implemented in a number of different forms. Typically, model 146 is trained using supervised learning techniques, and can be a linear regression model, logistic regression model, decision tree, support vector machine, naïve bayes model, random forest model, or artificial neural network (e.g., a feedforward neural network, deep neural network, recurrent neural network, and/or convolutional neural network), or a combination of two or more of these.

Environment 100 can further include a machine-learning system 110 for training the machine-learning models 146 in the library of packages 140a-n. System 110 can be implemented one or more computers in one or more locations, and may be accessible to the user's system 102 and classification system 120 via a direct connection or by connection over a network 106. In some implementations, the functionality of system 110 can be integrated with the user's system 102, classification system 120, or both. System 110 can include a training data receiver 112, training engine 114, and model provider 116. Receiver 112 is configured to receive training data 132 for training a machine-learning model. Different training data sets 132 can be used to train different models corresponding to different medical conditions. For example, a first training data set may be constructed for training a model that screens for pre-eclampsia, while a second training data set may be constructed for training a model that screens for endometriosis. The training data 132 can include a collection of training samples 130, each training sample 130 comprising (i) a set of filtered or unfiltered sequence data describing sequences of nucleic acids from a biological sample of a person and (ii) a label indicating whether or not the patient exhibits a specified medical condition. The label can serve as a target output for the model 146 when evaluated on a methylation profile derived from the sequence data in the training sample. Further details of a process for training a machine-learning model 146 is described below with respect to FIGS. 5 and 6. Training engine 114 receives the training data 132 from receiver 112 and processes this data 132 to train model 146. Trained model 116 can then be returned by model provider 116 and stored in a library in the package for the corresponding medical condition on which the model 116 was trained.

FIG. 2 depicts a functional illustration of a process 200 for classifying a person as either exhibiting a specified medical condition or not based on methylation patterns in nucleic acids recovered from biological samples such as plasma, cerebrospinal fluid, or stool. Filtering engine 124 receives initial sequence data 124, which contains information describing sequences of nucleic acids (e.g., cell-free DNA) from the biological sample. The initial sequence data 124 may include a relatively low fraction of sequences for target nucleic acids related to the specified medical condition and a relatively high fraction of sequences for background nucleic acids that are substantially unrelated to the medical condition. Filtering engine 124 performs a filtering operation on the initial sequence data 202 to generate filtered sequence data 204. Filtered sequence data 204 virtually enriches target nucleic acids in the specimen by discarding sequence data for all or some of the background nucleic acids. As a result, filtered sequence data 204 may include a higher fraction of sequences for the target nucleic acids than the fraction of the same sequences in initial sequence data 202. In some implementations, it is unnecessary for the filtering operation to completely distill the cohort of sequences to just those sequences for target nucleic acids. While complete distillation may be achieved in some cases, for many applications it is sufficient for the filtering operation to increase the fraction of sequences for the target nucleic acids relative to the fraction of sequences for the background nucleic to a threshold level such that the data is usable to generate a classification result with a minimally specified confidence level. Methylation profiler 126 then processes the filtered sequence data 204 to generate a methylation profile 206 that identifies a methylation status (e.g., methylation level) at each locus of a set of genomic loci within the target nucleic acids, or within a combination of target and background nucleic acids, represented in filtered sequence data 204. Machine-learning model evaluator 128 then uses a trained machine-learning model to process methylation profile 206 and generate a classification result 208 indicating the system's prediction as to whether the patient exhibits the specified medical condition.

Referring to FIG. 3, an illustration is shown of a filtering operation that results in increasing the fraction of sequences for target nucleic acids in a set of sequence data. The top bar in FIG. 3 represents a collection of initial sequence data, e.g., initial sequence data 202. The initial sequence data 202 includes a relatively small fraction 302 of target nucleic acid sequences and a relatively high fraction 304 of background (non-targeted) nucleic acid sequences. Due to the overwhelming presence of sequences for background nucleic acids in the initial data set 202, the significance of the sequences having greater predictive power for the target nucleic acids is diminished. After the filtering operation, however, a filtered sequence data set 204 results in a significantly higher fraction of target sequences 302 than occurs in the initial data set 202, and a lower fraction of non-target sequences 304.

FIG. 4 depicts a representation of a set of methylation patterns 404a-n on multiple fragments of cell-free DNA of an individual. Methyl groups (CH₃) bonded to the DNA at certain CpG sites form a methylation pattern on each fragment. In some implementations, methylation sites are detected by first treating the DNA with bisulfite to convert methylated cytosines to uracil, and interpreting the uracil sites as methylated sites in the original DNA strand. Other techniques for detecting methylated loci in nucleic acids may also be employed rather than or in addition to bisulfite conversion, although FIG. 4 illustrates a suitable method.

In some implementations, the systems and methods disclosed herein can be applied to generate a classification result indicating whether a patient likely does or does not have a specified medical condition according to the process 500 depicted in FIG. 5. The process 500 is based on the observation that the methylome(s) of certain tissue(s) in a person or other mammal can be affected by certain medical conditions that reflect diseases or abnormalities in the body (e.g., preeclampsia, endometriosis, necrotizing enterocolitis, or other conditions described herein), and that the changes of these methylomes can lead to changes in the methylation patterns of the DNA fragments found in a biological sample (e.g., a plasma, stool, urine, or saliva sample), which are released by these tissue(s). The term “methylome,” as used herein, refers to the amount or pattern of methylation at different sites or regions within a population of cells. The methylome can correspond to all of the genome, a subset of the genome (e.g., repeat elements in the genome), or a portion of the subset (e.g., those areas found to be associated with a specified medical condition). A methylome from plasma can be referred to a “plasma fluid methylome” or a “plasma fluid DNA methylome.” The plasma fluid methylome is an example of a cell-free methylome that includes cell-free DNA (cfDNA). A methylome from stool can be referred to a “stool fluid methylome” or a “stool fluid DNA methylome.” The stool fluid methylome is an example of a cell-free methylome that includes cell-free DNA (cfDNA).

The process 500 was developed to identify changes of methylation patterns in the methylome of a biological sample caused by phenotypes of certain tissues affected by the abnormal medical condition (e.g., intestinal tissue for necrotizing enterocolitis or uterine tissue for endometriosis). One insight behind this process 500 was that the methylome of the DNA fragments in these biological samples is a mixture of a variety of component methylomes, and that the proportion of these different component methylomes in the mixture varies from subject to subject, even among the population with normal tissue phenotype. By constructing a model methylome for a biological sample as a linear combination of various component methylomes, the process 500 can accurately predict the methylation patterns of a new biological sample under the hypothesis that it is from a normal individual. Consequently, the process 500 can exhibit high sensitivity in detecting abnormal methylation patterns in a biological sample caused by changes of the methylomes of some tissues (e.g., intestinal tissues) when the sample is from an affected individual. The process 500 can be performed by any of the computing systems described herein, such as systems 102, 110, and 120 shown in the environment of FIG. 1.

Let i be any CpG site in human genome, z_i,j be the methylation level of CpG site i in a biological sample j, p_i,r,j be the proportion of the r^th component methylome m_r,j of particular tissue origin in sample j at site i, m_i,r,j be the methylation level of CpG i in methylome m_r,j. The system models the scenario as follows:

z_i,j=Σ_r=1^Rp_i,r,jm_i,r,j  (1)

where p_i,r,j, m_i,r,j>=0, m_i,r,j<=1, p_i,1,j+ . . . +p_i,R,j=1.

The model assumes that there is a set of CpG sites S such that, for any CpG site i in S, and any biological sample of a particular type (e.g., plasma, stool, cerebrospinal fluid, saliva, or urine) j from a normal individual, it has m_i,r,j=m_i,r and p_i,r,j=p_r,j.

That is, the model assumes that in any biological sample from a normal individual, the proportions of different component methylomes in the mixture are the same for all CpG sites in S. The model further assumes that by restricting to the set of CpG sites S, biological samples from all normal individuals have the same set of component methylomes. They are called restricted reference component methylomes (RRCM), and are labeled as m₁^S, . . . , m_R^S or simply m₁, . . . , m_R when there is no confusion. For any biological sample j from a normal individual, its methylome restricted to set of CpG sites in S can be expressed as a weighted average of the restricted reference component methylomes. More precisely, let z_j^S be the methylome of biological sample C restricted to S, then for some mixture vector p_j=[p_j,1 . . . p_j,R]^T, it has:

z_j^s=[m₁^S, . . . ,m_R^S]p_j  (2)

The model also assumes that the set S is the union of two disjoint subsets C and T, where T is a union of K non-empty sets T_k such that T=U_k=1^K T_k where the index k represents the k^th type of abnormal tissue (e.g., intestinal tissue) phenotype. T_k's do not need to be disjoint. Moreover, T_k itself is the union of two disjoint sets D_k and V_k. Either D_k or V_k could be empty, but not both. It is assumed that for any biological sample, including one from an abnormal individual, when restricted to CpG sites in C, its methylome can always be expressed as a weighted average of the restricted reference component methylomes. That is, it has: z_j^C=[m₁^C, . . . , m_R^C]p_j regardless whether j is from an abnormal individual. C is called the set of reference CpG sites. On the other hand, for a biologic sample l from an abnormal individual, when restricted to CpG sites in S=CUT, its methylome can no longer be expressed as a weighted average of the restricted reference component methylomes. That is, it has: w₁^S≠[m₁^S, . . . , m_R^S]p_l for any mixture vector p_i. More specifically, for a biologic sample l from the k^th type of abnormal individual, it has: 1) w_j^C=[m₁^C, . . . , m_R^C]p_l, 2) if D_K is non-empty, then w_lDK=[m_1,k^Dk, . . . , m_R,k^Dk]p_l such that [m₁^Dk, . . . , m_R^Dk]≠[m_1,k^Dk, . . . , m_R,k^Dk], and 3) if V_k is non- empty, then w_l^Vk=[m₁^Vk, . . . , m_R^Vk]q_l such that p_l≠q_l. In other words, in a biologic sample from the k^th type of abnormal individual, if the set D_k is not empty, the component methylomes of the sample l restricted to D_k are no longer the same as the reference component methylome restricted to D_k. If the set V_k is not empty, in this biologic sample, the proportion of the reference component methylomes restricted to V_k is no longer the same as the proportion of the reference component methylome restricted to R.

T is called the target set of CpG sites, D_k is called the differential methylation target set, V_k is called the copy number variation target set, and T_k is called the target set for the k^th type of abnormal individual.

Certain operations of the process 500 are depicted in the flowchart of FIG. 5, which may be carried out by a system of one or more computers (e.g., systems 102, 110, and/or 120). First, the system identifies the sets of reference CpG sites C, and T₁, . . . , T_K for the list of K types of abnormal individuals (502). The system then estimates the restricted reference component methylomes m₁, . . . ,m_R, or R predictor methylomes n₁, . . . , n_R that are independent linear combinations of the reference component methylomes such that n_r=[m₁, . . . , m_R]q_r for R linearly independent mixture vectors q₁, . . . , q_R (504). If the reference component methylomes are available, the system estimates the proportions of these components at the reference CpG sites C for the test biologic samples (506). The system predicts the methylation level of the test biologic samples at the target set T_k of CpG sites, under the hypothesis that the sample is from a normal individual (508). Then, the predicted methylation levels at D_k and V_k are compared against the observed methylation levels, and the system rejects the null hypothesis that a test sample is from a normal individual if the observed methylation levels are significantly different from the predicted levels (510). In this manner, a classification result can be generated indicating whether the person or other mammal from whom the biologic sample was obtained either has the specified medical condition (abnormal) or does not have the specified medical condition (normal).

Process 500 of FIG. 5 can be implemented in a number of ways. For example, given the methyl-sequence data for a set of biologic samples from normal individuals, an expectation-maximization (EM) technique or data augmentation technique can be applied to estimate the component methylomes, and then the maximum likelihood method used to estimate the proportion of these component methylomes in the test sample. Below are certain implementations that employ linear regression.

In some implementations of the presently disclosed process 500, it is assumed the restricted methylome of a biologic sample from a normal individual can be approximated by a mixture of two restricted reference methylomes, one representing the DNA fragments from a first specific tissue region (e.g., intestinal tissue region for necrotizing enterocolitis), another representing the DNA fragments from a second specific tissue region. It is further assumed that the estimations of these two reference component methylomes are available. The implementation of the process 500 includes the following steps.

To begin, identify the reference set C, and the target sets T₁, . . . , T_K (502). First, collect the methylation data for a set of first cell type samples, a set of second cell type samples, and a set of biologic samples, all from normal individuals. For each type of abnormal individuals, collect a set of first cell type samples, a set of second cell type samples, and a set of biologic samples from that type of abnormal individuals. All these samples should have matched age, race, and other relevant parameters. These are the training data. Next, let x_i,j be the observed methylation level of CpG site i in a normal first cell type sample j, and y_i,l the observed methylation level of CpG site i in a normal second cell type sample l, s_x,i² the sample variance of x_i,j over all normal first cell type samples, s_y,i² the sample variance of y_i,j over all normal second cell type samples. Identify the CpG sites S₀ such that for any i∈S₀, it has both s_x,i²<c₀ and s_y,i²<c₀ for some constant c₀. These are CpG sites with stable methylation levels in each type of normal cells. Next, let x_i,j be the observed methylation level of CpG site i in a first cell type sample j, including normal and abnormal, and y_i,l the observed methylation level of CpG site i in a second cell type sample l, including normal and abnormal, s_x,i² the sample variance of x_i,j over all first cell type samples, including normal and abnormal, s_y,i² the sample variance of y_i,j over all second cell type samples, including normal and abnormal. Identify the CpG sites S₁ such that for any i∈S₁, it has both s_x,i²<c₀ and s_y,i²<c₀ for some constant c₀, and that the statistical test for the difference between {x_i,j0: j0 is a normal first cell type sample}, and {x_i,jk: jk is a first cell type sample of the kth abnormal phenotype}, is not significant for all abnormal phenotypes of first cell type, and that the statistical test for the difference between {y_i,j0: j0 is a normal second cell type sample} and {y_i,jk: jk is a second cell type sample of the kth abnormal phenotype} is not significant for all abnormal phenotypes of the second cell type. These are CpG sites with stable methylation levels in each type of cells, and with no difference in methylation level between normal and any abnormal samples. Let x_i be the sample mean of x_i,j over all first cell type samples, including normal and abnormal, y_i the sample mean of y_i,j over all second cell type samples, including normal and abnormal. Identify the subset C₀ of S₁ such that for any i∈C₀, it has |x_i−y_i|>c₁ for some constant c₁. These are CpG sites that are stably methylated in each cell type, with no difference between the normal and abnormal samples of the same cell type, and differentially methylated between different types of cells. Next, let x^R0 be the vector of x_i for all i∈C₀, and y^C0 be the vector of y_i for all i∈C₀, where x_i is the mean methylation at site i in all first cell type samples y_i the mean methylation at site i in all second cell type samples. Note that by the way the set C₀ is selected, there is no difference in the methylation level of any CpG sites in C₀ between normal and abnormal first cell type samples, or between normal and abnormal second cell type samples. Let z_j^C0 be the observed methylation levels of CpG sites in C₀ for a biologic sample j of the k^th abnormal type. (For convenience, the normal biologic sample is called as sample of the 0^th abnormal type). For each sample j belonging to the k^th abnormal type, regress z_j^C0 against x^C0 and y^C0, with the constraints that the intercept must be 0, and the coefficients must be non-negative and add to 1, and get the residual e_j^C0. Identify the subset C₀^k of C₀ such that for any CpG i in C₀^k, it has

$\frac{e_{i,k}^2}{s_{i,k}}<c_2,$

and e_i,k²<c₃ for some constants c₂ and c₃, where e_i,k² is the mean of the squared difference between estimated and observed methylation levels of CpG site i in all biologic samples of the k^th abnormal type, and s_i,k² the sample variances of methylation levels of CpG site i in the same set of biologic samples. Repeat the above procedure for each type of abnormal biologic samples, the intersection of the subsets C=∩_k=0^KC₀^k is the reference set of CpG sites. These are CpG sites where their methylation levels in both normal and any type of abnormal biologic samples can be accurately predicted by the reference component methylomes from normal individuals.

Next, let T₀=S₀\S₁. Let x^C and x^T0 be the vectors of x_i and x_h for all i∈C and h∈T₀ respectively, and y^C and y^T0 be the vectors of y_i and y_h for all i∈C and h∈T₀ respectively, where x_i, x_h, y_i, and y_h are mean methylation level of sites for a normal first cell type sample or second cell type sample at sites i and h respectively. Let z_j^C and z_j^T0 and be the observed methylation levels of CpG sites in C and T₀ respectively for a normal biologic sample j, w_lk^C and w_lk^T0 the observed methylation level of CpG sites in C and T₀ respectively for a biologic sample l_k from an individual with the k^th type of abnormality, w_lg^C and w_lg^T0 the observed methylation level of CpG sites in C and T₀ respectively for a biologic sample l_g from an individual with the g^th type of abnormality, where g≠k. For each j, l_k, and l_g, regress z_j^C, w_lk^C, and w_lg^C respectively against x^C and y^C, with the constraints that the intercept must be 0, and the coefficients must be non-negative and add to 1. Apply the fitted models respectively to x^T0 and y^T0 to predict z_j^T0, w_lk^T0, and W_lg^T0 respectively, and get the differences e_j^T0, e_lk^T0 and e_lg^T0 between the predicted values and observed values. Let e_i, e_i,k, and e_i,g be the means of the sets of differences {e_j^T0: j is a normal biologic sample}, {e_lk^T0: l_k is a biologic sample of the k^th abnormal type} and {e_lg^T0: l_g is a biologic sample of the g^th abnormal type} for CpG site i respectively. Identify the subset T_k of T₀ such that for any i∈T_k, it has |e_i|<c_2,0, |e_i,k|>c_2,k, and |e_i,k−e_i,g|>c_3,k, for some constants c_2,0, c_2,k, and c_3,k, for all g≠k. T_k is the target set for the k^th type of the abnormal individual. These are the sites where the methylation of a normal biologic sample can be accurately predicted, the observed methylation in a biologic sample of the k^th abnormal type will deviate from the prediction, and deviation will be different from that of a biologic sample of any other abnormal type

Next, the system estimates the fraction of the new biologic samples to be tested. Recall that x^c and y^c are mean vectors of the methylation levels of the training first cell type and training second cell type data for the CpG sites in the reference set C. For any new biologic sample t to be tested, let z_t^C be the observed methylation levels of CpG sites in C. Regress z_t^C against x^C and y^C, with the constraints that the intercept must be 0, and the coefficients must be non-negative and add to 1. The estimated coefficients for x^C are the estimated fractions of the two cell types for the biologic sample t.

The system can then test if the new biologic samples are from the k^th type of abnormal individual. For the new biologic sample (e.g., plasma) t, let x^Tk and y^Tk be mean vectors of the methylation levels of the training the first cell type data and the second cell type data for the CpG sites in the target set T_k identified in step 1 of this process 500, apply the fitted regression models obtained from the step 2 of this process 500 to x^Tk and y^Tk to predict the methylation levels of CpG sites in T_k for sample t under the hypothesis that sample t is from a normal. Let n_k be the number of CpG sites in T_k. Define functions f_k(x₁, . . . , x_nk)=Σ_i(−1)^{I_(ei,k−ei)}x_i and f_k,g(x₁, . . . , x_nk)=Σ_i(−1)^{I_(ei,k−ei,g)}x_i, where I_(⋅)=I_−∞,0)(⋅), that is, the indicator function for the interval (−∞, 0), e_i, e_i,k and e_i,g are estimations obtained from step 1.5 of the process 500. It will be said the sample is from an individual with the k^th type of abnormal phenotype if f_k(e_1,t−e₁, . . . , e_nk,t−e_nk)>c_4,k, and f_k,g(e_1,t−e_1,g, . . . , e_nk,t−e_nk,g)>c_5,g for all g≠k, where e_i,t is the difference between the observed methylation level of the CpG site i∈T_k for sample t and the predicted value by the fitted model obtained from step 2, and g≠k is any type of abnormal phenotype that is different form the k^th type of abnormal phenotype.

Other ways of implementing the process 500 can be developed by modifying the implementation presented above. Specifically, it does not need to assume that there are only two component reference methylomes that make up the biologic methylomes, nor does it need to estimate them directly. Instead, a set of predictor methylomes can be collected that are mixtures of component reference genomes, as long as the number of the predictor methylomes is the same as the number of the reference component methylomes, and the mixture vectors of the predictor methylomes are linearly independent. For example, they can be methylomes of biologic samples with known different proportion of first and second cell type DNAs.

In process 500, the difference between observed methylation levels in certain target regions and the predicted methylation levels as the test statistic to determine if in a biologic sample the methylome has been affect by some type of tissue abnormality. To illustrate the advantage of this approach, it is assumed that the mixture vector p_j for the methylome of a normal biologic sample j followed a Dirichlet's distribution with parameters α₁= . . . =α_R. Furthermore, for CpG site i, its methylation levels in the R reference vector p_j for component methylomes are m_i,r=(r−1)/(R−1). It can be shown that the methylation level of i in sample j then has a mean of 0.5, and a variance of

$\frac{R+1}{12\left(R-1\right)\left(R{\alpha }_1+1\right)}.$

If there is a methyl-seq library in sample j with a coverage of N for CpG site i, the variance of the measured methylation level z_i,j is

${\sigma }_1^2=\frac{1}{4N}+\frac{N-1}{N}\frac{R+1}{12\left(R-1\right)\left(R{\alpha }_1+1\right)}.$

In other words, if z_i,j is used as a test statistic to detect abnormal intestinal tissue using biologic sample, under the null hypothesis, the test statistic has a variance of σ₁². However, in process 500, it is first estimated the mixture vector p_j, then predicted z_i,j by Σ_rm_i,r p_r,j. Note that in a methyl-seq data, each library can cover millions of CpG sites, and that the variance of the coefficients in a linear regression model is inversely proportional to sample size. Thus it is possible to obtain highly accurate estimation of the mixture vector p_i, even if it is taken into account that adjacent CpG sites tend to have correlated methylation levels. Assuming an accurate estimate of Σ_rm_i,r p_r,j can be obtained, that is, the error of the estimation can be ignored, the variance of the difference z_i,j−Σ_rM_i,r p_r,j between the observed methylation level and the prediction will be

$\frac{1}{4N}-\frac{1}{N}\frac{R+1}{12\left(R-1\right)\left(R{\alpha }_1+1\right)}.$

In other words, under the null hypothesis, the test static z_i,j−Σ_r M_i,r p_r,j used in process 500 has a much smaller variance than the other candidate test statistic z_i,j. This in turns means that the presently disclosed test will achieve a higher power at the same level of type I error.

Additional techniques for detecting, assessing, monitoring, or treating preeclampsia can include those set forth in U.S. Application Ser. No. 62/832,157, which is incorporated by reference in this disclosure. Additional techniques for detecting, assessing, monitoring, or treating CNS conditions can include those set forth in U.S. Application Ser. No. 62/882,215, which is incorporated by reference in this disclosure. Additional techniques for detecting, assessing, or monitoring fetal aneuploidy can include those set forth in U.S. Application Ser. No. 62/928,156, which is incorporated by reference in this disclosure. Additional techniques for detecting, assessing, monitoring, or treating ovarian cancer can include those set forth in U.S. Application Ser. No. 63/007,218, which is incorporated by reference in this disclosure. Additional techniques for detecting, assessing, monitoring, or treating endometriosis can include those set forth in U.S. Application Ser. No. 63/007,204, which is incorporated by reference in this disclosure. Additional techniques for detecting, assessing, monitoring, or treating necrotizing enterocolitis can include those set forth in U.S. Application Ser. No. 63/007,208, which is incorporated by reference in this disclosure.

FIG. 6 is a flowchart of an example process 600 for processing sequence data for nucleic acids from a biological sample to determine a methylation profile of the sample and classify a person with respect to a specified medical condition. The process 600 can be carried out in whole or in part by systems of one or more computers, e.g., systems 102, 110, or 120. It should also be appreciated that aspects of the process 600 may incorporate operations from process 500 of FIG. 5 as previously described.

The process 600 can include obtaining an initial set of sequence data (602). The initial sequence data describes sequences of nucleic acids from a biologic sample of a person or other mammal. In some examples, the nucleic acids characterized in the initial sequence data include cell-free DNA. The initial sequence data may include cell-free DNA from various tissues of the person, including tissues affected by a specified medical condition and tissues that are not affected, or are substantially less affected, by the specified medical condition. DNA fragments corresponding to the affected tissues may be deemed target DNA (e.g., DNA from the intestinal region when assessing necrotizing enterocolitis), while fragments corresponding to the other tissues may be deemed background or non-targeted DNA. The system receives an indication of the specified medical condition that is to be screened, e.g., based on user input provided into a computing terminal (604). The initial sequence data can be filtered using a selected filter corresponding to the specified medical condition (606). In some implementations, the filtering is operable to increase a fraction of sequences for target DNA relative to other DNA. In some implementations, the filtering includes selecting target nucleic acids from an initial set of nucleic acids based a methylation characteristic (e.g., methylation status), a copy number characteristic of the target nucleic acids, or both, and enriching the target nucleic acids, e.g., to increase a fraction of the target nucleic acids after filtering relative to their fraction in a pre-filtered set. A methylation profile can be generated from the filtered sequence data (608), and the methylation profile processed with an appropriate machine-learning model corresponding to the specified medical condition to generate a classification result (610). In some implementations, the machine-learning model is a model corresponding to those described with respect to the process 500 of FIG. 5. The classification result can indicate that the person's condition is normal and does not exhibit the disease or specified medical condition, or the classification result can indicate that the person's condition is abnormal and does exhibit the disease or specified medical condition. The classification result can be provided as output by the system in a number of ways (612). For example, the classification result may be presented to one or more users on an electronic display, presented audibly through an intelligent assistant device including a speaker, transmitted to a user as a message via one or more channels (e.g., email, text messages, private social media messages). In some implementations, the system may alert the patient, his or her healthcare provider, or both, of the classification result. The healthcare provider can use the classification result to diagnose the patient with the specified medical condition, inform a determination of how and whether to treat the patient for the specified medical condition, and/or inform a determination of whether further testing or assessment should be performed to screen the patient for the specified medical condition or another medical condition (614).

FIG. 7 is a flowchart of an example process 700 for training a machine-learning model to generate, from a methylation profile of a patient, a classification result indicating whether a specified medical condition of the patient is or is not normal. The system can be performed by a system of one or more computers, e.g., system 110 from the environment of FIG. 1. While process 700 relates to the training of certain types of models (e.g., feedforward neural networks) using a supervised learning technique, further description of training techniques is disclosed with respect to FIG. 5. The system obtains a set of training samples (702). Each training sample can include (i) a set of filtered or unfiltered sequence data describing sequences of nucleic acids from a biological sample of a person and (ii) a label indicating whether or not the patient exhibits a specified medical condition. The label can serve as a target output for the model 146 when evaluated on a methylation profile derived from the sequence data in the training sample. The machine-learning model is initialized (704). For example, the weights or other parameters of the model may be set to default or randomized values. An initial training sample is then selected (706), and the sequences from the selected training sample can be filtered to increase a fraction of sequences for target nucleic acids in the sequence (708). The system determines a methylation profile from the filtered sequence data (710), processes the methylation profile to generate an estimated classification (712), compares the estimated classification to the labeled/target classification to determine an error in the classification result (714), and then the error can be back-propagated through the model to update the model based on the error (e.g., using gradient descent) (716). The system checks if the training process is complete (718), e.g., by checking whether a training termination condition is satisfied. For example, training may terminate once all training samples have been consumed. The trained model can then be provided for use, e.g., by storing the model in association with a package for the corresponding medical condition (720).

FIG. 8 shows an example of a computing device 800 and a mobile computing device 850 that can be used in some embodiments to implement the techniques described herein. The computing device 800 is intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The mobile computing device is intended to represent various forms of mobile devices, such as personal digital assistants, cellular telephones, smart-phones, and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the inventions described and/or claimed in this document.

The computing device 800 includes a processor 802, a memory 804, a storage device 806, a high-speed interface 808 connecting to the memory 804 and multiple high-speed expansion ports 810, and a low-speed interface 812 connecting to a low-speed expansion port 814 and the storage device 806. Each of the processor 802, the memory 804, the storage device 806, the high-speed interface 808, the high-speed expansion ports 810, and the low-speed interface 812, are interconnected using various busses, and may be mounted on a common motherboard or in other manners as appropriate. The processor 802 can process instructions for execution within the computing device 800, including instructions stored in the memory 804 or on the storage device 806 to display graphical information for a GUI on an external input/output device, such as a display 816 coupled to the high-speed interface 808. In other implementations, multiple processors and/or multiple buses may be used, as appropriate, along with multiple memories and types of memory. Also, multiple computing devices may be connected, with each device providing portions of the necessary operations (e.g., as a server bank, a group of blade servers, or a multi-processor system).

The memory 804 stores information within the computing device 800. In some implementations, the memory 804 is a volatile memory unit or units. In some implementations, the memory 804 is a non-volatile memory unit or units. The memory 804 may also be another form of computer-readable medium, such as a magnetic or optical disk.

The storage device 806 is capable of providing mass storage for the computing device 800. In some implementations, the storage device 806 may be or contain a computer-readable medium, such as a floppy disk device, a hard disk device, an optical disk device, or a tape device, a flash memory or other similar solid state memory device, or an array of devices, including devices in a storage area network or other configurations. The computer program product may also contain instructions that, when executed, perform one or more methods, such as those described above. The computer program product can also be tangibly embodied in a computer- or machine-readable medium, such as the memory 804, the storage device 806, or memory on the processor 802.

The high-speed interface 808 manages bandwidth-intensive operations for the computing device 800, while the low-speed interface 812 manages lower bandwidth-intensive operations. Such allocation of functions is exemplary only. In some implementations, the high-speed interface 808 is coupled to the memory 804, the display 816 (e.g., through a graphics processor or accelerator), and to the high-speed expansion ports 810, which may accept various expansion cards (not shown). In the implementation, the low-speed interface 812 is coupled to the storage device 806 and the low-speed expansion port 814. The low-speed expansion port 814, which may include various communication ports (e.g., USB, Bluetooth, Ethernet, wireless Ethernet) may be coupled to one or more input/output devices, such as a keyboard, a pointing device, a scanner, or a networking device such as a switch or router, e.g., through a network adapter.

The computing device 800 may be implemented in a number of different forms, as shown in the figure. For example, it may be implemented as a standard server 820, or multiple times in a group of such servers. In addition, it may be implemented in a personal computer such as a laptop computer 822. It may also be implemented as part of a rack server system 824. Alternatively, components from the computing device 800 may be combined with other components in a mobile device (not shown), such as a mobile computing device 850. Each of such devices may contain one or more of the computing device 800 and the mobile computing device 850, and an entire system may be made up of multiple computing devices communicating with each other.

The mobile computing device 850 includes a processor 852, a memory 864, an input/output device such as a display 854, a communication interface 866, and a transceiver 868, among other components. The mobile computing device 850 may also be provided with a storage device, such as a micro-drive or other device, to provide additional storage. Each of the processor 852, the memory 864, the display 854, the communication interface 866, and the transceiver 868, are interconnected using various buses, and several of the components may be mounted on a common motherboard or in other manners as appropriate.

The processor 852 can execute instructions within the mobile computing device 850, including instructions stored in the memory 864. The processor 852 may be implemented as a chipset of chips that include separate and multiple analog and digital processors. The processor 852 may provide, for example, for coordination of the other components of the mobile computing device 850, such as control of user interfaces, applications run by the mobile computing device 850, and wireless communication by the mobile computing device 850.

The processor 852 may communicate with a user through a control interface 858 and a display interface 856 coupled to the display 854. The display 854 may be, for example, a TFT (Thin-Film-Transistor Liquid Crystal Display) display or an OLED (Organic Light Emitting Diode) display, or other appropriate display technology. The display interface 856 may comprise appropriate circuitry for driving the display 854 to present graphical and other information to a user. The control interface 858 may receive commands from a user and convert them for submission to the processor 852. In addition, an external interface 862 may provide communication with the processor 852, so as to enable near area communication of the mobile computing device 850 with other devices. The external interface 862 may provide, for example, for wired communication in some implementations, or for wireless communication in other implementations, and multiple interfaces may also be used.

The memory 864 stores information within the mobile computing device 850. The memory 864 can be implemented as one or more of a computer-readable medium or media, a volatile memory unit or units, or a non-volatile memory unit or units. An expansion memory 874 may also be provided and connected to the mobile computing device 850 through an expansion interface 872, which may include, for example, a SIMM (Single In Line Memory Module) card interface. The expansion memory 874 may provide extra storage space for the mobile computing device 850, or may also store applications or other information for the mobile computing device 850. Specifically, the expansion memory 874 may include instructions to carry out or supplement the processes described above, and may include secure information also. Thus, for example, the expansion memory 874 may be provided as a security module for the mobile computing device 850, and may be programmed with instructions that permit secure use of the mobile computing device 850. In addition, secure applications may be provided via the SIMM cards, along with additional information, such as placing identifying information on the SIMM card in a non-hackable manner.

The memory may include, for example, flash memory and/or NVRAM memory (non-volatile random access memory), as discussed below. The computer program product contains instructions that, when executed, perform one or more methods, such as those described above. The computer program product can be a computer- or machine-readable medium, such as the memory 864, the expansion memory 874, or memory on the processor 852. In some implementations, the computer program product can be received in a propagated signal, for example, over the transceiver 868 or the external interface 862.

The mobile computing device 850 may communicate wirelessly through the communication interface 866, which may include digital signal processing circuitry where necessary. The communication interface 866 may provide for communications under various modes or protocols, such as GSM voice calls (Global System for Mobile communications), SMS (Short Message Service), EMS (Enhanced Messaging Service), or MMS messaging (Multimedia Messaging Service), CDMA (code division multiple access), TDMA (time division multiple access), PDC (Personal Digital Cellular), WCDMA (Wideband Code Division Multiple Access), CDMA2000, or GPRS (General Packet Radio Service), among others. Such communication may occur, for example, through the transceiver 868 using a radio-frequency. In addition, short-range communication may occur, such as using a Bluetooth, WiFi, or other such transceiver (not shown). In addition, a GPS (Global Positioning System) receiver module 870 may provide additional navigation- and location-related wireless data to the mobile computing device 850, which may be used as appropriate by applications running on the mobile computing device 850.

The mobile computing device 850 may also communicate audibly using an audio codec 860, which may receive spoken information from a user and convert it to usable digital information. The audio codec 860 may likewise generate audible sound for a user, such as through a speaker, e.g., in a handset of the mobile computing device 850. Such sound may include sound from voice telephone calls, may include recorded sound (e.g., voice messages, music files, etc.) and may also include sound generated by applications operating on the mobile computing device 850.

The mobile computing device 850 may be implemented in a number of different forms, as shown in the figure. For example, it may be implemented as a cellular telephone 880. It may also be implemented as part of a smart-phone 882, personal digital assistant, or other similar mobile device.

Various implementations of the systems and techniques described here can be realized in digital electronic circuitry, integrated circuitry, specially designed ASICs (application specific integrated circuits), computer hardware, firmware, software, and/or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

These computer programs (also known as programs, software, software applications or code) include machine instructions for a programmable processor, and can be implemented in a high-level procedural and/or object-oriented programming language, and/or in assembly/machine language. As used herein, the terms machine-readable medium and computer-readable medium refer to any computer program product, apparatus and/or device (e.g., magnetic discs, optical disks, memory, Programmable Logic Devices (PLDs)) used to provide machine instructions and/or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term machine-readable signal refers to any signal used to provide machine instructions and/or data to a programmable processor.

To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a client computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.

Although various implementations have been described in detail above, other modifications are possible. In addition, the logic flows depicted in the figures do not require the particular order shown, or sequential order, to achieve desirable results. In addition, other steps may be provided, or steps may be eliminated, from the described flows, and other components may be added to, or removed from, the described systems. Accordingly, other implementations are within the scope of the following claims.

OTHER EMBODIMENTS AND EXAMPLES

Example 1—Methods of Assessing DNA Methylation for Pregnancy Phenotyping and Disease Diagnosis

Introduction of this Example

The example relates to methods for diagnosing, prognosing, monitoring and/or treating pregnancy-associated disorders in a pregnant subject.

Background of this Example

Early detection of pregnancy-associated conditions and prenatal disorders, including potential complications during pregnancy or delivery, is crucial, as it allows early medical intervention necessary for the safety of both the mother and the fetus. For example, the pregnancy-associated condition preeclampsia affects 2%-8% of all pregnancies and contributes to 15% of preterm deliveries and between 9% and 26% of maternal deaths worldwide. A number of risk factors for preeclampsia have been identified including hypertension and diabetes, and the consequences of preeclampsia can be far reaching and can include an elevated lifetime risk of cardiovascular disease both in the mother and the infant.

Prenatal diagnosis is typically reliant on invasive procedures such as chorionic villus sampling and amniocentesis. However, such early gestational placental biopsies for prediction of complex gestational diseases are not feasible due to the costly and invasive nature of such procedures. In addition, these invasive procedures are associated with higher risks of spontaneous abortion or fetal death, procedure-induced limb defects and oromandibular hypogenesis, fetomaternal hemorrhage, persistent leakage of amniotic fluid and vertical transmission of infection, and can also cause maternal discomfort and anxiety. With regard to preeclampsia, the only reliable predictor is a previous occurrence.

Alternatives to such invasive approaches have been developed for prenatal screening following the discovery that plasma from pregnant women contains significant numbers of genome equivalents that are derived from the fetoplacental unit. It has also been demonstrated that fetoplacental RNAs are detectable in maternal plasma and that these can be targeted for non-invasive early gestational phenotyping (Koh et al., Proc Natl Acad Sci USA. 2014; 111:7361-7366 and Ngo et al., Science. 2018; 360:1133-1136). However, very little is understood regarding the potential of DNA methylation in maternal plasma to provide non-invasive diagnostic and phenotypic information during pregnancy. This is significant because numerous studies have identified altered DNA methylation in the placentas of mothers affected by complex gestational diseases (Lapaire et al., Fetal Diagn Ther. 2012; 31:147-153; Winn et al., Pregnancy Hypertens. 2011; 1:100-108; Kang et al., J Hypertens. 2011; 29:928-936; Chaouat et al., J Reprod Immunol. 2011; 89:163-172; Varkonyi et al., Placenta. 2011; 32 Suppl:S21-29; Yuen et al., Eur J Hum Genet. 2010; 18:1006-1012; Blair et al., Mol Hum Reprod. 2013; 19:697-708; and Chu et al., PLoS One. 2014; 9:e107318). However, it is not known to what extent the maternal plasma provides a window into the fetoplacental DNA methylome and therefore little progress has been made with respect to whether this approach has value for diagnosis and phenotyping of the fetoplacental unit during early gestation. Therefore, there is a need for non-invasive methods for diagnosis of pregnancy-associated conditions, e.g., preeclampsia, and prenatal conditions.

Summary of this Example

This Example relates to methods for diagnosing, prognosing, monitoring and/or treating pregnancy-associated disorders, e.g., preeclampsia, in a pregnant subject. It is based, at least in part, on the discovery that DNA is differentially methylated in blood samples from pregnant women with preeclampsia as compared to pregnant women that do not have preeclampsia. This Example further provides algorithms and kits for diagnosing, prognosing, monitoring, classifying and/or treating pregnancy-associated disorders.

In certain embodiments, a method for diagnosing, prognosing, classifying and/or monitoring a pregnancy-associated disorder in a pregnant subject can include obtaining a biological sample from the subject, determining the methylation status and/or level of one or more genomic loci in the biological sample, comparing the methylation status and/or level of the one or more genomic loci to a reference and diagnosing a pregnancy-associated disorder in the subject. In certain embodiments, a difference in the methylation status and/or level of the one or more genomic loci in the biological sample compared to the reference indicates the presence of the pregnancy-associated disorder in the subject. In certain embodiments, the reference is the methylation status and/or level of the one or more genomic loci in a biological sample obtained from a pregnant subject that does not have the pregnancy-associated disorder. In certain embodiments, the reference is the methylation status and/or level of the one or more genomic loci in a biological sample obtained from the pregnant subject being tested for the pregnancy-associated disorder. In certain embodiments, the reference is the methylation status and/or level of the one or more genomic loci in a biological sample obtained from a non-pregnant subject. In certain embodiments, the pregnancy-associated disorder is preeclampsia, preterm labor, preterm birth, hyperemesis gravidarum, ectopic pregnancy or intrauterine growth retardation. For example, but not by way of limitation, the pregnancy-associated disorder is preeclampsia. In certain embodiments, the pregnancy-associated disorder is preterm birth.

This Example further provides methods for diagnosing, prognosing and/or monitoring a pregnancy-associated disorder, e.g., preeclampsia in a pregnant subject. For example, but not by way of limitation, the methods can include obtaining a biological sample from the subject, determining the methylation status and/or level of one or more genomic loci in the biological sample, comparing the methylation status and/or level of the one or more genomic loci to a reference and diagnosing the pregnancy-associated disorder, e.g., preeclampsia, in the subject. In certain embodiments, a difference in the methylation status and/or level of the one or more genomic loci in the biological sample compared to the reference indicates the pregnancy-associated disorder, e.g., preeclampsia, in the subject. In certain embodiments, the reference is the methylation status and/or level of the one or more genomic loci in a biological sample obtained from a pregnant subject that does not have the pregnancy-associated disorder, e.g., preeclampsia. In certain embodiments, the reference is the methylation status and/or level of the one or more genomic loci in a biological sample obtained from the pregnant subject being tested for the pregnancy-associated disorder, e.g., preeclampsia. In certain embodiments, the reference is the methylation status and/or level of the one or more genomic loci in a biological sample obtained from a non-pregnant subject.

This Example further provides methods for determining if a pregnant subject has an increased risk of having a preterm birth. In certain embodiments, the method includes obtaining a biological sample from the subject, determining the methylation status and/or level of one or more genomic loci in the biological sample, comparing the methylation status and/or level of the one or more genomic loci to a reference and determining that the subject is at an increased risk of preterm birth. In certain embodiments, a difference in the methylation status and/or level of the one or more genomic loci in the biological sample compared to the reference indicates that the subject is at an increased risk of preterm birth.

In certain embodiments, a method for diagnosing, prognosing and/or monitoring a pregnancy-associated disorder in a pregnant subject can include obtaining a biological sample from the subject, determining the fraction of fetal nucleic acid (fetal fraction) in the biological sample, determining the methylation status of one or more genomic loci in placental nucleic acids present in the biological sample and diagnosing the subject with the pregnancy-associated disorder by analyzing the fetal fraction and the methylation status of the genomic loci in the placental nucleic acid. In certain embodiments, the methylation status is the methylation rate of the one or more genomic loci. In certain embodiments, the fetal fraction is determined by: analyzing the methylation status of one or more reference genomic loci in the biological sample, analyzing the methylation status of the one or more reference genomic loci in a reference sample of maternal blood cells and analyzing the methylation status of the one or more reference genomic loci in a reference sample of placental nucleic acids. In certain embodiments, the methylation status of the one or more genomic loci is determined by: analyzing the methylation status of the one or more genomic loci in the biological sample and analyzing the methylation status of the one or more genomic loci in a reference sample of maternal blood cells. In certain embodiments, the methylation status of the one or more genomic loci is determined by: analyzing the methylation status of the one or more genomic loci in the biological sample and analyzing the methylation status of the one or more genomic loci in a reference sample of blood cells from a non-pregnant individual. In certain embodiments, the methylation status of the one or more genomic loci is determined by: analyzing the methylation status of the one or more genomic loci in the biological sample and analyzing the methylation status of the one or more genomic loci in a reference sample of plasma from a non-pregnant individual.

This Example disclosure also provides for methods of treating a pregnancy-associated disorder in a pregnant subject. In certain embodiments, the method can include obtaining a biological sample from the subject, determining the methylation status and/or level of one or more genomic loci present in the biological sample, comparing the methylation status and/or level of the one or more genomic loci to a reference, diagnosing a pregnancy-associated disorder in the subject, wherein the difference in the methylation status and/or level of the one or more genomic loci in the biological sample compared to the reference indicates the presence of the pregnancy-associated disorder in the subject, and treating the subject diagnosed with the pregnancy-associated disorder. In certain embodiments, a method of treating a pregnancy-associated disorder in a pregnant subject can include diagnosing a pregnancy-associated disorder in the subject by utilization of the algorithm disclosed in Example Embodiment B and treating the subject diagnosed with the pregnancy-associated disorder. In certain embodiments, the pregnancy-associated disorder preeclampsia, preterm labor, hyperemesis gravidarum, ectopic pregnancy or intrauterine growth retardation. For example, but not by way of limitation, the pregnancy-associated disorder is preeclampsia. In certain embodiments, the pregnancy-associated disorder is preterm birth. In certain embodiments, the method of treating preeclampsia can include any method known in the art, e.g., can include one or more of the following treatments: administration of an anti-hypertensive medication, administration of HMG-CoA reductase inhibitors, delivery, administration of a corticosteroid and/or administration of an anti-convulsant medication.

In certain embodiments, an increase in the level of methylation of the one or more genomic loci in the biological sample indicates the presence of the pregnancy-associated disorder or preeclampsia in the subject. In certain embodiments, a decrease in the level of methylation of the one or more genomic loci in the biological sample indicates the presence of the pregnancy-associated disorder or preeclampsia in the subject. Alternatively and/or additionally, a decrease in the level of methylation of at least one of the one or more genomic loci in the biological sample and the increase in the level of methylation of at least one of the one or more different genomic loci in the biological sample indicates the presence of the pregnancy-associated disorder or preeclampsia in the subject.

In certain embodiments, the subject is human. In certain embodiments, the biological sample is a blood sample, stool sample, saliva sample and/or urine sample obtained from the subject. For example, but not by way of limitation, the biological sample can be obtained from the pregnant subject anytime during the pregnancy but prior to onset of clinical systems. In certain embodiments, the biological sample can be obtained from the pregnant subject between week 10 and week 13 of gestation, e.g., for early diagnosis of preeclampsia. In certain embodiments, the one or more genomic loci are present within maternal nucleic acids isolated from the biological sample, e.g., the maternal nucleic acids are obtained from cells, e.g., leukocytes, in the biological sample or are cell-free nucleic acids, e.g., placental nucleic acids, in the biological sample. Alternatively and/or additionally, the one or more genomic loci are present within fetal nucleic acids isolated from the biological sample, e.g., the fetal nucleic acids are cell-free nucleic acids in the biological sample. In certain embodiments, the one or more genomic loci comprise one or more CpG sites.

This Example provides for algorithms for diagnosing and/or monitoring a subject with a pregnancy-associated disorder. In certain embodiments, the algorithm can be used to classify a pregnancy-associated disorder of a subject.

This Example provides for kits for diagnosing, monitoring, classifying and/or treating a subject with a pregnancy-associated disorder. In certain embodiments, a kit of this Example includes a means for determining and/or detecting the methylation status of one or more genomic loci. In certain embodiments, the kit can further include instructions for diagnosing, monitoring and/or treating a subject with a pregnancy-associated disorder, e.g., preeclampsia.

Description of this Example

This Example provides methods for diagnosing, prognosing, monitoring, classifying and/or treating pregnancy-associated disorders, e.g., preeclampsia, in a pregnant subject. It is based, at least in part, on the discovery that DNA is differentially methylated in blood samples from pregnant women with preeclampsia as compared to pregnant women that do not have preeclampsia. For example, but not by way of limitation, the methods disclosed herein include determining the methylation status of one or more genomic loci in a biological sample of a pregnant subject. In certain embodiments, the methods disclosed herein include the use of an algorithm to diagnose, prognose, monitor, classifying and/or treat pregnancy-associated disorders, e.g., preeclampsia, in a pregnant subject.

Definitions of this Example

Unless defined otherwise, all technical and scientific terms used in this Example have the meaning commonly understood by a person skilled in the art to which this invention belongs. The following references provide one of skill with a general definition of many of the terms used in this invention: Singleton et al., Dictionary of Microbiology and Molecular Biology (2nd ed. 1994); The Cambridge Dictionary of Science and Technology (Walker ed., 1988); The Glossary of Genetics, 5th Ed., R. Rieger et al. (eds.), Springer Verlag (1991); and Hale & Marham, The Harper Collins Dictionary of Biology (1991). As used herein, the following terms have the meanings ascribed to them below, unless specified otherwise.

The terms “comprise(s),” “include(s),” “having,” “has,” “can,” “contain(s),” and variants thereof, as used herein, are intended to be open-ended transitional phrases, terms or words that do not preclude the possibility of additional acts or structures. The singular forms “a,” “an” and “the” include plural references unless the context clearly dictates otherwise. This Example also contemplates other embodiments “comprising,” “consisting of” and “consisting essentially of,” the embodiments or elements presented herein, whether explicitly set forth or not.

As used herein, the term “about” or “approximately” means within an acceptable error range for the particular value as determined by one of ordinary skill in the art, which will depend in part on how the value is measured or determined, i.e., the limitations of the measurement system. For example, “about” can mean within 3 or more than 3 standard deviations, per the practice in the art. Alternatively, “about” can mean a range of up to 20%, preferably up to 10%, more preferably up to 5%, and more preferably still up to 1% of a given value. Alternatively, particularly with respect to biological systems or processes, the term can mean within an order of magnitude, preferably within 5-fold, and more preferably within 2-fold, of a value.

As used herein, the term “biomarker” refers to a marker (e.g., DNA methylation status) that allows detection of a disease and/or disorder in an individual, including detection of disease in its early stages. Early stage of a disease, as used herein, refers to the time period between the onset of the disease and the time point that signs or symptoms of the disease emerge. In certain non-limiting embodiments, the presence, absence and/or level of a biomarker in a biological sample of a subject is compared to a reference control.

The terms “reference sample,” “reference control,” “control” or “reference,” as used interchangeably herein, refers to a control for a methylation status of a genomic locus that is to be detected in a biological sample of a subject. In certain embodiments, a reference sample can be a sample from a healthy pregnant individual, e.g., an individual that does not have a pregnancy-associated disorder. In certain embodiments, a reference sample can be a sample from an individual that is not pregnant. In certain embodiments, a reference sample can be a sample from an individual that does not have preeclampsia. In certain embodiments, the reference sample can be an earlier sample taken from the same subject, e.g., while they were not pregnant or during an earlier healthy pregnancy. In certain embodiments, a control or reference can be the presence, absence and/or particular level of a methylation state of a genomic locus in a healthy pregnant individual. In certain embodiments, a control can be the presence, absence and/or particular level of a methylation state of a genomic locus in a healthy individual that underwent treatment for a pregnancy-associated disorder, wherein the healthy individual is non-symptomatic. In certain embodiments, a reference can be the presence, absence and/or particular level of a methylation state of a genomic locus in a healthy individual that has never had the disease. In certain embodiments, the reference can be a predetermined presence, absence and/or particular level of a methylation state of a genomic locus that indicates a subject does not have preeclampsia.

The term “pregnancy-associated disorder,” as used herein, refers to any condition or disease that may affect a pregnant woman or both the woman and the fetus. Such a condition or disease may manifest its symptoms during a limited time period, e.g., during pregnancy or delivery, or may last the entire life span of the fetus following its birth. Non-limiting examples of pregnancy-associated disorders include preeclampsia, preterm labor, preterm birth, hyperemesis gravidarum, ectopic pregnancy, intrauterine growth retardation and genomic abnormalities (e.g., aneuploidy and chromosomal abnormalities such as trisomy 21, trisomy, 18, trisomy 13 and extra or missing copies of the X chromosome and Y chromosome). In certain embodiments, the pregnancy-associated disorder is preeclampsia. In certain embodiments, the pregnancy-associated disorder is preterm birth.

As used herein, “preeclampsia” refers to a pregnancy-associated disorder that manifests as new onset hypertension. For example, but not by way of limitation, a subject suffering from preeclampsia can has a systolic blood pressure greater or equal to 140 mmHg and a diastolic pressure greater than or equal to 90 mmHg and protein in the urine at a concentration greater than 300 mg/dL/24 hr after 20 weeks gestation (i.e., arterial pressure >140/90 mmHg and proteinuria >300 mg/dL/24 hours after 20 weeks gestation). See, e.g., Chu et al., PLoS One. 2014; 9:e107318.

The term “slightly invasive or non-invasive method” refers to a method that does not involve the removal of tissues or fetal cells by biopsy and/or effraction from the placental barrier. In certain embodiments, slightly invasive or non-invasive methods, as disclosed herein, includes the extraction of a biological sample from a subject by venipuncture.

The term “patient” or “subject,” as used interchangeably herein, refers to any pregnant warm-blooded animal, e.g., human or non-human. Non-limiting examples of non-human subjects include non-human primates, dogs, cats, mice, rats, guinea pigs, rabbits, fowl, pigs, horses, cows, goats, sheep, etc. In certain embodiments, the subject is human.

As used herein, the term “biological sample” refers to a sample of biological material obtained from a pregnant subject. In certain embodiments, a sample of biological material obtained from a pregnant subject a pregnant human subject, including a biological fluid. Non-limiting examples of a biological fluid include urine, amniotic fluid, saliva, tears, sweat, blood, plasma and serum. In certain embodiments, the biological sample is a peripheral blood sample from a pregnant subject. In certain embodiments, the blood sample can be a fractionated portion of peripheral blood, such as a plasma sample. In certain embodiments, the biological sample can be a tissue sample obtained from the fetoplacental unit, e.g., as obtained by chorionic villus sampling and/or amniocentesis. In certain embodiments, the biological sample can be maternal leukocytes obtained from a blood sample. In certain embodiments, the biological sample can be circulating fetal cells obtained from a maternal blood sample. In certain embodiments, the biological sample can be a stool sample, a sample from the embryo and/or a sample from the blastocyst.

The term “nucleic acid,” “nucleic acid molecule” or “polynucleotide” includes any compound and/or substance that comprises a polymer of nucleotides. Each nucleotide is composed of a base, specifically a purine- or pyrimidine base (i.e., cytosine (C), guanine (G), adenine (A), thymine (T) or uracil (U)), a sugar (i.e., deoxyribose or ribose) and a phosphate group. In certain embodiments, the nucleic acid molecule is described by the sequence of bases, whereby said bases represent the primary structure (linear structure) of a nucleic acid molecule. The sequence of bases is typically represented from 5′ to 3′. These terms encompass deoxyribonucleic acid (DNA) including, e.g., complementary DNA (cDNA) and genomic DNA, ribonucleic acid (RNA), in particular messenger RNA (mRNA), synthetic forms of DNA or RNA, and mixed polymers comprising two or more of these molecules. The herein described nucleic acid molecule can contain naturally occurring or non-naturally occurring nucleotides. Examples of non-naturally occurring nucleotides include modified nucleotide bases with derivatized sugars or phosphate backbone linkages or chemically modified residues.

The term “isolated” (e.g., isolated genomic DNA) refers to a biological component that has been substantially separated or purified away from other biological components in the cell of the organism in which the component naturally occurs, e.g., other chromosomal and extra-chromosomal DNA and RNA, proteins and organelles. Nucleic acids, e.g., DNA, that have been “isolated” include nucleic acids purified by standard purification methods.

The term “genomic locus” or “genomic DNA locus,” as used herein, refers to any fixed position in a genome. For example, but not by way of limitation, a genomic locus can refer to a genomic element, a chromosomal region, a gene, a region of a gene, e.g., an exon or intron, a regulatory region of a gene, e.g., a promoter or enhancer, a CpG site, a CpG island or a CpG island shore. For example, but not by way of limitation, a genomic locus can include one or more CpG sites, e.g., between about 1 to about 100 CpG sites. In certain embodiments, a genomic locus can be of any particular length, e.g., between about 1 to about 10,000 nucleotides in length.

As used interchangeably herein, “methylation state,” “methylation profile,” “methylation status” and “methylation level” refer to the presence, absence, percentage and/or quantity of methylation at a particular nucleotide, or nucleotides, within a DNA region, e.g., a genomic locus. The methylation status of a particular DNA sequence (e.g., a genomic locus) can indicate the methylation state of every nucleotide in the sequence, indicate the methylation state of any of the nucleotides (e.g., cytosines) in the sequence, can indicate the methylation state of a subset of the nucleotides (e.g., of cytosines), can indicate the percentage or fraction of methylated cytosines at any particular stretch of nucleotides within the sequence or can indicate the average rate of methylation of all the cytosines (or a subset of the cytosines) present in a nucleic acid.

As used herein, a “methylated nucleic acid molecule” refers to a nucleic acid molecule that contains one or more methylated nucleotides that is/are methylated.

As used herein, a “CpG site” or “methylation site” is a nucleotide within a nucleic acid that is susceptible to methylation either by natural occurring events in vivo or by an event instituted to chemically methylate the nucleotide in vitro.

A “CpG island,” as used herein, describes a segment of a nucleic acid, e.g., DNA sequence, that have a high frequency of CpG dinucleotide repeats. See, e.g., Illingworth and Bird, FEBS Letters 2009; 583:1713-1720. For example, but not by way of limitation, Yamada et al. (Genome Research 2004; 14:247-266) have described a set of standards for determining a CpG island: it must be at least 400 nucleotides in length, has a GC content greater than 50% and an OCF/ECF ratio greater than 0.6. Others (Takai et al., Proc. Natl. Acad. Sci. U.S.A. 2002; 99:3740-3745) have defined a CpG island less stringently as a sequence of at least 200 nucleotides in length, having a greater than 50% GC content and an OCF/ECF ratio greater than 0.6.

A “CpG island shore,” as used herein, refers to methylation hotspots that are present a short distance, e.g., less than 2 kb, from CpG islands.

The term “methylome,” as used herein, refers to the amount or pattern of methylation at different sites or regions within a population of cells. The methylome can correspond to all of the genome, a subset of the genome (e.g., repeat elements in the genome) or a portion of the subset (e.g., those areas found to be associated with a pregnancy-associated disorder). A “fetal methylome” corresponds to a methylome of a fetus of a pregnant female. The fetal methylome can be determined using a variety of fetal tissues or sources of fetal DNA, including placental tissues and cell-free fetal DNA in maternal plasma. A methylome from plasma can be referred to a “plasma methylome.” The plasma methylome is an example of a cell-free methylome since plasma and serum include cell-free DNA (cfDNA). The plasma methylome is also an example of a mixed population methylome because it is a mixture of the fetal/maternal methylome. A “maternal methylome” corresponds to a methylome of a pregnant female. The maternal methylome can be determined using a variety of material tissues or sources of maternal DNA, including cell-free maternal DNA in maternal plasma and DNA from leukocytes.

As used herein, the term “increase” refers to alter positively by at least about 2%, including, but not limited to, alter positively by about 5%, by about 10%, by about 15%, by about 20%, by about 25%, by about 30%, by about 35%, by about 40%, by about 45%, by about 50%, by about 55%, by about 60%, by about 65%, by about 70%, by about 75%, by about 80%, by about 85%, by about 90%, by about 95% or by about 100%.

As used herein, the terms “reduce,” “reduction” or “decrease” refers to alter negatively by at least about 2%, including, but not limited to, alter negatively by about 5%, by about 10%, by about 15%, by about 20%, by about 25%, by about 30%, by about 35%, by about 40%, by about 45%, by about 50%, by about 55%, by about 60%, by about 65%, by about 70%, by about 75%, by about 80%, by about 85%, by about 90%, by about 95% or by about 100%.

Methods of this Example

This Example provides methods for diagnosing, monitoring, classifying and/or treating pregnancy-associated disorders in a pregnant subject, e.g., by analyzing the methylation status of one or more genomic loci in a biological sample of a pregnant subject. In certain embodiments, the methods can utilize the algorithm disclosed herein. For example, but not by way of limitation, methods of this Example can be used to diagnose, monitor and/or treat pregnancy-associated disorders including, but not limited to, preeclampsia, preterm labor, preterm birth, hyperemesis gravidarum, ectopic pregnancy and/or intrauterine growth retardation. In certain embodiments, methods this Example allow the early diagnosis of a pregnant subject with preeclampsia, e.g., within the first trimester.

In certain embodiments, the biological sample can be a blood sample (including plasma or serum), stool sample, saliva sample and/or urine sample. In certain embodiments, the biological sample can be obtained from an embryo or blastocyst, e.g., during an in vitro fertilization procedure. The step of collecting a biological sample can be carried out either directly or indirectly by any suitable technique. For example, and not by way of limitation, a blood sample from a subject can be carried out by phlebotomy or any other suitable technique, with the blood sample processed further to provide a serum sample or other suitable blood fraction for analysis.

In certain embodiments, the biological sample, e.g., blood, is extracted from a pregnant subject at any time during the pregnancy. For example, but not by way of limitation, the biological sample can be obtained during the first trimester, second trimester or third trimester of the subject's pregnancy. In certain embodiments, the biological sample can be obtained any time during the pregnancy before the onset of pregnancy-associated disorder and/or before the onset of symptoms of the pregnancy-associated disorder. In certain embodiments, the biological sample is obtained during the first trimester of a subject's pregnancy for early diagnosis of the subject with a pregnancy-associated disorder, e.g., preeclampsia. In certain embodiments, a biological sample can be extracted from a pregnant subject between the 10th and 34th week of the pregnancy. In certain embodiments, the biological sample is obtained before the 34th week of pregnancy, before the 33rd week of pregnancy, before the 32nd week of pregnancy, before the 31st week of pregnancy, before the 30th week of pregnancy, before the 29th week of pregnancy, before the 28th week of pregnancy, before the 27th week of pregnancy, before the 26th week of pregnancy, before the 25th week of pregnancy, before the 24th week of pregnancy, before the 23rd week of pregnancy, before the 22nd week of pregnancy, before the 20th week of pregnancy, before the 21st week of pregnancy, before the 20th week of pregnancy, before the 19th week of pregnancy, before the 18th week of pregnancy, before the 17th week of pregnancy, before the 16th week of pregnancy, before the 15th week of pregnancy, before the 14th week of pregnancy, before the 13th week of pregnancy, before the 12th week of pregnancy, before the 11th week of pregnancy or before the 10th week of pregnancy. For example, but not by way of limitation, the biological sample can be obtained from a pregnant subject between about week 10 and about week 13 of pregnancy, e.g., to diagnose a subject with preeclampsia.

In certain embodiments, multiple biological samples (e.g., two or more, three or more, four or more, five or more, six or more or seven or more biological samples) can be obtained during a subject's pregnancy (e.g., serially obtained samples). For example, but not by way of limitation, multiple biological samples can be obtained during the first trimester. In certain embodiments, one or more samples can be obtained during the first trimester and one or more samples can be obtained during the second or third trimester.

Diagnostic, Prognostic, Classification and Monitoring Methods of this Example

This Example provides diagnostic and prognostic methods for diseases and/or disorders that are characterized by differential methylation of genomic loci. For example, but not by way of limitation, this Example provides methods for diagnosing, prognosing, classifying and/or monitoring a pregnancy-associated disorder in a subject that includes analyzing the methylation status of certain genomic loci. In certain embodiments, the pregnancy-associated disorder is preeclampsia. In certain embodiments, the pregnancy-associated disorder is preterm birth.

In certain embodiments, the analyzed genomic loci can include one or more genomic loci that exhibit differential methylation in a biological sample from a subject that has a pregnancy-associated disorder compared to a reference sample. For example, but not by way of limitation, the methods of this Example include assessing the methylation status of one or more genomic loci, e.g., about 5 or more, about 10 or more, about 50 or more, about 100 or more, about 500 or more, about 1,000 or more, about 5,000 or more, about 10,000 or more, about 25,000 or more, about 50,000 or more or about 100,000 or more genomic loci in a biological sample of a subject. In certain embodiments, the genomic loci can be selected from the genes, or a region within the genes, provided in FIGS. 15A-15B.

In certain embodiments, the one or more genomic loci can be one or more promoter regions of one or more genes, one or more exons of one or more genes, one or more introns of one or more genes, one or more CpG sites, one or more CpG islands, one or more CpG island shores, one or more enhancers of one or more genes or a combination thereof. In certain embodiments, the genomic loci are present on a particular chromosome. For example, but not by way of limitation, the genomic loci are present on chromosome 19.

In certain embodiments, this Example provides for diagnosing, prognosing and/or monitoring a pregnancy-associated disorder in a pregnant subject by detecting the DNA methylation profiles associated with the pregnancy-associated disorder. For example, and not by way of limitation, the method can include (a) obtaining a biological sample from the subject, (b) determining the methylation status of one or more genomic loci present in the biological sample, (c) comparing the methylation status of the one or more genomic loci to a reference and (d) diagnosing a pregnancy-associated disorder in the subject, wherein the difference in the methylation status of the one or more genomic loci in the biological sample compared to the reference indicates the presence of the pregnancy-associated disorder in the subject. In certain embodiments, the difference in the methylation status can also indicate the severity of the pregnancy-associated disorder.

In certain embodiments, a method for diagnosing, prognosing and/or monitoring a pregnancy-associated disorder in a pregnant subject includes (a) obtaining a biological sample from the subject, (b) determining the level of methylation of one or more genomic loci present in the biological sample, (c) comparing the level of methylation of the one or more genomic loci to a reference and (d) diagnosing a pregnancy-associated disorder in the subject, wherein the difference in the level of methylation of the one or more genomic loci in the biological sample compared to the reference indicates the presence of the pregnancy-associated disorder in the subject. In certain embodiments, the difference in the methylation level can also indicate the severity of the pregnancy-associated disorder.

In certain embodiments, this Example further provides methods for monitoring a subject at risk of developing preeclampsia. In certain embodiments, a subject a risk of developing preeclampsia is an individual that suffered from preeclampsia in an earlier pregnancy. For example, and not by way of limitation, the method can include determining the methylation status of one or more genomic loci in the biological sample obtained from the pregnant subject prior to a diagnosis of preeclampsia and determining the methylation status of the one or more genomic loci in a biological sample obtained from the subject at one or more later timepoints during the subject's pregnancy. In certain embodiments, a change in the methylation status of the one or more genomic loci in the second or subsequent samples, relative to the first sample can indicate that the subject has developed preeclampsia.

In certain embodiments, this Example further provides methods for determining if a pregnant subject has an increased risk of having a preterm birth. In certain embodiments, the method includes obtaining a biological sample from the subject, determining the methylation status and/or level of one or more genomic loci in the biological sample, comparing the methylation status and/or level of the one or more genomic loci to a reference and determining that the subject is at an increased risk of preterm birth. In certain embodiments, a difference in the methylation status and/or level of the one or more genomic loci in the biological sample compared to the reference indicates that the subject is at an increased risk of preterm birth.

In certain embodiments, diagnosis of a subject with a pregnancy-associated disorder, monitoring a subject at increased risk of developing a pregnancy-associated disorder or determining if a subject is at risk of having a preterm birth can be based on a higher or lower methylation level of the genomic locus in the biological sample of the subject relative to the methylation level in a reference sample, e.g., a biological sample from a non-pregnant woman or a pregnant woman that does not have the pregnancy-associated disorder. In certain embodiments, the reference sample can be a biological sample from the subject prior to being pregnant and/or during an earlier pregnancy where the subject did not have a pregnancy-associated disorder. In certain embodiments, a difference of greater than about 5%, greater than about 10%, greater than about 15%, greater than about 20%, greater than about 25%, greater than about 30%, greater than about 35%, greater than about 40%, greater than about 45%, greater than about 50%, greater than about 55%, greater than about 60%, greater than about 65%, greater than about 70%, greater than about 75%, greater than about 80%, greater than about 85%, greater than about 90% or greater than about 95% in the methylation (e.g., level, percentage and/or fraction) of the one or more genomic loci in a biological sample obtained from a subject compared to a control is indicative that the subject has the pregnancy-associated disorder or at risk of developing the pregnancy-associated disorder. In certain embodiments, the difference can be a decrease in methylation (e.g., level, percentage and/or fraction) of the genomic loci in the biological sample of the subject. Alternatively, the difference can be an increase in methylation (e.g., level, percentage and/or fraction) of the genomic loci in the biological sample of the subject. In certain embodiments, the difference can be a decrease in the methylation of a genomic locus and an increase in the methylation of a different genomic locus in the sample obtained from the subject. In certain embodiments, a decrease in the level of methylation of one or more genomic loci in the biological sample and the increase in the level of methylation of one or more different genomic loci in the biological sample indicates the presence of the pregnancy-associated disorder or preeclampsia in the subject.

In certain embodiments, diagnosis of a subject with a pregnancy-associated disorder, e.g., preeclampsia, can be based on the methylated or unmethylated state of a genomic locus, e.g., a CpG site. In certain embodiments, a genomic locus, e.g., a CpG site, in a sample from a subject diagnosed with a pregnancy-associated disorder can be methylated and the genomic locus, e.g., the CpG site, in a reference sample can be unmethylated. Alternatively, a genomic locus in a sample from a subject diagnosed with a pregnancy-associated disorder can be unmethylated and the genomic locus in a reference sample can be methylated.

In certain embodiments, this Example provides for diagnosing, prognosing, classifying and/or monitoring preeclampsia in a subject by assessing the DNA methylation profiles associated with preeclampsia. For example, and not by way of limitation, the method can include (a) obtaining a biological sample from the subject, (b) determining the methylation status and/or level of one or more genomic loci present in the biological sample, (c) comparing the methylation status and/or level of the one or more genomic loci to a reference and (d) diagnosing preeclampsia in the subject, wherein the difference in the methylation status and/or level of the one or more genomic loci in the biological sample compared to the reference indicates the presence of preeclampsia or a sub-type of preeclampsia in the subject. For example, but not by way of limitation, the method can be used to indicate whether the subject has mild preeclampsia or severe preeclampsia.

Diagnostic, Prognostic, Classification and Monitoring Methods Using an Algorithm of this Example

This Example further provides diagnostic and prognostic methods for diseases and/or disorders that are characterized by differential methylation of genomic loci by using an algorithm, as disclosed in Example Embodiment B. For example, but not by way of limitation, this Example provides methods for diagnosing, prognosing, classifying and/or monitoring a pregnancy-associated disorder in a subject that includes analyzing the methylation status of certain genomic loci and/or genomic fractions. In certain embodiments, the pregnancy-associated disorder is preeclampsia. In certain embodiments, the pregnancy-associated disorder is preterm birth.

By accurately representing the methylome of the maternal plasma as a mixture of multiple DNA methylomes of both fetal and maternal origin, the algorithm of this Example is able to provide a patient's specific reference methylation pattern for all genomic loci used as a biomarker, reduce the variance of the estimated deviation of the methylation pattern of a genomic locus in a test sample from the reference pattern, and achieve higher power when testing if the deviation is statistically significant.

Methods of Treatment of this Example

This Example further provides methods of treating a subject with a pregnancy-associated disorder, e.g., preeclampsia. In certain embodiments, a method of treating a pregnancy-associated disorder can include diagnosing a subject with the pregnancy-associated disorder as disclosed in herein, followed by the treatment of the subject. Any of the diagnosis methods disclosed herein can be used in treating a subject.

In certain embodiments, the treatment method can include (a) obtaining a biological sample from the subject, (b) determining the methylation status and/or level of one or more genomic loci present in the biological sample, (c) comparing the methylation status and/or level of the one or more genomic loci to a reference, (d) diagnosing a pregnancy-associated disorder in the subject, where the difference in the methylation status and/or level of the one or more genomic loci in the biological sample compared to the reference indicates the presence of the pregnancy-associated disorder in the subject and (e) treating the subject diagnosed with the pregnancy-associated disorder.

In certain embodiments, the methods of treatment can include diagnosing the subject with a pregnancy-associated disorder by use of the algorithm of this Example. For example, but not by way of limitation, the treatment method can include (a) diagnosing the subject with a pregnancy-associated disorder by use of the algorithm of this Example and (b) treating the subject diagnosed with the pregnancy-associated disorder.

In certain embodiments, the treatment method can include (a) obtaining a biological sample from the subject, e.g., a blood sample, (b) determining the fraction of fetal nucleic acid in the biological sample, (c) determining the methylation status of one or more genomic loci in placental nucleic acids present in the biological sample, (d) diagnosing the subject with the pregnancy-associated disorder by analyzing the fetal fraction and the methylation status of the genomic loci in the placental nucleic acid and (e) treating the subject diagnosed with the pregnancy-associated disorder.

In certain embodiments, this Example provides methods for treating preeclampsia. For example, but not by way of limitation, if a subject is diagnosed with preeclampsia, the subject can be treated by any method known in the art to treat preeclampsia. For example, but not by way of limitation, the subject can be treated by administration of medication to reduce the blood pressure of the subject, e.g., by administration of an anti-hypertensive medication. In certain embodiments, delivery can be used as a treatment for preeclampsia. Additional methods of treatment include administration of a corticosteroid administration of HMG-CoA reductase inhibitors and/or administration of an anti-convulsant medication.

In certain embodiments, the information provided by the methods described in this Example can be used by a physician in determining the most effective course of treatment (e.g., preventative or therapeutic) for the subject. A course of treatment refers to the measures taken for a patient after the prognosis or the assessment of increased risk for development of a pregnancy-associated disorder is made. For example, when a subject is identified to have an increased risk of developing a pregnancy-associated disorder, e.g., preeclampsia, the physician can determine whether frequent monitoring of DNA methylation changes can be performed as a prophylactic measure. Also, when a subject is diagnosed with pregnancy-associated disorder, e.g., preeclampsia (e.g., based on the presence of a DNA methylation pattern in a sample from a subject), it can be advantageous to follow such detection with a therapeutic treatment.

In certain embodiments, this Example further provides methods for assessing the efficacy of a therapeutic or prophylactic therapy for preventing, inhibiting or treating a pregnancy-associated disorder in a subject, comprising determining the methylation status of one or more genomic loci obtained from a subject prior to therapy and determining methylation status of the one or more genomic loci present in a biological sample obtained from the subject at one or more time points during therapeutic or prophylactic therapy, wherein the therapy is efficacious for preventing, inhibiting and/or treating a pregnancy-associated disorder in a subject when there is a change in the presence and/or level of methylation of the one or more genomic loci in the second or subsequent samples, relative to the first sample. In certain embodiments, the first sample is obtained after therapeutic treatment has begun.

In certain embodiments, the methods for monitoring the response in a subject to prophylactic or therapeutic treatment (for example, administration of an anti-hypertensive to a subject diagnosed with preeclampsia) can include measuring the methylation status and/or level of one or more genomic loci in a biological sample of a subject at a first timepoint, administering a therapeutic agent, re-measuring the methylation status and/or level of the one or more genomic loci at a second timepoint, comparing the results of the first and second measurements and optionally modifying the treatment regimen based on the comparison. In certain embodiments, the first timepoint is prior to an administration of the therapeutic agent, and the second timepoint is after said administration of the therapeutic agent. In certain embodiments, the first timepoint is prior to the administration of the therapeutic agent to the subject for the first time. In certain embodiments, the dose (defined as the quantity of therapeutic agent administered at any one administration) is increased or decreased in response to the comparison. In certain embodiments, the dosing interval (defined as the time between successive administrations) can be increased or decreased in response to the comparison, including total discontinuation of treatment. In addition, the method of the present disclosure can be used to determine the efficacy of the therapeutic treatment, wherein a change in the methylation status of certain genomic loci present in a biological sample of a subject can indicate that the therapeutic treatment regimen can be altered, reduced and/or stopped.

Assays of this Example

This Example further provides assays and/or methods for determining the DNA methylation status and/or level of genomic loci that correlates with the presence, absence and/or severity of a pregnancy-associated disorder. For example, but not by way of limitation, the pregnancy-associated disorder is preeclampsia. In certain embodiments, a method can include comparing the methylation status and/or level of genomic loci present in a biological sample from a subject that has a pregnancy-associated disorder to the methylation status and/or level of genomic loci in a biological sample from a healthy subject to determine the methylation pattern, as disclosed above, that correlates with the presence of the pregnancy-associated disorder. In certain embodiments, a method can include comparing the methylation status and/or level of genomic loci in a biological sample from a subject that has a pregnancy-associated disorder at an early stage (or less severe case, e.g., mild preeclampsia) to the methylation status and/or level of genomic loci in a biological sample from a subject that has the pregnancy-associated disorder at a late stage (or a more severe case, e.g., severe preeclampsia), as disclosed above, to determine the methylation status and/or level that correlates with the different stages and/or severity of the pregnancy-associated disorder. Non-limiting examples of pregnancy-associated disorders include preeclampsia, preterm labor, preterm birth, hyperemesis gravidarum, ectopic pregnancy, intrauterine growth retardation and genomic abnormalities (e.g., aneuploidy and chromosomal abnormalities such as trisomy 21, trisomy, 18, trisomy 13 and extra or missing copies of the X chromosome and Y chromosome).

DNA Isolation Techniques of this Example

In certain embodiments, the methods of this Example include obtaining nucleic acid from a biological sample from a subject, e.g., a blood sample. There are several platforms that are known in the art and currently available to isolate nucleic acids from biological samples. For example, but not by way of limitation, isolation of DNA from a biological sample can be performed by extraction methods using organic solvents such as a mixture of phenol and chloroform, followed by precipitation with ethanol (see, for example, J. Sambrook et al., “Molecular Cloning: A Laboratory Manual,” 1989, 2nd Ed., Cold Spring Harbor Laboratory Press: New York, N.Y.). Additional non-limiting examples include salting out DNA extraction (see, for example, P. Sunnucks et al., Genetics, 1996, 144:747-756; and S. M. Aljanabi and I. Martinez, Nucl. Acids Res. 1997, 25:4692-4693), the trimethylammonium bromide salts DNA extraction method (see, for example, S. Gustincich et al., BioTechniques, 1991, 11:298-302) and the guanidinium thiocyanate DNA extraction method (see, for example, J. B. W. Hammond et al., Biochemistry, 1996, 240:298-300). There are also numerous commercially available kits that can be used to extract DNA from biological fluids or cells, for example, Qiagen's Gentra PureGene Cell Kit, QIAamp Circulating Nucleic Acid Kit, QiaAmp DNA Mini Kit, DNeasy Blood and Tissue Kit or QiaAmp DNA Blood Mini Kit (Qiagen, Hilden, Germany), GenomicPrep™ Blood DNA Isolation Kit (Promega, Madison, Wis.) and GFX™ Genomic Blood DNA Purification Kit (Amersham, Piscataway, N.J.) can be used to obtain DNA from a biological sample, e.g., blood sample or a tissue sample, from a pregnant woman.

In certain embodiments, the biological sample can be enriched or relatively enriched for maternal nucleic acids (e.g., maternal DNA) by one or more methods. For example, but not by way of limitation, maternal peripheral blood can be collected from a pregnant subject by venipuncture, e.g., during the first trimester of pregnancy, and DNA from leukocytes obtained from the blood sample, i.e., maternal DNA, can be isolated in the disclosed methods. In certain embodiments, placental nucleic acids can be isolated from a biological sample and used in the methods of this Example.

In certain embodiments, the biological sample can be first enriched or relatively enriched for fetal nucleic acids (e.g., fetal DNA) by one or more methods. For example, but not by way of limitation, discrimination between fetal and maternal DNA can be performed by detecting one or more of the following: single nucleotide differences between chromosome X and Y, chromosome Y-specific sequences, polymorphisms located elsewhere in the genome, size differences between fetal and maternal DNA and differences in the methylation pattern between maternal and fetal tissues. For example, but not by way of limitation, fetal nucleic acids can be enriched from a biological sample based on the differential methylation between fetal and maternal nucleic acids. In certain embodiments, separation of fetal and maternal nucleic acids can be based on the methylation status of a genomic locus, e.g., a CpG site or a genomic locus that includes a CpG site. In certain embodiments, the genomic locus of the maternal DNA is methylated and the genomic locus of the fetal DNA is unmethylated. In certain embodiments, the genomic locus of the maternal DNA is unmethylated and the genomic locus of the fetal DNA is methylated. In certain embodiments, the genomic locus of the maternal DNA is hypomethylated compared to the genomic locus of the fetal DNA. In certain embodiments, the genomic locus of the maternal DNA is less methylated compared to the genomic locus of the fetal DNA. In certain embodiments, the genomic locus of the maternal DNA is hypermethylated compared to the genomic locus of the fetal DNA. In certain embodiments, the genomic locus of the maternal DNA is more methylated compared to the genomic locus of the fetal DNA. See, e.g., U.S. 2010/0105049, the contents of which are incorporated by reference in their entirety.

Methylation Detection Techniques of this Example

Various methylation analysis procedures are known in the art, and can be used in the methods of this Example. These assays allow for determination of the methylation state of one genomic locus, e.g., one or more CpG sites or islands within a nucleic acid obtained from a biological sample. In addition, the methods can be used to quantify the methylation of a genomic locus. Such assays involve, among other techniques, DNA sequencing of bisulfite-treated DNA, PCR (for sequence-specific amplification), digital PCR and use of methylation-sensitive restriction enzymes.

In certain embodiments, methylation-specific PCR can be used to determine the methylation status of a genomic loci. Methylation-specific PCR is based on a chemical reaction of sodium bisulfite with DNA that converts unmethylated cytosines, e.g., of CpG dinucleotides, to uracil or UpG, followed by traditional PCR. Methylated cytosines will not be converted in this process and primers can be designed to overlap the methylation site, e.g., CpG site, of interest, thereby allowing one to determine the methylation status of the methylation site as methylated or unmethylated. Additionally, restriction enzyme digestion of PCR products amplified from bisulfite-converted DNA may be used, e.g., by using the method described by Sadri & Hornsby (Nucl. Acids Res. 1996; 24:5058-5059) or COBRA (Combined Bisulfite Restriction Analysis) (Xiong & Laird, Nucleic Acids Res. 1997; 25:2532-2534).

In certain embodiments, whole genome bisulfite sequencing, which is a high-throughput genome-wide analysis of DNA methylation, can be used to determine the methylation status of multiple genomic loci. It is based on sodium bisulfite conversion of genomic DNA, as described above, which is then sequenced on a next-generation sequencing platform. The sequences obtained are then re-aligned to the reference genome to determine the methylation states of cytosines, e.g., of CpG dinucleotides, present within the analyzed genomic loci based on mismatches resulting from the conversion of unmethylated cytosines into uracil.

In certain embodiments, genome-wide DNA methylation profiling can be performed using commercially-available arrays, thereby allowing the interrogation of multiple genomic loci, e.g., multiple CpG sites. Non-limiting examples of such arrays include HumanMethylation BeadChips (Illumina, San Diego, Calif.) and Infinium MethylationEPIC kit (Illumina). Additional methods for analyzing the methylation state of multiple genomic loci is provided in Yong et al., Epigenetics & Chromatin 2016; 9:26, which is incorporated by reference herein.

Kits of this Example

This Example provides kits for diagnosing, monitoring, classifying and/or treating a subject with a pregnancy-associated disorder that comprise a means for determining and/or detecting the methylation status of one or more genomic loci. For example, but not by way of limitation, a kit of this Example can be used to diagnose, monitor and/or treat a subject with preeclampsia. In certain embodiments, a kit of this Example can be used to diagnose, monitor and/or treat a subject with preterm labor and/or preterm birth.

Types of kits include, but are not limited to, packaged probe and primer sets (e.g., TaqMan probe/primer sets), arrays/microarrays, which further contain one or more probes, primers or other detection reagents for determining the methylation state and/or level of one or more genomic loci. For example, but not by way of limitation, a kit of this Example can include one or more probes or primers for detecting the methylation state of one or more genomic loci. In certain embodiments, the one or more genomic loci comprise a CpG site. In certain embodiments, one or more of the genomic loci do not comprise a CpG site. For example, but not by way of limitation, about 5% or more, about 10% or more, about 15% or more, about 20% or more, about 25% or more, about 30% or more, about 35% or more, about 40% or more, about 45% or more, about 50% or more, about 55% or more, about 60% or more, about 65% or more or about 70% or more of the one or more genomic loci detected by the primers or probes of this Example comprise one or more CpG sites.

In certain non-limiting embodiments, a primer and/or probe of this Example can be at least about 10 nucleotides or at least about 15 nucleotides or at least about 20 nucleotides in length and/or up to about 200 nucleotides or up to about 150 nucleotides or up to about 100 nucleotides or up to about 75 nucleotides or up to about 50 nucleotides in length.

In a further non-limiting embodiment, the oligonucleotide primers and/or probes can be immobilized on a solid surface or support, for example, on a nucleic acid microarray, wherein the position of each oligonucleotide primer and/or probe bound to the solid surface or support is known and identifiable.

In certain non-limiting embodiments, a kit of this Example can additionally include other components such as, but not limited to, a buffer, enzymes such as DNA polymerases or ligases, nucleotides such as deoxynucleotide triphosphates, positive control sequences, negative control sequences and the like necessary to carry out an assay or reaction to detect the methylation state of a genomic locus.

In certain embodiments, this Example provides for a kit that includes a container comprising one or more probes and/or primers for detecting the methylation state of one or more genomic loci. The kit can further include instructions for use, e.g., the instructions can describe that a particular methylation status of a genomic locus is indicative of a pregnancy-associated disorder in a subject. The instructions can be printed directly on the container (when present), or as a label applied to the container, or as a separate sheet, pamphlet, card or folder supplied in or with the container.

Reports, Programmed Computers and Systems of this Example

In certain embodiments, the diagnosis and/or monitoring of a pregnancy-associated disorder, e.g., preeclampsia, in a subject based on the methylation status of one or more genomic loci, can be referred to herein as a “report.” A tangible report can optionally be generated as part of a testing process (which can be interchangeably referred to herein as “reporting,” or as “providing” a report, “producing” a report or “generating” a report).

Examples of tangible reports can include, but are not limited to, reports in paper (such as computer-generated printouts of test results) or equivalent formats and reports stored on computer readable medium (such as a CD, USB flash drive or other removable storage device, computer hard drive, or computer network server, etc.). Reports, particularly those stored on computer readable medium, can be part of a database, which can optionally be accessible via the internet (such as a database of patient records or genetic information stored on a computer network server, which can be a “secure database” that has security features that limit access to the report, such as to allow only the patient and the patient's medical practitioners to view the report while preventing other unauthorized individuals from viewing the report, for example). In addition to, or as an alternative to, generating a tangible report, reports can also be displayed on a computer screen (or the display of another electronic device or instrument).

A report can include, for example, an individual's medical history, or can just include size, presence, absence or levels of one or more markers (for example, a report on computer readable medium such as a network server can include hyperlink(s) to one or more journal publications or websites that describe the medical/biological implications). Thus, for example, the report can include information of medical/biological significance as well as optionally also including information regarding the methylation status of relevant genomic loci, or the report can just include information regarding the methylation status of relevant genomic loci without other medical/biological significance.

A report can further be “transmitted” or “communicated” (these terms can be used herein interchangeably), such as to the individual who was tested, a medical practitioner (e.g., a doctor, nurse, clinical laboratory practitioner, genetic counselor, etc.), a healthcare organization, a clinical laboratory, and/or any other party or requester intended to view or possess the report. The act of “transmitting” or “communicating” a report can be by any means known in the art, based on the format of the report. Furthermore, “transmitting” or “communicating” a report can include delivering a report (“pushing”) and/or retrieving (“pulling”) a report. For example, reports can be transmitted/communicated by various means, including being physically transferred between parties (such as for reports in paper format) such as by being physically delivered from one party to another, or by being transmitted electronically or in signal form (e.g., via e-mail or over the internet, by facsimile, and/or by any wired or wireless communication methods known in the art) such as by being retrieved from a database stored on a computer network server, etc.

In certain embodiments, this Example provides computers (or other apparatus/devices such as biomedical devices or laboratory instrumentation) programmed to carry out the methods of this Example, e.g., to perform the algorithm of this Example (see Example Embodiment B). In certain embodiments, the system can be controlled by the individual and/or their medical practitioner in that the individual and/or their medical practitioner requests the test, receives the test results back and (optionally) acts on the test results to reduce the individual's pregnancy-associated disorder risk or treat the individual, such as by implementing a disorder management system.

The following Example Embodiments are offered to more fully illustrate the disclosure of the Example, but are not to be construed as limiting the scope thereof

Example Embodiment a of Example 1: DNA is Differentially Methylated in Healthy Pregnant Females and Pregnant Females with Preeclampsia

DNA methylation patterns are associated with cellular phenotypes, are altered in complex gestational disease states and are cell lineage-specific. Thus, DNA methylation signatures identified in plasma potentially contain information relating to both pathobiology and the cell lineage-specific origins of the signal.

The maternal plasma compartment contains biomarkers derived from both mother and fetus and offers several distinct theoretical advantages. Obtaining maternal plasma is less invasive than obtaining amniotic fluid, placental biopsy or fetal blood. Maternal blood is drawn routinely at several points in time during prenatal care and, thus, a plasma biomarker could be incorporated into the usual provision of care. Using maternal plasma for biomarkers of PTB also facilitates central processing and analysis. After minimal processing at the point of care, blood specimens may be transported to a central facility for processing and analysis.

The present Example Embodiment A used solution phase hybridization to undertake targeted region capture of bisulfite-converted DNA obtained from the plasma of pregnant women in early gestation and non-pregnant female controls as disclosed in Example Embodiment C. The present Example Embodiment A performed targeted sequencing of 80.4 Mb of the plasma methylome and generated an average genome read depth of ˜42× in 18 plasma samples. The present Example Embodiment A used these data to identify the pregnancy-specific characteristics of cell-free DNA (cfDNA) methylation in plasma and found that pregnancy resulted in clearly detectable global alterations in DNA methylation patterns that were modulated by genomic location. Similar data was analyzed from first trimester maternal leukocyte populations and gestational age-matched chorionic villus (CV) and confirmed that tissue-specific DNA methylation signatures in these samples had a significant influence on global and gene-specific methylation in the plasma of pregnant versus non-pregnant women. The subject matter disclosed in this Example can be used in the context of non-invasive prenatal testing with respect to phenotypic pregnancy monitoring and the early detection of complex gestational phenotypes such as preeclampsia and preterm birth.

Results

DNA was extracted from the plasma of both pregnant and non-pregnant women. Blood was drawn from the pregnant women between 10-13 weeks gestation. Plasma DNA was subjected to solution-phase hybridization capture and DNA sequencing. Summary sequencing data are shown in Table 1. It was previously found that CpG methylation levels in early gestational maternal leukocytes were distributed in a biphasic manner with two peaks reflecting largely low methylation (LM) (<20%) and high methylation (HM) (>80%) states respectively. Relatively few CpG sites existed in an intermediate methylation (IM) (20-80%) state in these samples (Chu et al., PLoS One. 2011; 6:e14723). Previous studies have also determined that, compared with maternal leukocytes, the early gestational chorionic villus (CV) contained relatively fewer HM CpG sites, more IM sites and a slight increase in LM sites (Chu et al., PLoS One. 2011; 6:e14723).

	TABLE 1
			Mapping	Mean
	Sample ID	Read Pairs	efficiency	Target Coverage
	EN1	65,822,964	84.53%	25.6
	EN2	82,220,206	84.84%	31.7
	EN3	124,211,203	84.56%	11.0
	EN4	82,129,212	87.74%	24.4
	EN5	57,426,035	84.90%	16.6
	EN6	60,396,896	85.30%	41.9
	EN7	73,976,467	84.51%	40.2
	EN8	76,707,685	84.44%	36.3
	EN9	65,869,746	84.73%	37.3
	EN11	114,262,664	84.85%	33.2
	EN12	152,428,644	84.82%	98.9
	PL12	100,790,172	83.47%	40.1
	PL230	94,809,078	84.76%	71.3
	PL891	78,878,511	83.43%	46.2
	PL897	117,999,412	83.71%	61.3
	PL1523	123,747,149	84.65%	59.5
	PL1746	88,845,825	84.80%	39.7

Differences in the methylation of DNA obtained from pregnant women were compared to non-pregnant women were analyzed. As shown in FIGS. 9A-9G, which provide the distribution of DNA methylation in plasma samples from pregnant and non-pregnant women, differences were observed between the methylation of various sites within DNA isolated from plasma samples obtained from pregnant women as compared to non-pregnant women. As shown in FIG. 9A, the CpG methylation distributions of circulating cell-free plasma DNA from non-pregnant women largely reflected those of maternal leukocytes. Although this distribution was preserved in the plasma of pregnant women, there was a notable reduction in HM sites and an increase in IM sites. Given that the early gestational placenta displayed a significant reduction in HM CpG sites (Chu et al. PLoS One. 2011; 6:e14723), the distribution of CpG methylation in the plasma of pregnant, compared to non-pregnant women reflected the presence in maternal plasma of significant numbers of placentally-derived circulating cfDNA fragments. Thus, patterns of CpG methylation in the CV genome were broadly visible via the window of maternal plasma.

Methylation distributions were influenced by the location of the CpG sites of interest. Those sites that are present in specific genomic elements, specifically: exons, introns, promoters, CpG island (CGIs), CpG island shores and enhancers were examined. To reduce potential bias created by the presence within other elements of CGIs, these regions of interest were filtered to exclude those that overlapped with CGIs (with the exception of course of CGIs themselves). As shown in FIGS. 9A-9G, it was found that methylation distributions were influenced by CpG site location. Those located in promoters consisted of relatively fewer numbers of HM sites (FIG. 9B) when compared to all sites (FIG. 9A). CpGs in introns and exons were skewed towards HM sites (FIGS. 9C and 9D). Relative distributions in CGIs were similar to those in promoters (FIG. 9E). CpGs in CGI shores and enhancers contained relatively more IM sites (FIGS. 9F and 9G) and CGI shores contained more HM sites than LM sites. Pregnancy resulted in a clear relative reduction in HM sites and an increase in IM sites in every case (FIGS. 9B-9G), reflecting the presence of maternal plasma of a significant number of relatively hypomethylated CV genome equivalents. This established that pregnancy has a significant and global impact on the methylation profile of circulating cfDNA and that this is likely dependent to a large degree on the influence of chorionic villus-derived cfDNA fragments.

The spatial differences between CpG methylation in cell-free plasma DNA from pregnant and non-pregnant women were examined. CpG methylation levels were plotted spatially both in a genome-wide fashion and with respect to each autosome. FIG. 10 provides a heatmap of plasma DNA methylation as compared between pregnant (PL) and non-pregnant women (EN). FIG. 10 shows that methylation levels were broadly reduced in DNA extracted from maternal plasma relative to non-pregnant female plasma. This relationship was consistent with respect to every autosome (FIG. 11), further demonstrating the impact of pregnancy as a significant contributor to the CpG methylation signature that was detectable in circulating cell free plasma DNA in the first trimester of pregnancy. As shown in FIG. 11, which provides chromosome-specific plots of spatial DNA methylation distributions, differences in DNA methylation in plasma samples obtained between weeks 10-13 were observed between pregnant and non-pregnant (control) women.

The impact of genomic location on the ability of the presently disclosed subject matter to detect spatial differences in DNA methylation signals in plasma from pregnant and non-pregnant women was examined. The methylation rates of CpG sites present in each of the same structural and regulatory genomic elements (see above) between plasma DNA from pregnant and non-pregnant women were compared. As shown in FIGS. 12A-12F, differential methylation of DNA obtained from plasma samples from pregnant and non-pregnant women differs by genomic element. For example, greater differences in methylation of exons and introns between pregnant and non-pregnant women were observed. As before, regions of interest were filtered to exclude those that overlapped with CGIs. CpGs within exons (FIG. 12B) and introns (FIG. 12C) displayed the most significant pregnancy-specific differences in methylation, followed by those in promoters (FIG. 12A). Sites within CGI shores (FIG. 12D) and enhancers (FIG. 12E) showed fewer differences in global methylation level between pregnant and non-pregnant plasma, and CGIs (FIG. 12F) displayed almost no difference at this resolution. This shows that the impact of the first trimester placental methylome on CpG methylation levels in maternal plasma is influenced significantly by genomic location.

To identify specific CpG sites whose methylation levels are altered between pregnant and non-pregnant plasma, a multiple testing significance filter of q=<0.1 and a % methylation filter of >+/−5% was used. These analyses confirmed the overall reduction in CpG methylation levels observed in maternal plasma compared to plasma from non-pregnant women. Specifically, 24,398 CpG sites were identified, whose methylation levels were significantly different between pregnant and non-pregnant subjects. A significant majority of these (19,470 sites or −80%) displayed lower cfDNA methylation levels in plasma from pregnant women than non-pregnant women. There were 10.35-fold more CpG sites in exons (2463/2702 or ˜91%) that were less methylated in the plasma of pregnant compared to non-pregnant women. Similar ratios were identified for introns, with 8.82-fold more CpG sites in introns (6032/6803 or ˜89%) that were less methylated in the plasma of pregnant compared to non-pregnant women. There were 4.53-times more CpG sites in promoters that were less methylated in pregnant v non-pregnant plasma (2078/2536 or ˜82%) and similar numbers for CGI (4.32-fold, 2987/3671, ˜81%). Similarly of 78 CpGs in enhancers (excluding those overlapping with CGIs) 63 (˜81%) (4.2-fold) were less methylated in pregnant v non-pregnant plasma. Notably, there were only 2.14-fold more CpGs in CGIs (510/1092 or ˜55%) that were less methylated in cfDNA from the plasma of pregnant v non-pregnant women. These results are reflected in the data presented in FIG. 11, and provided considerably more insight into the degree of the impact of pregnancy on cfDNA methylation profiles in plasma.

It was next determined whether the specific loci that are known to have distinct epigenetic signatures in the CV genome may influence maternal plasma DNA methylation profiles in a predictable fashion. To identify such loci, CpG methylation between early gestational CV tissue samples (11-13 weeks) and gestational age-matched maternal leukocytes (MBC) were compared. cfDNA from CV and MBC contributed the majority of circulating DNA fragments to maternal plasma during pregnancy. These assays were performed using solution-phase hybridization followed by bisulfite DNA sequencing. A multiple testing significance filter of q=<0.1 was used for these analyses. The CpG sites that were differentially methylated between the CV and MBC genomes AND that were differentially methylated between plasma samples from the pregnant and non-pregnant women were identified. It was found, in every instance, that if a CpG site was differentially methylated between CV and MBC and between plasma from pregnant and non-pregnant women, the direction of change in each comparison was the same. That is, if a CpG site was less methylated in CV compared to MBC then it was also less methylated in the plasma of pregnant women compared to non-pregnant women. A total of 6,558 CpG sites followed this trend. Examples of such genomic loci were shown in FIGS. 18A-18L. Similarly, if a CpG site was more methylated in CV compared to MBC then it was also more methylated in the plasma of pregnant women compared to non-pregnant women. A total of 1,707 CpG sites followed this trend. Examples of this were shown in FIGS. 18M-18X. No instances were found where these directional relationships did not hold true. These results demonstrate the strong influence of circulating cfDNA fragments originating in the chorionic villus on the molecular phenotype of the maternal plasma DNA methylome in early gestation.

Differences between the distribution of DNA methylation in chorionic villus sampling (CVS) samples and maternal leukocyte samples obtained from pregnant women between 10-13 weeks gestation were also observed. As shown in FIG. 13, differences were observed between the methylation of various sites within DNA isolated from maternal leukocyte samples and CVS samples. FIG. 13A shows the methylation of all sites analyzed, FIG. 13B shows the methylation of sites present within promoters, FIG. 13C shows the methylation of sites present within introns of genes, FIG. 13D shows the methylation of sites present within exons of genes, FIG. 13E shows the methylation of sites present within CpG island shores, FIG. 13F shows the methylation of sites present within CpG islands and FIG. 13G shows the methylation of sites present within enhancers. In particular, significant differences in the methylation status of CpG islands, CpG island shores and enhancers were observed. Differences in the methylation of genomic elements between maternal leukocyte samples and CVS samples were also analyzed. As shown in FIG. 14, differential methylation of DNA isolated from maternal leukocyte samples and CVS samples differed based on the genomic element. Greater differences in methylation were observed in sites present within exons, intron and enhancers.

To determine if genomic regions are differentially methylated in preeclampsia patients, maternal plasma samples were collected from women that later developed severe preeclampsia or mild preeclampsia between 10-13 weeks of gestation. As shown in FIG. 15, which provides a heatmap of the differentially methylated CpG sites within specific genes, significant differences in CpG methylation were observed between pregnant women that would develop severe preeclampsia (SPE; FIG. 15A) or mild preeclampsia (MPE; FIG. 15B) and pregnant women that do not develop preeclampsia (Ctr). FIG. 16 shows the DNA methylation patterns of chromosome 1 in plasma samples obtained at 10-13 weeks of gestation from women who went on to develop severe preeclampsia or mild preeclampsia compared to normal controls. Specific regions within chromosome 1 were differentially methylated in plasma samples from pregnant women that would develop severe preeclampsia or mild preeclampsia compared to pregnant women that do not develop preeclampsia (FIG. 16). Tables 2-4 provide the genomic loci that are differentially methylated in preeclampsia patients. Table 2 provides the genomic loci that are differentially methylated in patients that have severe preeclampsia as compared to normal patients. Table 3 provides the genomic loci that are differentially methylated in normal patients as compared to patients that have mild preeclampsia. Table 4 provides the genomic loci that are differentially methylated in patients that have severe preeclampsia as compared to patients that have mild preeclampsia.

TABLE 2
Severe Preeclampsia versus Normal
chr	start	end	strand	pvalue	qvalue	meth.diff
chr14	3630780	34630780	*	1.23757164166253e−20	1.042273646659133e−14	−69.4444444444444
chr14	100562866	1005626866	*	1.93875488245721e−10	0.343459581332533e−05	−66.677822400714
chr10	80316236	803162236	*	1.36257656350231e−07		−65.2777777777778
chr9	6639395	6639395	*	5.13477476101507e−06	0.0740092321987059	−64.0449438202247
chr3	184083996	184063996	*	5.69057647450106e−06		−61.8041765543427
chr17	67836273	67835273	*	2.06550818522239e−09	0.000221397954240477	−61.4474408365438
chr12	52433824	52433824	*	1.3901734845378e−09	0.000167256226154071	−58.2053501180173
chr14	59713230	59713230	*	4.75620035556604e−	0.070625341303741	−59.0909090909091
chr4	8366705	8366705	*	2.46392999035932e−06	0.0457065024985565	−58.4877622377622
chr3			*	1.73676157460199e−06	0.0376426916901624	−57.3343824713252
chr20	38698763	38698763	*	1.53214328471838e−16	6.94608947471698e−11	−57.2191166599007
chr14	100562865	100562865	*	3.5318390553361e−	0.0583232963007572	−56.6965517241379
chr4	8366877	8366877	*	2.50030133599156e−07	0.010075342199196	−56.6508092282977
chr4	8366676	8366676	*	2.01821805612074e−07	0.00649663983877575	−56.3573625694972
chr7	2123618	2123618	*		4.63379604746384e−05	−56.3555347091932
chr5	181193978	181193978	*	4.82857178702083e−10	6.7764433307093e−05	−56.1562323592302
chr20			*		7.15442052471664e−08
chr11	125932408	125932408	*	3.26700517199238e−09		−53.6016949152542
chr1	205849727	205849727	*	5.19656282371184e−07	0.01770684052329949	−53.5017730496454
chr1	147098664	147096684	*	.9151506676131e−11	7.69372391425696e−05	−53.1922043010753
chr7	4833972	4833972	*	3.50606198251754e−08	0.0580605895994088	−52.6116939883646
chr5	181193979	181193979	*	3.20366650351068e−15	1.1195386542593e−09	−51.1572498298162
chr10	94390089	94390089	*		0.0616405238802056	−50.9295499021527
chrX	142972654	142972654	*	2.77792054960497e−16	1.16977195343523e−10
chrX	5828406	5828406	*	1.63509075637424e−11	3.57015829285575e−06	−48.1912144702842
chr15	50944108	50944108	*	1.03069739500366e−06		−48.1324278438031
	180432021	180432021	*	4.42630224445324e−11	8.41760592191312e−06	−46.7545787545788
chr2	121226155	121226155	*	9.18590257807462e−07	0.0250713348175674	−46.5711503686187
	14803557	14803557	*	4.10811700269213e−06	0.0649018977494697	−46.5405911928851
chr5	180432022	180432022	*	1.45171617533349e−07	0.008846696479283
chr12	128705935	128705935	*		4.22659018652328e−23	−44.7227983579116
chr17	78429801	78429801	*	7.25226999556526e−07	0.0213773240807728	−44.0374003795066
chr6	28444550	28444550	*	4.74288643278353e−09	0.000417327780639565	−43.6251920122888
chr2	240900828	240900828	*	1.1644271575075e−74		−43.3805682467162
chr5	175555631	175555631	*	3.38562904250277e−06	0.0584880066848964	−43.1745031746032
chr1	205849838	205849838	*	6.19096061641065e−07	0.0195227571945974
chrX	142972653	142972653	*	1.22864900146637e−07	0.00622598807228257	−42.7534418022528
chr5	181193879	181193879	*	3.1232729731788e−13	8.36944452391397e−08	−42.6925926926927
chr22	50300221	50300221	*	8.27605998982079e−	0.0934679031257298	−42.0496894409938
chr7	904231	904231	*	1.09939542198965e−	0.0279367133743123	−41.3101983949422
chr22	49239411	49239411	*	2.06341099607852e−07	0.00656656607429472	−39.9092970521542
chr13	114087703	114087703	*	2.50115545257257e−	0.0472767137975261	−39.8302006550456
chr1	121006621	121006621	*	4.47885349679804e−06	0.0662284122113361	−39.6768826700404
chr2	105599292	105599292	*	2.48712460856619e−	0.048995072435054	−39.3647181381715
chr2	240900827	240900827	*	4.24010299264063e−	1.24884392286775e−29	−38.461422543701
chr10	80273713	80273713	*	1.23242597855127e−30		−38.1627828435339
chr1	205849826	205849826	*	9.22990309950144e−06	0.0990711541344773	−38.0831212692282
chr2	105599293	105599293	*	1.31975309310901e−08	0.00105141370157715
chr14	92667121	92667121	*	2.22643778184888e−06	0.0437520776596981	−37.1031746031746
chr8	123851131	123851131	*	3.94219081356403e−	0.0629525059881015	−36.7587232045063
chr7	143443324	143443324	*	6.10229062628313e−06	0.0802319051571692	−36.3359707851491
chr11	8031559	8031559	*	2.7366575657266e−06	0.495141213330146	−36.2747784781219
chr2	232520493	232520493	*	1.18172568538255e−	0.000954340133444324	−36.1954887218048
chr7	904312	904312	*	2.05612595771965e−06	0.0414358634417034	−35.4140829543889
chr17	6997044	6997044	*	1.36283520517278e−14	4.22862444121564e−09	−35.1351351351351
chr10	129048340	129048340	*		0.000282761486039124	−35.0382845915636
chr10	129046339	129046339	*	1.73859926079335e−07	0.007065007040002	−34.95337995338
chr1	169158038	169158038	*	1.56325481535507e−09	0.00017950025742979	−34.7366421052632
chr12	127705934	127705934	*	2.68039058344032e−07	0.0104647697913332	−34.5158002038736
chr4	26083923	26083923	*	4.33199808088278e−06		−34.3992773261066
chr10	131397640	131397640	*	7.37261632374229e−06	0.0581625385929518	−34.3293117171463
chr18	42795335	42795335	*	1.13715064226165e−	0.000931096892695178	−34.2955473660022
chr4	92697208	92697208	*	3.81808787652981e−07	0.0134566624568968	−34.2697594501718
chr1	205650256	205650256	*	2.02202569728084e−	0.0413907747513456	−34.1249226953513
chr20	3650911	3650911	*	4.87572798743508e−09	0.000422707475934433	−34.1176470588235
chr17	6997043	6997043	*	1.1019113537954e−23	1.0826916323106e−17	−33.6709677419355
chr4	99072347	99072347	*	8.06514713089625e−07	0.0231778715158391	−33.3548458885056
chr12	64828874	64828874	*	4.96051999439397e−06		−33.3397832517337
chr14	51508130	51508130	*	8.55511095832576e−	0.00460336644979361	−33.2306111967129
chr14	92687198	92687198	*	5.27577991954763e−07	0.0177728876122088	−33.1851179673321
chr16	33932399	33932399	*	1.40436685477237e−07	0.00673106986650923	−33.1529561529582
chr14	92687205	92687205	*	1.11121583813308e−06	0.0334048397934732	−32.7631578947368
chr1	4161376	4161376	*	1.40538280258833e−06	0.0330088417892059	−32.6213090392195
chr1	24595268	24595268	*	5.67954148117829e−09	0.000485258954093333	−32.5809940286068
chr21	45304888	45304888	*	3.62045084752562e−07	0.0134568624568968	−32.3863836363636
chr1	109712040	109712040	*	3.01185893127661e−06	0.0525324068488898	−31.9859885915705
chr10	44910843	44910843	*	7.7263708029507e−07	0.0225493252148527	−31.5325056088044
chr1	205550472	205550472	*	4.72952773994055e−07	0.0153053843117403	−30.8545027599036
chr1	21428121	21428121	*	5.89429147947907e−06	0.0759587300954092	−30.8253968253968
chr16	78046399	78046399	*	2.06183471870158e−06	0.0414358864417034	−30.7692307692308
chr16	87963156	87963156	*	6.542277206852E−06	0.0629440737732974	−30.5769230769231
chr15	25177929	25177929	*	5.20175126558262e−	0.0032280135492946	−30.3661616151616
chr17	81159480	81159480	*	1.17788440569205e−07	0.0060382933021247	−30.1676481192184
chr22	49239412	49239412	*	2.56600853755404e−07	0.0101289543974943	−29.2403100775194
chr2	130808958	130808958	*	1.01978919977991e−09	0.000127912623247126	−29.2139195277303
chrX	46574452	46574452	*		0.022728747705268	−28.9164077623111
chr14	68695029	68695029	*	8.97786573456088e−	0.0959577598056528	−28.6007438758497
chr21	45304887	45304887	*	1.70459115138157e−06	0.0370817576772098	−28.4476451724158
chr2	130193974	130193974	*	3.38444555514297e−08	0.0022176897988434
chr4	169154581	169154581	*	4.9899650680501e−08	0.00312952931851425	−27.8231621488588
chr10	133272250	133272250	*	1.44490804768508e−06	0.0334048397934732	−27.7777777777776
chr15	52204700	52204700	*		0.0439055437354353	−27.702056432329
chr2	32203968	32203968	*	4.25470681648478e−06	0.0656622300716585	−26
chr10	44910843	44910843	*	1.80546716538467e−07	0.00787736633553591	−25.9353937202038
chr18	80147494	80147494	*		0.0656161957474253	−25.787728028534
chr18	861938	861938	*	6.06231367941984e−08	0.00357394427630267	−25.6210604375232
chr6	144354406	144354406	*	2.18050500676714e−06	0.0429927546827077	−25.5763212161269
chr7	70809483	70809483	*			−25.5308641975309
chr4	187997035	187997035	*	2.69104557651176e−06	0.049354247370596	−25.0264214753752
chr6	159836629	159836629	*	7.20872854132426e−07	0.0213557664991661	−24.9026544734441
chr2	130193993	130193993	*	4.19993792913393e−07	0.0146509418988426	−24.6996949257108
chr5	150846762	150846762	*	6.71723460796621e−07	0.020518356940058	−24.6074348770578
chr15	81118715	81118715	*	8.71178281447527e−	0.09581934279192	−24.4988625719256
chr19	34838866	34838866	*	7.43877477057478e−06	0.0681277435158626	−24.1642084562439
chr9	39133013	39133013	*	4.68242526760355e−06	0.070077245951675	−23.9316239316239
chr17	58520372	58520372	*	8.65429821377254e−07	0.0242952614641285	−23.7095141700405
chr2	130193992	130193992	*	5.08232778437366e−	0.017378240735756	−29.5759267939756
chr6	109901748	109901748	*	7.57378186133299e−	0.0887675391150375	−22.9825899097725
chr6	1427666	1427666	*	4.1656054954838e−06	0.0653130520338026	−22.8012564249801
chr2	130193975	130193975	*	7.63997909945208e−	0.000100059616296525	−22.7777777777778
chr1	111778211	111778211	*	1.38526468537416e−07	0.00670439793176889	−22.6514654418198
chr12	131804120	131804120	*	1.60355088648069e−	0.0355394316415577	−22.4025974025974
chr15	94715715	94715715	*	8.03990294313617e−06	0.0922541353104587	−22.2307104660846
chr6	27213934	27213934	*	5.30074322138427e−06	0.0753204952188882	−21.8158372897251
chr19	34838821	34838821	*	4.2255574048317e−06		−21.6517657142857
chr6	27214027	27214027	*	8.17644643418618e−	0.0928993827561695	−21.4678272742789
chr5	132948556	132948556	*	1.09679422056492e−07	0.00562520947551778	−21.4397496087637
chr1	148310450	148310450	*	2.92561301205199e−07		−21.4387601152105
chr5	13770123	13770123	*	2.97750526917204e−	0.0522530956076664	−21.0654555962025
chr10	119028861	119028861	*	5.41505577263468e−07	0.018045942850302	−21.010101010101
chr14	54349504	54349504	*	1.32500782359076e−06	0.0371607265944355	−20.9900005847611
chr2	9122698	9122698	*		0.0553922940139813	−20.7370043435617
chr 8	56117913	56117913	*	1.31209250928954e−06	0.0314440676356203	−20.2140367129831
chr3	192571473	192571473	*	1.51012441582648e−08	0.00117140887427228
chr3	192571509	192571509	*	1.23198728176775e−06	0.0301059625728883	−19.57246581581
chr1	109712391	109712391	*	8.36381052244206e−06		−19.2891080054867
chr14	103075201	103075201	*	4.97080244677135e−07	0.0170375612163756	−19.135410044501
chr14	103075202	103075202	*	8.04742970205609e−07	0.0123778715158391	−19.1011235955056
chr3	192571475	192571475	*	4.29345309935699e−09	0.000389406069984431	−19.0929251913166
chr3	128078786	128078786	*	8.93677019482845e−	0.0989577598058528	−18.9655172413793
chr3	192571457	192571457	*	3.18199115275514e−06	0.0545318081291326	−18.9288971897668
chr3	192571464	192571464	*	8.16265991559531e−06	0.0928993827561696	−18.8854401805869
chr	1229	1229	*	9.25465083046279e−	0.0990711541344773	−18.8479252572811
chr2	74501497	74501497	*			−18.7544483985765
chr9	137435110	137435110	*	3.27132912037723e−06	0.0553922940439813	−18.4988303357945
chr18	21896998	21896998	*	2.57621167333829e−07	0.0101289543974943	−18.4446322907861
chr9	41269669	41269669	*	8.20157694827191e−06	0.0929829554947384	−18.3809956063631
chr2	240647539	240647539	*		0.0871757023007231	−18.3847663010248
chr10	1774890	1774890	*	7.41911120927606e−	0.0882377435158626	−18.2339658729901
chr9	133788456	133788456	*	5.85389135968004e−06	0.0786094346155254	−18.1078276659677
chr20	63007936	63007936	*	2.36283979197505e−06	0.0453738126355073	−18.0519996468615
chr19	49155243	49155243	*	4.41322599883591e−08	0.067492892065868	−18.048303783331
chr12	8321565	8321565	*	1.87307308936261e−07	0.00800175059286567	−17.9609115552061
chr3	192571414	192571414	*	8.5009966803365e−06	0.0949172818272803	−17.9395881550856
chr10	86403169	86403169	*	1.88645415553304e−06	0.039437240590281	−17.8405001834759
chr10	59283484	59283484	*	2.87524309202921e−06	0.0509925697803757	−17.8802242781048
chr17	8888553	8888553	*	8.66305234324867e−08	0.09581934279192	−17.5944490309392
chr14	52043983	52043983	*	6.42607874239494e−07	0.0200445950219647	−17.5354453019117
chr10	119028855	119028855	*	5.48484355960944e−06	0.0764422128269312	−17.3903638151426
chr1	80197189	80197189	*	6.46758553616037e−	0.000544695156969233	−17.2946556247137
chr10	133497580	133497580	*	5.25908496378768e−07	0.0177728876122088	−17.2454876702245
chr3	192571454	192571454	*	1.87361668787714e−06	0.0393124540648982	−17.2451241134752
chr4	16034373	16034373	*	1.74430702931042e−06		−17.1641791044776
chr17	52043983	52043983	*	1.9812424421928e−06	0.0406608739611971	−17.1368951250225
chr11	3123574	3123574	*	5.78086117798307e−	0.0783425368589402	−17.1212121212121
chr11	12193269	12193269	*	2.17360181038483e−06	0.0429927546527077	−16.9700289514095
chr5	502298	502298	*	5.45495798489112e−08	0.00331543744240762
chr1	228211541	228211541	*	7.69335762497404e−	0.0394579677608044	−16.7807869187532
chr2	231592876	231592876	*	1.56145627015732e−06	0.0348686611284209	−16.3443670150987
chr17	3695455	3695455	*	8.02230489421483e−06	0.6922541353104587	−16.1024844720497
chr6	317262	317262	*	2.58873014807748e−06	0.0481062202469341	−16.0769860769881
chr7	73770020	73770020	*	5.54099620562508e−06	0.0755741040514544	−16.0227272727273
chr7	98399974	98399974	*	8.17917720593016e−07	0.0232942453156757	−15.5682309688581
chr12	64320160	64320160	*	6.46807216516429e−07	0.0200445950218647	−15.4077540105952
chr6	12857835	12857835	*	6.8014319507603e−06	0.0846874616877994	−15.3610243238581
chr14	54349491	54349491	*	4.11736743542011e−06	0.0649018977494697	−15.0819672131148
chr4	80197333	80197333	*	4.40212680119678e−09	0.000393213106631316	−14.9885730221029
chr2	216810316	216810316	*	8.65636492696555e−	0.09581934279192	−14.9299719867955
chr3	51717773	51717773	*	5.93012904896357e−06	0.0790954038425797	−14.4593068071349
chr7	98626784	98626784	*	6.95479892954345e−06	0.0857039652361363	−14.3399372692262
chr4	80197401	80197401	*	6.51606438527839e−06	0.0829440737732974	−14.3055230060152
chr4	106681728	106681728	*	1.01959742093521e−06		−14.3023036509735
chr4	80197434	80197434	*	2.47665753584192e−07		−14.2822601644162
chr4	80197448	80197448	*	3.90192206510683e−09	0.000371019105496151	−13.8938431904116
chr15	81134210	81134210	*	1.46473540854918e−07	0.00885927281633657
chr2	238139939	238139939	*	6.72797903159993e−06	0.0842118288254895	−12.7743498146194
chr12	8321458	8321458	*	3.16414431293741e−07	0.0116585804267766	−12.7116456741669
chr15	81134204	81134204	*	3.14989784152142e−07	0.0116585804267766
chr13	51263036	51263036	*	8.24457527543057e−06	0.0932910405302952	−12.5668418171866
chr4	80197386	80197386	*	4.41138286705164e−06	0.057402892065868	−12.3397435897436
chr4	184592009	184592009	*	4.08890824143861e−06	0.0647997880208238	−12.3244719732236
chr17	41522424	41522424	*	4.85064838768374e−	0.0714905096742304	−12.2570026063612
chr4	80197352	80197352	*	7.9042025396316e−06	0.0913686602537442	−12.2314622314622
chr15	81134172	81134172	*	5.82590267260604e−06	0.0785643200074141	−11.8460725623563
chr10	133178861	133178861	*	4.13593463961443e−06	0.0650207227971104	−11.4820298828914
chr5	141413367	141413367	*	7.76408812915108e−	0.0901023497242646	−11.0053151407691
chr6	27601892	27601892	*	1.12696056779619e−07	0.00598198535935319	−10.8045212765957
chr15	81134247	81134247	*	1.61260371448941e−	0.0356062114996405	−10.2122216407931
chr21	17454767	17454767	*	1.57513331385615e−06	0.0350413488922337	−10.1555164319249
chr10	89537766	89537766	*	4.24059356977105e−	0.0656161957474253
chr7	103969589	103969589	*	6.37032458331714e−06	0.0619984146752947	−5.3475935828877
chr17	48735209	48735209	*		0.049354247370586	2.86168521462639
chr14	90062594	90062594	*	6.96526939592211e−06	0.0657039652351353
chr5	176630707	176630707	*	4.70708171486644e−06	0.0701633113531462	4.59459459459459
chr2	236169565	236169565	*	8.7670482328519e−06	0.0956901512029661	4.7652326965885
chr13	87677946	87677946	*	7.09937990209725e−07	0.0211380343473178	6.23052959501558
chr11	17544470	17544470	*	3.25102908072152e−	0.0553922940139513	7.59020846867064
chr17	9115741	9115741	*	9.15941125734579e−07	0.0250713348175674	7.77777777777778
chr17	52160362	52160362	*	6.28550414474443e−	0.081440051545509	7.9088253177322
chr10	99236564	99236564	*	4.21794729967644e−06	0.0656161957474253	8.84050589932943
chr4	25245665	25245665	*	8.05135851458758e−08	0.0922541351304587	9.07484890748469
chr2	102620063	102620063	*		0.0882897675169175	9.13461538461538
chr1	89843771	89843771	*	6.11840110519317e−06	0.0740092321987059	9.32642489046632
chr10	29409440	29409440	*	1.43616184275359e−	0.0334048397934732	9.69897818282242
chr1	34177008	34177008	*	2.0333104672896e−06	0.0414358834417034	10.2297892606263
chr7	1647690	1647690	*	8.90170165376592e−06	0.0959577598068528	10.605243526773
chr3	9422752	9422752	*	6.82295671075496e−06	0.096322946992369	10.7162924836528
chr10	38402836	38402836	*	7.60787694357665e−	0.0659675592264456	10.7569721115536
chr2	23381331	23381331	*	6.45880698427465e−06	0.094625420271301	11.0674009965906
chr13	114115441	114115441	*	2.00688445021296e−06	0.0412235620765496	11.1227757197731
chr5	39455822	39455822	*	3.97383310973769e−	0.0631458897627837	11.6715875539405
chr17	7767401	7767401	*	9.85317560721615e−07	0.0258676453893068	12.2577519379845
chr11	1299404	1299404	*	7.02662324507093e−06	0.0857039652361363	12.4037322823389
chr7	56229631	56229631	*	9.25954077480037e−06	0.0990711541344773	12.6457762904371
chr6	160679141	160679141	*	3.391955445635E−06	0.0554880066848954	12.6694944043794
chrX	128165501	128165501	*	7.32783881549843e−06	0.0878051642006674	12.8392662662035
chr6	160679196	160679196	*	7.50383914366681e−		12.9055696139103
chr17	9115755	9115755	*	3.61917824126845e−06	0.0592675321864645	13.1250042727435
chr3	67648882	67648882	*	9.72353899380297e−07	0.0259382970300902	13.2977588171549
chr17	2376166	2376166	*	6.63550000349147e−06	0.0634084755830776	13.3778047301395
chr13	110719297	110719297	*	8.55748057908093e−06	0.0953673297910076	13.5954770528915
chr5	16617985	16617985	*	6.09043427333241e−07	0.0195137083815149	13.8078108400913
chr2	120466164	120466164	*	8.97978301427997e−06	0.0959577598056528	13.8459877890251
chr13	36675352	36675352	*	6.63185693860932e−06	0.09581934278192	14.094129932125
chr7	98293439	98293439	*	2.55905278327878e−06	0.0478936639644374	14.2638830112381
chr5	141340409	141340409	*	2.91719063513637e−06	0.0514905720337703	14.4652837607971
chr1	108640932	108640932	*	4.52019430927724e−06		14.5024071753611
chr20	20365677	20365677	*	6.42058331347275e−06	0.0824650297766311	14.5833333333333
chr1	17579648	17579648	*	5.14707366002753e−06	0.0740092321987059
chr6	164040155	164040155	*	2.31449394623397e−06	0.044590669746554	14.7955809109636
chr6	164040129	164040129	*	6.14142579203506e−06	0.0803391129964423	14.9056603773585
chr10	1644256	1644256	*	7.26996022323746e−06	0.0672890798211452	14.9419582540602
chr14	92323937	92323937	*	6.33397268615149e−06	0.0817088990664582	14.9422345715121
chrX	102769280	102769280	*	2.3819544205987e−06	0.0455923631416016	15.2221105527638
chr6	75493950	75493950	*	1.16229603673797e−06	0.0290345349321572	15.2689429546563
chr10	72326102	72326102	*	1.66613645697655e−	0.0366509424001043	15.3494809688581
chr14	2878	2878	*	3.78860341767504e−06	0.0615292686847942	15.3676072452571
chr2	52572195	52572195	*	2.29713811045743e−06	0.0444013976002324	15.5818353831599
chr7	158154286	158154286	*	6.31946156301339e−06	0.0937784182927845	15.5825201041698
chr5	140876767	140876767	*	6.47272912187058e−06	0.0764422128269312	15.5911587945897
chr14	105137166	105137166	*	9.30176934484768e−06	0.0993426775395648	15.6930154958416
chr7	158671573	158671573	*	6.14602328945107e−06	0.0803391129964423	15.8153638814016
chr6	15504589	15504589	*	1.80400817128343e−06	0.0384376186546574	15.8588076338458
chr19	39336029	39336029	*		0.0802319051571692	15.9090909090909
chrX	25006255	25006255	*	1.46199983883199e−06		16.0092005171533
chr12	125142124	125142124	*	2.11915250516777e−06	0.00150519751372088	16.1316966708088
chr6	157511117	157511117	*	6.1711967992935e−06	0.0804897308137579	16.2272625196497
chr6	150679242	150679242	*	1.8335930004598e−06	0.0387443261450957	16.2345679012346
chrX	104584408	104584408	*	1.10803256096536e−06	0.0279407196102626	16.2540682414698
chr17	74270497	74270497	*	4.93447428179247e−06	0.0721847100589141	16.3354893651534
chr5	142426222	142426222	*	1.51460789516886e−06	0.0342027277222956	16.354156661861
chr6	33908181	33908181	*	8.37424153186831e−06	0.0936883143768792	16.362760747395
chr12	34361526	34361526	*	5.63398219765226e−	0.0776034580536978	16.4900551562761
chr15	24773347	24773347	*	8.0890471917745e−06	0.0922541353104587	18.4955259862353
chr3	11582396	11582396	*	5.30314452351505e−06	0.0753204952188082	16.5818661497873
chr2	64636343	64636343	*	1.771487478579E−06	0.0381150853306115	16.6337683763957
chr1	28612745	28612745	*	2.24757603314162e−06	0.0438749689544513	16.6666666666667
chr6	158869351	158869351	*	3.03955849810629e−06	0.0528608695986296	16.7341462047384
chr12	116726927	116726927	*	9.38518755291245e−07	0.0254999228716589	16.9503206622
chr20	19975153	19975153	*	5.83948863226354e−	0.0785977438122122	17.0103092783505
chr7	928847	928847	*	2.30726259098789e−07	0.00944591221805046	17.0478536242083
chr1	153452347	153452347	*	9.0960369712121e−07	0.0250580669445832	17.072192513369
chr6	151616062	151616062	*	1.07739679765797e−	0.0275235483198223	17.0825967904059
chr17	75828646	75828646	*	5.31492486512653e−06	0.0753204952188082	17.1014219794708
chr2	4380984	4380984	*	5.70066406879171e−	0.0780769936964471	17.1465968588387
chr1	232767857	232767857	*	4.75809525975101e−06	0.070628341303741	17.1577275935074
chr11	3156499	3156499	*	8.99722307684049e−07	0.0209396233880114	17.1949313765457
chr5	41869578	41869578	*		0.0342027277222966	17.3694779116466
chr22	26757154	26757154	*	3.84822502918888e−06	0.0617577667911084	14.4166713718405
chr17	6995809	6995809	*	5.51834419781712e−	0.0765741040514544	17.5181937855493
chr20	19975175	19975175	*	3.55499472990289e−06	0.0585416969719241	17.563025210084
chr12	34361520	34361520	*	6.34375976257521e−07	0.0198929069601162	17.6046176046176
chr1	2955499	2955499	*	4.95345048031268e−06	0.072207374487384	17.6112412177966
chr17	9115517	9115517	*	8.45813208293672e−	0.0047945791923793	17.7335428571207
chr17	6996074	6996074	*	7.53937521614787e−	0.0885403049266502	17.8347754200365
chr2	24304224	24304224	*	9.86884626589841e−07	0.0262073302819578	18.042328042328
chr20	19975167	19975167	*	6.47133536675212e−08	0.0825774316790541	18.0620456905504
chr10	133170251	133170251	*	1.15487541922153e−06	0.0289718780343888	18.1174854002061
chr8	19652708	19652708	*	7.19644872008835e−	0.0870543809739352	18.1283787330079
chr15	100553730	100553730	*	8.72606758944996e−	0.09581934279192	18.1361484252569
chr14	95071900	95071900	*	6.82343982140052e−06	0.0848874818877994	18.1432838035877
chr2	231924719	231924719	*	2.7317235815661e−06	0.0495141213330146	18.1804306504307
chr13	99566717	99566717	*	2.79109069942135e−06	0.0499222052083915	18.1597804949519
chr20	19975163	19975163	*	2.81454084761736e−06	0.0501289877413247	18.428826543385
chr20	19975160	19975160	*	1.10903198558053e−	0.0279407196102626	18.4659090909091
chr20	19975171	19975171	*		0.0105227571945974	18.7731624142323
chr5	1410033	1410033	*	1.49557110781536e−07	0.00654244932579815	18.8241742832741
chr5	50377869	50377869	*	6.19047972859498e−	0.0805629715155258	18.8826205641492
chr3	128434766	128434766	*	2.5680972780805e−06	0.047910837291266	19.0173796791444
chrX	43078653	43078653	*	1.47290098991755e−06	0.0336560559387906	19.0642663443805
chr2	91559538	91559538	*	1.2798441911153e−06	0.0307964310716694	19.0743405275779
chr17	8995833	8995833	*	5.80912520858866e−06	0.0785477267285294	19.1361189435605
chr11	45930246	45930246	*	6.67636406131693e−08	0.09581934279192	19.1517314890558
chr12	4911898	4911898	*	5.1999302791302e−06	0.074406294590844	19.4045661303299
chr7	928902	928902	*	1.42073307431449e−09	0.000157514289349215	19.5591142177787
chr3	184017372	184017372	*	3.09142055769524e−07	0.0116082782869539	19.6161648515909
chr17	6996045	6996045	*	1.3474988752516e−06	0.0319034805099525	19.7234678624813
chr13	112980491	112980491	*	6.05595949101343e−08	0.079570207197914	19.8325344694952
chr3	158801256	158801256	*	1.38068648098188e−07	0.00670439793176869	19.8975199973408
chr14	33475011	33475011	*	6.52813522959558e−07	0.0200445950218647	19.9644633972992
chr8	48331785	48331785	*	1.53438473949785e−07	0.0070002592590762	19.9948407807618
chr18	86681145	86681145	*	2.79446277187232e−	0.0499222052083915	20.0137879832423
chr11	45930309	45930309	*	3.87504524032989e−	0.0520761359197938	20.031746031746
chr16	87954981	87954981	*	3.76282271941594e−07	0.0134443306851155	20.1140964298859
chrX	107717006	107717006	*	2.28477009577724e−06	0.04430760700502	20.1513167466899
chr11	72103654	72103654	*	1.32947831334E−06	0.0316037285944355	20.1845419018671
chr8	1515412	1515412	*	3.27940881714947e−	0.00217231402944738	20.1887816091954
chr7	928884	928884	*	5.776099753223E−07	0.0187099518416535	20.5057107427951
chr7	45548154	45548154	*	3.8464420181436e−06	0.0617677667911084	20.2754976958525
chr17	81255325	81255325	*	2.94355442400736e−	0.00197198303895888	20.2867383512545
chr5	56585077	56585077	*	1.79517057959436e−07	0.00787738633553591	20.332669941025
chr11	1299316	1299316	*	6.4988054748583e−07	0.0200445950218847	20.3922148842589
chr16	86681244	86681244	*	9.56783330012095e−06	0.000794446441441805	20.4052181944124
chr5	142426207	142426207	*	3.34377581110171e−06	0.012243924656305	20.4150688021656
chr5	169299958	169299958	*	3.68583716161777e−06	0.0600256603680437	20.6111957349581
chr10	46911213	46911213	*	2.78829720129378e−06	0.0499222052083915	20.6247827598193
chr5	140842538	140842538	*	7.94033601486232e−06	0.091606712737502	20.5908789778777
chr15	21955106	21955106	*	4.91225838152047e−06	0.0721847100589141	20.7339284509702
chr1	15066339	15066339	*	8.97309417940359e−06	0.0989577598056528	20.8438038670017
chr16	30398862	30398862	*	7.03617940967512e−06	0.0857039652361363	20.9139816965438
chr12	90954040	90954040	*	7.8885532383007e−06	0.0913669129032541	20.9183673469388
chr7	4706798	4706798	*		0.0571220483819076	20.0357900614152
chr18	14748232	14748232	*	1.98227666222751e−06	0.0406508739611971	20.9880300263745
chr5	671276	671276	*	6.75153422827891e−06	0.0958901512029551	21.0102040816327
chr10	133565671	133565671	*	3.12406149733364e−06	0.0538946630166299	21.0105134703839
chrX	19344014	19344014	*	2.8014511627033e−07	0.0108654780971641	21.0144057623049
chr2	109130298	109130298	*	2.61302838066889e−06	0.0484424812103089	21.0662740226077
chrX	15645314	15645314	*	3.6550945343122e−09	0.000353246731107763	21.113030279239
chr20	61394416	61394416	*	5.58178210281432e−07	0.0184393499525971	21.1275210398152
chr13	112689062	112689062	*	7.01003959212573e−06	0.0857999652361383	21.2597245433066
chr12	75391126	75391126	*	6.3123204346952e−06	0.0816081564423863	21.2828770799785
chr8	8893931	8893931	*	2.05853107488063e−06	0.0414358834417034	21.2893700787402
chr10	69337894	69337894	*		0.0825774316790542	21.2999195683804
chr6	41198259	41198259	*	8.88973999706272e−06	0.0853306691228229	21.4809364164223
chr10	121527085	121527085	*	6.98851969048408e−07	0.0209396233880114	21.7685862007488
chr15	101065769	101065769	*	7.17708053664482e−		21.8864666522771
chr22	25757078	25757078	*	3.13168016121987e−	0.0538646630177199	21.8995765275257
chr7	928936	928936	*		0.00513966443612768	21.9341430499325
chr5	82334505	82334505	*	3.9636195228669e−06	0.0631458876257637	22.0043231728329
chr7	1024437	1024437	*	7.62328118855421e−08	0.0889675592264456	22.1361921778837
chr17	9116060	9116060	*	1.55088676018667e−06	0.0347639130761953	22.222456957173
chr16	8644774	8644774	*	1.82713707034827e−06	0.0387443261450957	22.4154068445032
chr17	9116020	9116020	*	1.07750438072861e−07	0.00582520947551778	22.4231601731602
chr19	29671725	29671725	*	4.56454934144679e−	0.0691703554407462	22.4514991181658
chrX	47145243	47145243	*	7.64848063702281e−07	0.0224330589636318	22.4900817308735
chr7	127251316	127251316	*	5.35818201189688e−06	0.075751420262087	22.5546975546976
chr2	3316654	3316654	*	6.16402809086827e−08	0.0928993827561695	22.6568941823179
chr13	18600074	18600074	*	5.18309937342403e−08	0.0743458964634313	22.8951956500997
chr6	13877799	13877799	*	8.95126736806239e−06	0.0959577598056528	22.9665071770335
chr19	37294559	37294559	*	1.6220954636522e−09	0.000150430483497275	23.0094030365769
chr7	108415322	108415322	*	1.24614314085159e−	0.0302403896561507	23.0595268218623
chr2	168499447	168499447	*	5.43835452559394e−	0.0761543594497282	23.0726878669358
chr20	45304879	45304879	*	6.90524030830571e−06	0.0853433705154194	23.2261256803563
chrX	64205954	64205954	*	1.52003175779843e−06	0.0342027277222966	23.3647941417225
chr19	7862084	7862084	*	5.52098871833697e−06	0.0765741040514544	23.3613445378151
chr5	25710448	25710448	*	1.23582956553838e−06	0.0301959625728663	23.4370850374462
chr4	1218866	1218866	*	8.37698638800132e−06	0.0938883143768792	23.4448114505918
chr3	184017291	184017291	*	3.61320552585674e−07	0.0129884758275872	23.4491540840458
chr17	6995830	6995830	*	1.36074437699724e−	0.0320882411227902	23.4804754402543
chr17	49014770	49014770	*	2.06640079572259e−06	0.0414358634417034	23.4702790000165
chrX	153781252	153781252	*	5.5987266684809e−07	0.0184393499825971	23.5335132703554
chr22	137400	137400	*	1.54261018011994e−07	0.0070002592590752	23.7164750957854
chr3	13933198	13933198	*	9.59812325404932e−07	0.0258614954261138	23.7825446209268
chrX	63351268	63351268	*	2.7521979719894e−06	0.0496182325919689
chr2	131830061	131830061	*	1.92095230523419e−07	0.00814725209338524	23.9108409321176
chr11	15069378	15069378	*	4.77078686740945e−10	6.77764433307093e−05	23.9130434782609
chr5	141246131	141246131	*	6.59054316005384e−	0.083376692804708	23.9220696263175
chr2	239231927	239231927	*	5.13244577684371e−	0.0740092321987059	23.9330024813896
chrX	49234431	49234431	*	1.67470205968947e−06	0.0367024154776301	24.0277562075934
chr18	24870694	24870694	*	6.81210387660693e−06	0.0846874816877994	24.1156319910515
chr16	14337567	14337567	*	6.81187505369222e−07	0.020700189258845	24.1374353078481
chr19	19514983	19514983	*	3.26632886702103e−06	0.0553922940139813	24.1576605212969
chr2	239232124	239232124	*	8.27114020935035e−07	0.0234429046248908	24.1879597913497
chr2	131830062	131830062	*	5.40964187031731e−06	0.0761138805996409	24.1937856223571
chr18	14458892	14458892	*	1.18500214028664e−06	0.0295329361879514	24.1952309985097
chr7	27411345	27411345	*	6.65982139582401e−08	0.0543084755630776	24.3007477153143
chr2	3595133	3595133	*	1.06394407915843e−06	0.0273900416385782	24.3924050632911
chr20	19975301	19975301	*	4.7967037772016e−06	0.0709023155809478	24.4444444444444
chr13	99586817	99586817	*	5.90848886095403e−08		24.5086431445839
chr11	45930252	45930252	*	7.45553160300389e−08	0.00426727829797806	24.5537163258682
chr5	68225408	68225408	*	1.09193639577545e−07	0.00582520947551778	24.8198276717996
chr17	16689974	16689974	*	2.03999637741829e−07	0.00852942306639986	24.6512136453094
chr17	78140883	78140883	*	8.8705310348995e−06	0.009581934279192	24.7621554544429
chr1	93592170	93592170	*	3.27918112553502e−06	0.0553922940139813
chr11	58830739	58830739	*	7.47083099100034e−06		24.8744939271255
chr17	4946839	4946839	*	8.91482794177763e−08	0.0248428819074298	24.9246868558744
chr9	63781683	63781683	*	4.23812400795643e−08	0.0586161957474253	25.0008648424257
chr11	45930240	45930240	*	1.53082434776836e−07	0.0070002592590762	25.0130982045876
chr1	93592233	93592233	*	2.83100456694509e−07	0.0109083381311835	25.0330667830588
chr18	1459220	1459220	*	3.38616499749607e−06	0.0564880066848964	25.1025464025641
chr2	131263832	131263832	*	2.08273921361247e−06	0.00150519751372088	25.1808406647116
chr3	2083862	2083862	*	2.53773885194839e−09	0.000267158055850311	25.2170666612228
chr18	14459023	14459023	*	2.84654382432768e−06	0.0505462758889119	25.2670716360394
chr7	89228935	89228935	*	1.81723746139605e−07	0.00787736633553591	25.5062571103527
chr2	11511040	11511040	*	9.25184599489317e−06	0.0990711541344773	25.5118574164089
chr9	55376788	55376788	*	5.75437260355062e−07	0.0187099518416835	25.6744186046512
chr1	16353352	16353352	*	4.65527420517651e−06	0.0700113694379202	25.8187134502924
chr17	6995802	6995802	*	4.79049616505616e−06	0.0709023155809878	25.8361204013378
chrX	15231803	15231803	*	2.73807433589792e−06	0.0495141213330146	25.9012345679012
chr7	928856	928856	*	6.84664612433188e−08	0.00395719163619035	25.9324155193992
chr16	88937576	88937576	*	1.25160657680843e−06	0.0302403896581507	25.9938328194016
chrX	23908198	23908198	*	2.2949207621491e−07	0.00944591221805040	26.234796404019
chr13	99565890	99565890	*	3.58056479295213e−06	0.0587985289984288	25.4184907834101
chrX	52995515	52995515	*	3.12651175527464e−06	0.0538646630166299	26.5151515151515
chr12	121654239	121654239	*	6.73629885054117e−	0.00393196229897439	26.7460319460317
chr10	94383771	94383771	*	1.57870254186226e−	0.00017950025742979	26.8034215111674
chr2	131253902	131253902	*	1.61220629993114e−	0.00121852914927674	27.1579077819275
chr17	44210845	44210845	*	1.94982343283369e−06	0.0404749497369219	27.1787858065497
chr1	1509036	1509036	*	1.45502579270725e−	0.0335073513357785	27.1883269124668
chr22	43129256	43129256	*	8.63505007185784e−06	0.0242952614341285	27.1971829380967
chr20	45304619	45304619	*	5.75997712730997e−06	0.0783485368589402	27.2538272538272
chr2	130455808	130455808	*	2.89043106423357e−07	0.0110649961709326	27.5730778678457
chr21	44921095	44921095	*	2.83830583449049e−07	0.0248428619074298	27.7561770977048
chr1	53146300	53146300	*	5.99018904485462e−06	0.0792393419764003	27.8161886227838
chr16	33029245	33029245	*	2.62618153607755e−06	0.0485487901506917	27.8814732761008
chrX	107776374	107776374	*	1.38742731797769e−07	0.00670439793176869	28.1957186544343
chr9	135485006	135485006	*	7.06210067890364e−06	0.0858423382496341	28.2051282051252
chr22	40536393	40536393	*	2.31951733013113e−10		26.2352941176471
chr19	33442882	33442882	*	1.73077443193989e−09	0.000188953997315065	28.3827780313403
chrX	64395518	64395518	*	5.54626257453056e−06	0.0785741040514544	28.5048247625287
chr7	928941	928941	*	1.62734713385636e−07	0.0073234931621066	28.8420376712329
chr12	2106958	2106958	*	1.85235829723081e−08	0.00136231581398794	28.9473684210526
chr13	99565943	99565943	*	8.72358591444338e−06	0.09562934279192	28.989080662205
chr2	131830531	131830531	*	4.93224498006621e−06	0.0721847100589141	29.6087533156499
chr9	63781708	63781708	*	2.0943910600261e−06	0.00150519751372088	30.2474247292239
chr4	1883408	1883408	*	7.2163450433967e−06	0.0870543809739352	30.3359462486002
chr19	37294432	37294432	*	3.13392568998976e−06	0.0538646630168299	30.3885590933621
chr12	120103765	120103765	*	1.29292545702216e−07		30.6754560231172
chr6	44311573	44311573	*	6.96283160484396e−16	2.56551927582013e−10	30.7361762460485
chr2	89649492	89649492	*	3.80279779206735e−06	0.0615901443820628	30.8474576271186
chr1	53146299	53146299	*	1.01516567163954e−06	0.0286055408882385	31.2574671445635
chr8	142832875	142832875	*	4.57395088049963e−06	0.070077245951875	31.3136979897148
chr18	76767817	76767817	*	9.60701295160221e−07	0.0258614954261138	31.5458937198068
chr18	76349531	76349531	*	2.52388135007034e−06	0.00173013445877657	31.6992645216443
chr14	52122802	52122802	*	3.0710009730901e−07	0.0116055233164768	31.7062085264976
chr16	33029286	33029286	*	1.70183579048149e−07	0.00760069130447552	31.7708333333333
chr5	33502742	33502742	*	7.49883997457326e−06	0.0882987675169175	31.8063305000752
chr2	3125150	3125150	*	7.38867223422674e−06	0.068175881558769	31.8300065935383
chrX	154805593	154805593	*	6.62749295173053e−06	0.0834084755630776	31.9047619047619
chr7	151174648	151174648	*	1.51846708716497e−06	0.0342027277222966	32.2258592471358
chr14	100883428	100883428	*	5.71978412508625e−06	0.0781769936084471	32.367244905659
chr14	99040860	99040860	*	3.52457275698086e−07	0.012747594857546	32.4862385321101
chr4	16256532	16256532	*	2.24632470908301e−06	0.0438748689544813	32.5367813661219
chr	1769483	1769483	*	9.05358605864304e−06	0.0975759257953621	32.6356641697454
chr13	110639275	110639275	*	2.57719038357563e−07	0.0101289543974943	32.861744866443
chr17	3697140	3697140	*	4.32888016037048e−	0.00280442315019776	32.8896916690433
chr7	1759295	1759295	*	1.85698686464837e−06	0.0390985069340864	32.9693295539253
chr14	100883427	100883427	*	6.79411347487157e−	0.0646874616877994	33.0891866781425
chr14	23580472	23580472	*	1.17801547070557e−06	0.0293029956589775	33.1623931623932
chr7	89228934	89228934	*	5.7213360363665e−06	0.0780769936064471	33.2051282051282
chrX	154398474	154398474	*	7.01302409398192e−06	0.0857039652361363	33.3094213295074
chr11	8313237	8313237	*	1.48263199504415e−	0.001165417344545	33.5684729064039
chr9	63781672	63781672	*	8.12820452220173e−11	1.45207835804088e−08	33.9124889054726
chr2	240913913	240913913	*	6.46215759307487e−05	0.0825774316790542	34.0201465201465
chr18	14179615	14179615	*	7.034048053248E−06	0.0857039652361363	34.3085105382979
chr10	112842985	112842985	*	8.1277498359986e−06	0.092960282194563	34.3089030988867
chr6	170168066	170168066	*	5.95588546571617e−06	0.0792393419764003	34.4703389830508
chr7	1759296	1759296	*		7.75922693444485e−09	34.5021037868163
chr7	158973450	158973450	*	2.98695163460174e−06	0.0522530956076654	34.8612172141564
chr10	103456497	103456497	*	1.82355256133036e−12	4.47968395537135e−07	35.2497195764178
chr4	16360900	16360900	*	1.76438224548031e−06	0.0370817576772098	35.9553123575011
chrX	142971347	142971347	*	6.97291165033871e−07	0.0209396233880114	35.9930110658125
chr12	5889121	5889121	*	4.19457086012719e−07	0.0146509416988426	35.993265993266
chr12	5889122	5889122	*	3.24681426757704e−10	4.90797352813289e−07	36.144576313253
chr12	132849363	132849363	*	1.42672418428752e−07	0.00678909204811647	36.3135018465184
chr7	2714483	2714483	*	2.63523298089516e−06	0.0485487901506917	36.3550063907701
chr20	18227868	18227868	*	1.89319155567353e−06	0.0394382373080044	36.5421455939697
chr7	158973719	158973719	*	6.2576347272967e−06	0.0612575515670304	36.7965367965368
chr13	110539274	110539274	*	3.65465862545053e−11	7.42947613146915e−	37.1058067648038
chr7	1769484	1769484	*	9.76538684309593e−10	0.000125152665483359	37.4363861993184
chr18	1458079	1458079	*	2.73473845547679e−06	0.0495141213330146	37.5944737911418
chr7	158973714	158973714	*	4.24186387872778e−07	0.014710151948704	38.110861288556
chr12	13095512	13095512	*	2.26280767494508e−	0.00156809957596004	38.4369114877589
chr7	34939470	34939470	*	5.6620661324981e−07	0.0185443646548435	38.6921850079745
chr18	14458993	14458993	*	2.39376120192636e−06	0.0456700741763526	38.9165835825856
chr12	132843222	132843222	*	3.38268172042491e−07	0.012309927489174	39.1886597570966
chr11	50279012	50279012	*	4.30883976796929e−06	0.0683240355080424	39.314606741573
chr7	1764018	1764018	*	1.22055159642621e−06	0.029981563301402	39.3270049898556
chr2	9245093	9245093	*	5.43430493275317e−06	0.0761543594497282	40
chr7	100847116	100847116	*	3.64229197056794e−06	0.0594808172903373	40.0865800685801
chr16	8902070	8902070	*	1.16752761648575e−07	0.00603770316698126	40.1101946516082
chr6	59258597	59258597	*	1.15209360303514e−07	0.00601061200563929	40.3251231527094
chrX	100406950	100406950	*	2.06158285185436e−06	0.0414358834417034	40.3743315508021
chr1	151852674	151852674	*	2.25762286037092e−10	3.6957728469288e−05	40.4853833425262
chr13	110569482	110569482	*	2.34599317702309e−06	0.0016278046100216	40.655443109808
chr4	141247044	141247044	*	6.53851793505649e−06	0.0829440737732974	40.8432147562582
chr16	88006372	88006372	*	1.03047403394103e−06	0.0286503210330817	40.983606557377
chr11	76025058	76025058	*	3.34086790592703e−06	0.0552595792448322	41.2037087037037
chr11	15072035	15072035	*	2.46322768519033e−06	0.0457065024985586	41.6014924302217
chrX	153606805	153606805	*	6.09115908826083e−16	2.39396643050005e−10	41.5272655834358
chr2	90245085	90245085	*	6.72528026354226e−06	0.0842118280264895	41.8832891246654
chr16	5185557	5185557	*	80.9899999484523e−07	0.0231778715158391	42.0454545454545
chr6	118688131	118688131	*	7.63612132900429e−06	0.0889675592284456	42.2027792074112
chr2	90245076	90245076	*	1.00357097038975e−06	0.0255309377110954	42.3719958202717
chr10	103458498	103458498	*	8.87300946042895e−32	1.74364899011705e−25	43.0265356767126
chr19	7554442	7554442	*	1.07846741577602e−06	0.0275235483198223	43.2249322493225
chr8	71668017	71668017	*	2.09855280766445e−06	0.0417962736452911	43.4541203974284
chr17	40014144	40014144	*	3.36438847291582e−06	0.0564880066848964	43.466172381835
chr8	44311572	44311572	*	4.0038430934445e−13	1.02921153475302e−07	43.5798882855708
chr8	23344333	23344333	*	4.25627604564246e−09	0.000389406859984431	43.7608434538903
chr16	8902069	8902069	*	6.27942692393337e−06	0.60916294830927e−05	43.8190490844745
chr	1489565	1489565	*	2.52602036762017e−06	0.0474635582490079	43.9298245614035
chr13	114103908	114103908	*	1.30173061753499e−07	0.006503520677028	44.4522144522144
chr17	1517860	1517860	*	2.98607382645789e−06	0.0522530956076884	44.4577831132453
chr12	131585340	131585340	*	6.1255332108104e−07	0.0195200777501359	44.4612455197133
chr11	3617050	3617050	*	1.56072555590698e−08	0.00119493749660254	44.5192307692308
chr2	194200772	194200772	*	1.83775461270924e−07	0.0079081760166216	45.2539912917271
chr4	4037085	4037085	*	4.62505552560616e−06	0.0699142129015274	45.7295968934993
chr4	87927456	87927456	*	5.15240314806726e−11	9.49225138804652e−06	45.890823738383
chr5	127476033	127476033	*	4.65493880975636e−06	0.0700113094379202	46.9618445215282
chr17	1526574	1526574	*	1.1458599113541e−14	3.75291210601205	47.354790972993
chr14	57366558	57366558	*	3.2636021578232e−09	0.000326442866774403	47.4576271186441
chr11	1872395	1872395	*	1.42527266706343e−06	0.033343162468073	47.7839514149115
chr1	15219581	15219581	*	6.8864757723842e−19	4.51090714281087e−13	47.8225778225478
chr5	92374220	92374220	*	3.15624406540522e−07	0.0116585804267786	48
chr5	1705803	1705803	*	1.11374371059014e−09	0.000136769700895759	48.206599713056
chr1	15219580	15219580	*	1.94284510722883e−08	0.00143171825929577	48.250055029716
chr11	71589734	71589734	*	1.19350487339216e−08	0.0294398551248445	48.2698412898413
chr5	17095802	17095802	*	3.84338065355121e−19	2.8325578257214e−13	48.2810474462017
chr5	1926307	1926307	*	5.99468471007297e−06	0.0792393419764003	48.936170212786
chr19	3831767	3831767	*	9.92393775997153e−08	0.00541713487976249	49.8668280871671
chr7	34939471	34939471	*	1.98497366394687e−10	3.34345958133233e−05	50.2493287303414
chr1	153348938	153348938	*	4.40575383775038e−08	0.00282386085625328	51.3026452489411
chr9	63703081	63703081	*	5.77752168001692e−06	0.0783425388589402	51.8048128342246
chr9	122620923	122620923	*	7.429894660979E−06	0.662377435168626	52.6705276705277
chr12	95638101	95638101	*	5.71858504259934e−06	0.00344016660304417	52.7537633758539
chr10	111583910	111583910	*	4.95882892516532e−06	0.00312952931851425	52.7851458885942
chr10	131724260	131724260	*	1.23715424482524e−18	6.07802535514812e−11	52.8746177370031
chr19	3623239	3623239	*	8.97578013043069e−07	0.0248428819074295	53.0175792184528
chr10	132803598	132803598	*	2.13628040558972e−06	0.0424044251560342	53.2599928859333
chr11	71589733	71589733	*	0.0482592293217e−08	0.00357394427630267	53.3441316769469
chr16	30710079	30710079	*	1.01993186164307e−06	0.0266055408862385	53.4782608695652
chr2	130697316	130697316	*	2.65157858657125e−08	0.00179677883512725	54.689366786141
chr22	50471299	50471299	*	1.39941300365533e−11	3.17305663571782e−06	55.8558558558559
chr11	17567503	17567503	*	2.08475930379464e−06	0.041662902941552	56.7846100996221
chr19	3823238	3823238	*	1.78037894556353e−06	0.0381670970021494	57.0016939582157
chrX	144231760	144231760	*	7.22087983908766e−06	0.0570543809739352	57.133152173913
chr11	71590085	71590085	*	5.33441584465839e−06	0.00327885938881253	57.7435257561811
chr5	149427683	149427683	*	5.39341347686118e−07	0.018045942850302	59.3939393939394
chr10	132803137	132803137	*	5.97330999592518e−06	0.0792393419764003	60.3435571520678
chr5	150098459	150098459	*	3.09372760861357e−18	1.82385977762572e−12	61.1111111111111
chr22	50471298	50471298	*	3.39559018818451e−09	0.000333734627440756	61.2155745489079
chr22	37103346	37103346	*	5.92202191736626e−07	0.0190778000789806	61.7066287653523
chr11	71612643	71612643	*	9.79275135454918e−08	0.00539548292708294	63.1568828505511
chr11	71602277	71602277	*	2.51302646524811e−07	0.010078342199198	63.519882179676
chr11	71590064	71590064	*	4.08725788966055e−09	0.000352473072032365	64.4620720856583
chrX	144230634	144230634	*	5.29944949997185e−06	0.0753204952188082	64.872325741891
chr10	131659639	131659639	*	5.35757653519535e−10	9.94851868850657e−05	66.2058371735791
chr10	131659638	131659638	*	7.45230122527055e−10	9.95497765261251e−05	67.9128856624319
chr14	90556073	90556073	*	1.05901238808213e−11	2.49729819730085e−06	69.3899018232819
chr11	71602082	71602082	*	1.86133735136998e−11	3.91901056085249e−06	73.0769230769231
chr14	90556042	90556042	*	4.88075839593414e−17	2.50860756203605e−11	74.3318409985077
indicates data missing or illegible when filed

TABLE 3
Normal versus Mild Preeclampsia
chr	start	end	strand	pvalue	qvalue	meth.diff
chr2	85868078	85868078	*	3.5194993513349e−08	0.0024216712743042	−67.3618007282357
chr17	78802280	78802280	*	6.13863318081308e−08	0.00347585652497927	−60.3174603174603
chr4	27977807	27977807	*	6.00742796105181e−06	0.0750690131438125	−58.6326950466846
chr11	68373945	68373945	*	5.63363012523847e−06	0.0718853055408169	−54.2424242424242
chr4	169154580	169154580	*	2.24673501570411e−06	0.00172010475820479	−53.5947712418301
chr7	151724728	151724728	*	1.73522362030776e−06	0.0338074675249276	−52.9741560232656
chr5	12865273	12865273	*	7.21504868351372e−07	0.0195713318720594	−52.8753292361721
chr1	172748286	172748286	*	6.19845329082826e−11	1.34773564953995e−05	−52.6436781609195
chr22	45185017	45185017	*	7.60715104015446e−06	0.0842174212589503	−52.5516447131649
chr9	580319	580319	*	4.7056910445104e−08	0.00291084454940201	−52.3475609756098
chr13	112760954	112760954	*	4.16617535963161e−09	0.000435507455756272	−51.0442773600668
chr4	128742393	128742393	*	6.14079857577175e−08	0.00347585652497927	−50.3827227993439
chr6	167179087	167179087	*	2.08281818520008e−08	0.00161738984826254	−49.4380360839988
chr19	55209873	55209873	*	1.79574052904895e−06	0.0348615702805375	−46.0450636430594
chr5	102752373	102752373	*	4.07786605470401e−07	0.0129796760687627	−47.779013718947
chr5	112628386	112628386	*	6.91364726458152e−08	0.00383479837203912	−45.9870782483848
chr1	205850058	205850058	*	1.99409658808528e−10	3.737744217138023e−05	−44.9964763918252
chr1	205849819	205849819	*	4.06367622655203e−07	0.0129796760687527	−44.7817836812144
chr1	3427237	3427237	*	4.94921003694122e−16	6.72569680983604e−10	−44.6690873081367
chr7	1005317	1005317	*		0.00553706552855676	−44.6091115815886
chr1	205850082	205850082	*	1.0874671424508e−08	0.000985204269798062	−44.0997800014194
chr1	205850052	205850052	*	5.92294789074457e−13	2.92689146798765e−07	−44.0811772681053
chr1	81426518	81426518	*	7.0075613123647e−07	0.0192381412046252	−44.055171984823
chr1	205850104	205850104	*	8.32717753369843e−10	0.000115648611832455	−43.7239521965999
chr22	31248629	31248629	*	1.30305810600284e−10	2.52968906182125	−43.672514619883
chr1	205850063	205850063	*	8.95932248908814e−10	0.000118782566448294	−42.9183300066534
chr15	77908672	77908672	*	8.25765425136711e−06	0.0887089768390727	−42.8461538461538
chr9	28387669	28387669	*	6.78713472366506e−06	0.0805531232868902	−42.7870166299339
chr19	38499706	38499706	*	4.0105344158003e−08	0.00269140230907561	−42.5654786210115
chr18	31045897	31045897	*	5.76796584170878e−06	0.0730847523432371	−42.5531914893617
chr3	134916416	134916416	*	3.86432738108714e−06	0.0569257731851756	−42.3262032085562
chr5	64638081	64638081	*	1.50636275359321e−09	0.000186096533876145	−42.0508274231678
chr8	141170848	141170848	*	2.44040133950921e−08	0.00921213204921482	−41.7976760082023
chr10	132731408	132731408	*	1.81067249455354e−08	0.00757799771517082	−41.6142557651992
chr22	19586819	19586819	*	2.19977848674928e−08	9.19807596799142e−07	−41.2582675910431
chr2	3322275	3322275	*	7.75243231045903e−06	0.085019457373052	−40.9548921744044
chr1	205850124	205850124	*	1.15667512203153e−08	0.000998003897585052	−40.8009197164208
chr15	42015868	42015868	*	1.7744955377753e−07	0.0075357474276211	−40.5063291139241
chr2	1767572	1767572	*	5.10151486155981e−06	0.0678011781053311	−40.4225598044351
chr1	205849960	205849960	*	2.10584671188882e−06	0.0388030739134601	−40.3949967083608
chr1	205849963	205849963	*	5.83181082753211e−10	8.56767720789663e−05	−40.342413115874
chr1	205879990	205879990	*	4.71467738890653e−07	0.0143172741726291	−40.1776700989299
chr8	30580122	30580122	*	1.7559577996275e−07	0.00751573996245069	−39.9242424242424
chr3	184636251	184636251	*	1.56812206485296e−11	5.14503005743297e−06	−39.5348837209302
chr20	29324391	29324391	*	3.20372220684376e−06	0.0518442908913507	−39.4916911045943
chr9	138148159	138148159	*	6.72911273393646e−06	0.0800392473571132	−39.4342290893015
chr6	22061410	22061410	*	8.05457544212907e−08	0.00420988959320805	−39.3617021276596
chr1	3019823	3019823	*	7.37248524298958e−06	0.0840150180277284	−39.0967601178139
chr19	38499705	38499705	*	3.30451465849165e−08	0.00230289677733946	−38.4651686761661
chr1	205849983	205849983	*	2.3854596580419e−07	0.00913087271952168	−38.3088482673545
chr1	205849966	205849966	*	6.20884365998158e−09	0.000624997609944309	−38.2948958791319
chr1	205850010	205850010	*	3.82525017154793e−07	0.0126019366218156	−38.1735911147676
chr10	129046340	129046340	*	3.81908059518629e−06	0.0565902170382831	−37.5781042676896
chr16	66566575	66566575	*	1.94541872919422e−09	0.000224996947814191	−37.5532338613315
chr22	20667703	20667703	*	4.66511945897494e−06	0.0646024380866661	−37.4254049445865
chr11	3513408	3513408	*	3.45106486370145e−06	0.0539057720069594	−36.8334917115405
chr1	102235052	102235052	*	1.14434594785233e−06	0.0257160290067815	−36.8228849665246
chr1	163457381	163457381	*	2.3776888675431e−08	0.00177049031866852	−36.6312508853945
chr12	33075435	33075435	*	1.67889766512072e−06	0.0330656096982751	−36.4961581183368
chrX	26557760	26557760	*	4.33954022125379e−06	0.061750680383889	−35.8247524361857
chr1	3037601	3037601	*	1.49094094034888e−07	0.00668118895524527	−
chr6	168365090	168365090	*	2.21567226328442e−09	0.000250914451693387	−35.6463161861692
chr11	3509333	3509333	*	4.51713330314854e−06	0.0630098770047527	−35.2490421455939
chr1	205849961	205849961	*	7.78431846852264e−08	0.0041787742269341	−35.0458960876927
chr10	16779672	16779672	*	6.17403446716301e−06	0.0759290860445895	−34.7663247028171
chr1	163457380	163457380	*	3.82072721438564e−06	0.0565902170382831	−34.7237642083003
chr3	184636250	184636250	*	8.17991192563509e−14	4.94046139540635e−08	−34.7050728238847
chr21	44647264	44647264	*	2.01864170955556e−06	0.037887088288121	−34.4262295081967
chr7	6827387	6827387	*	2.22558665429373e−07	0.00896651164504118	−34.3949044585987
chr16	89227775	89227775	*	3.39275313069829e−06	0.0533012675267715	−34.3201288586803
chr1	3427238	3427238	*	1.9052592842424e−07	0.00778688637634848	−33.7758534316181
chr1	205850256	205850256	*	4.90916426283835e−08	0.00293242939139185	−33.3575757575758
chr20	63035180	63035180	*	1.77780270352192e−09	0.000210081167672482	−33.1344118606928
chr10	15534759	15534759	*	5.38024743424222e−07	0.0156394701158399	−33.0625272015088
chr1	3021702	3021702	*	1.37827391839617e−06	0.0292655684785	−32.5902915272752
chr12	132397001	132397001	*	2.38615965648831e−10	4.28858148243318e−05	−32.1428571428571
chr8	130356676	130356676	*	1.38664713585591e−06	0.0293287955185215	−31.9586312563841
chr11	3514122	3514122	*	7.54594048629173e−07	0.0199116636412291	−31.8506493506494
chr14	102870468	102870468	*	1.43061391863649e−07	0.00668118895524527	−31.8193660442901
chr1	3021701	3021701	*	2.34477902190371e−06	0.0420649796789997	−31.7332626119755
chr8	141170849	141170849	*	1.6090718815087e−11	5.14503005743297e−06	−31.6405612661119
chr1	3064014	3064014	*	1.29773472846529e−07	0.00624264840344132	−30.8998133003274
chr1	3037602	3037602	*	7.1015688340521e−06	0.0828380326767182	−30.7593769215003
chr11	44964070	44964070	*	6.70420308835351e−06	0.0799178351719117	−30.2449810198411
chr1	3037639	3037639	*	2.02816676163321e−07	0.00804719427047447	−29.9275760909424
chr3	56718947	56718947	*	4.37372185793998e−09	0.000448576670917393	−29.9081920903955
chr10	48143	48143	*	4.7167014617407e−08	0.00291084454940201	−29.5219159182558
chr2	27442609	27442609	*	1.00949626362308e−05	0.0999525235500815	−29.4665718349929
chr11	12263379	12263379	*	1.06493930045846e−06	0.0247743913021197	−29.2680210282456
chrX	44310321	44310321	*	9.18197996635613e−07	0.0225842389002894	−29.1681079211346
chr4	3746827	3746827	*	6.96526254368032e−06	0.0814227807271212	−29.1328828828829
chr3	164004174	164004174	*	2.03729755505485e−06	0.037887088288121	−29.0270070292268
chr21	46296433	46296433	*	7.47949420766429e−06	0.0842174212589503	−28.9809863339275
chr12	124373681	124373681	*	1.60188462882896e−08	0.00132565058057724	−28.8348985980672
chrX	4430373	4430373	*	6.10455334794975e−07	0.0173732839635593	−28.8343377053054
chrX	44310310	44310310	*	9.56085149428863e−07	0.0230980568757846	−28.7814078522341
chr22	49056296	49056296	*	1.59042499855762e−06	0.0320192253929983	−28.3665407076605
chr11	3514056	3514056	*	8.16779083433761e−06	0.0880517957435133	−27.7608401084011
chr4	80197352	80197352	*	4.3595024220583e−06	0.06187276741191	−27.250936329588
chrX	44210362	44210362	*	3.61680160045793e−12	1.40429400735197e−06	−27.1900464797228
chr2	170710457	170710457	*	5.08393634459985e−06	0.0677331593128208	−27.1844660194175
chrX	44310376	44310376	*	2.19768070891085e−06	0.0399534971832748	−27.0766928544706
chr3	9946410	9946410	*	4.22291676057923e−07	0.0132686827348675	−26.4705882352941
chr10	129511181	129511181	*	3.88951239902586e−06	0.0570144381016749	−26.1064578722076
chr17	82205372	82205372	*	6.91865435316407e−09	0.000683786208143929	−26.0267261791362
chr8	133062901	133062901	*	1.58663673825078e−06	0.0320192253929983	−26
chr2	54976913	54976913	*	3.92736270705501e−07	0.0128544689445323	−25.6405202995664
chr17	82205371	82205371	*	1.35259720519151e−08	0.00114881448757073	−25.5639097744361
chr8	133062902	133062902	*	4.72055907788727e−06	0.0647977078309181	−25.2553693033002
chrX	44310363	44310363	*	3.0568152516151e−08	0.00215794245032894	−25.074725011956
chr3	9946415	9946415	*	1.91077871827886e−06	0.036444074654644	−24.9589490968801
chr7	128271403	128271403	*	4.15985869640936e−06	0.0599789170971243	−24.9576181394363
chr10	133215684	133215684	*	5.97344697921834e−06	0.075013789176032	−24.7714069209396
chr6	33521699	33521699	*	5.06317248260392e−06	0.0676222637308449	−24.5228245828246
chr16	72877874	72877874	*	2.8222481563497e−07	0.0104361245917236	−24.5461273890897
chrX	44310393	44310393	*	2.04771099783168e−08	0.00161317307268514	−24.5285580665601
chr1	218217201	218217201	*	3.15086638737154e−06	0.0512796334908626	−24.3989397446427
chr11	44834245	44834245	*	9.04979987108319e−06	0.0935221803126185	−24.0559696932665
chr5	13810207	13810207	*	6.8148686533429e−06	0.0805545754455256	−23.8458566737493
chrX	44310356	44310356	*	9.19463608933693e−06	0.0941242256005261	−23.8120693751485
chr4	69658238	69658238	*	4.6858747658948e−06	0.0646481122327589	−23.5000252308624
chr5	142375677	142375677	*	7.43846107282538e−05	0.0842174212589503	−23.4626072248399
chr2	217151607	217151607	*	8.09371351800198e−05	0.0874663966294308	−24.4375
chr8	95073358	95073358	*	8.29263379253513e−06	0.0889090397652541	−23.2689832689833
chr9	138123005	138123005	*	2.56199816863536e−11	7.17415789972436e−06	−22.6535415384916
chr13	29503282	29503282	*	4.65835941888654e−08	0.00291084454940201	−22.0588235294118
chr9	124838915	124838915	*	9.65643759778869e−06	0.0972040970864485	−21.6374269005848
chr6	57576937	57576937	*	7.58831019639142e−06	0.0842174212589503	−20.9876543209877
chr4	8975138	8975138	*	2.36994988415395e−08	0.00177049031866852	−20.8104112471036
chr22	15263712	15263712	*	4.99613504645456e−08	0.00295194141152244	−20.662385877183
chr20	58852896	58852896	*	4.85861078536543e−07	0.0145111594413236	−20.5533887703289
chrX	44310418	44310418	*	3.18657761623654e−07	0.0111421082907189	−20.5533887703289
chr6	165876732	165876732	*	5.00828194858739e−06	0.0672129485204955	−20.0587043265715
chr7	157994682	157994682	*	1.00838152158822e−05	0.0999525235500815	−19.9902959728287
chr19	38976130	38976130	*	8.89952692923179e−07	0.0224868678698749	−19.8201155462185
chr12	1519454	1519454	*	9.51916762309728e−06	0.0963911891443587	−19.2817087507933
chr19	50133862	50133862	*	5.29888080543803e−06	0.0687971092962942	−19.1765842742346
chr14	104851983	104851983	*	5.256835520871E−06	0.0686900397879824	−19.0168490533264
chrX	44310597	44310597	*	1.50262289414418e−06	0.0307064603751643	−19
chr10	125023522	125023522	*	5.25685147171326e−06	0.0686900397879824	−18.3318813333945
chr2	144522921	144522921	*	7.20103464503185e−06	0.0880173843820925	−17.1608925425719
chr2	218608011	218608011	*	3.10787704929277e−06	0.0508212671698537	−16.9748090085485
chrX	44310441	44310441	*	2.25038216983616e−06	0.0407752296295299	−16.7364507924085
chr13	25096005	25096005	*	2.44576073169025e−10	4.28858148243318e−05	−16.0790011839414
chr12	131840590	131840590	*	6.81284556136893e−07	0.0188247134508835	−15.8986928104575
chr5	502298	502298	*	7.32287737422104e−06	0.083759733246032	−15.617780091764
chr20	29324198	29324198	*	8.65861964295904e−06	0.09210626131328	−15.4665314401623
chr2	110212166	110212166	*	8.84368595140172e−07	0.0223591991436254	−15.2765957446809
chr13	25096001	25096001	*	6.32943826954945e−06	0.0764564355116043	−14.9168062534897
chr13	25096012	25096012	*	1.31589577987864e−07	0.00627448426927547	−14.4155844155844
chr17	82861233	82861233	*	5.45495194711744e−06	0.0704097356507249	−14.2081447963801
chr2	869008	869008	*	2.60457766326837e−06	0.0449326131402911	−
chr17	8226706	8226706	*	6.26829788264717e−06	0.0762540805535126	−
chr20	62799630	62799630	*	3.24363770746426e−06	0.0521339513270659	−13.3646552700102
chr3	158104239	158104239	*	2.42084069740981e−06	0.0427849084025574	−13.2600919775166
chr13	25096033	25096033	*	1.7308866422865e−06	0.0338074575249276	−12.9296066252588
chr2	63055926	63055926	*	6.04540592624519e−06	0.0752135681501139	−12.1212121212121
chr10	126918347	126918347	*	1.31421353359581e−06	0.0282362361911788	−11.9043006586594
chr16	57591317	57591317	*	1.63522053441961e−06	0.0325593013379642	−11.7856526243094
chr10	127194852	127194852	*	1.14487449618787e−06	0.0257160290067815	−11.3061760492386
chr5	141189533	141189533	*	4.23822707387975e−06	0.0606264329849779	−10.4203389830508
chr17	79079547	79079547	*	8.0514906785991e−06	0.0871834336493453	−9.15492957746479
chr3	12554272	12554272	*	5.15835485708764e−06	0.0680574058614823	−9.03543719857412
chr17	18698467	18698467	*	3.37397705540429e−06	0.0531599326213426	4.81481481481481
chr12	46831525	46831525	*	1.46787107548532e−06	0.0301094921791514	5.1640350877193
chr22	20268607	20268607	*	7.75085541403057e−06	0.085019457373052	6.03015075376884
chr2	227324859	227324859	*	8.76241329583653e−06	0.0923419046394581	6.82926829268293
chr3	49686913	49686913	*	6.63026139255662e−06	0.0792101153645342	7.18562874251497
chr19	48837317	48837317	*	5.95262870231243e−06	0.075013789176032	7.8740157480315
chr19	19170227	19170227	*	6.26547971610748e−06	0.0762540805536126	8.37988826815642
chr10	122114233	122114233	*	2.42992025342566e−06	0.00178493207238441	8.55263157894737
chr8	98748402	98748402	*	5.42434191178619e−06	0.0702035636457087	8.60760921679781
chr7	153749251	153749251	*	1.00568402479447e−05	0.0999525235500815	8.66547085201794
chr15	59206996	59206996	*	5.61937943310756e−06	0.0718721802791079	9.34258949481432
chr14	56805622	56805622	*	9.01381221769034e−07	0.0224868678698749	9.46401394606871
chr7	150737454	150737454	*	2.78826334492208e−09	0.000309313661132254	9.51231608530336
chrX	9890831	9890831	*	3.55431923686348e−06	0.0548877161802339	9.69402024948258
chr14	76131559	76131559	*	6.50249846229761e−06	0.0780472704441361	10.3896103896104
chr15	78810637	78810637	*	1.67390804690697e−06	0.0330656096982751	10.4477611940298
chr8	98949296	98949296	*	3.31967115855298e−06	0.0525531981597828	10.6397371472748
chr21	29019121	29019121	*	4.11629724581241e−06	0.0595086743219646	10.8333333333333
chr14	65279792	65279792	*	3.78902349229063e−06	0.0565902170382831	10.9355742296919
chr4	182895022	182895022	*	8.07231904374769e−07	0.0208949057036708	10.9874881743927
chr14	103158052	103158052	*	9.25268817307087e−06	0.0942489382327313	11.1216181931407
chr22_	147890	147890	*	2.99931586206499e−06	0.0495550248656549	11.1397962767826
Kl27073
chr16	87644983	87644983	*	2.67899555916917e−06	0.0452938349135597	11.1776967170401
chr6	63275283	63275283	*	6.09301763917949e−06	0.0756170476360577	11.1965358784023
chr17	1907284	1907284	*	5.13254724852875e−06	0.0679910831660104	11.2093171817718
chr2	233446634	233446634	*	8.76571078818957e−06	0.0923419046394581	11.3687985654513
chr2	11640232	11640232	*	3.03783376535419e−06	0.0498881504472339	11.4973614775726
chr20	2656783	2656783	*	5.57071321764643e−06	0.0714177784668368	11.5076228119706
chr5	52787012	52787012	*	4.4119824134481e−06	0.0624545294048918	11.5702479338843
chr8	7355190	7355190	*	5.35951186567492e−07	0.0156394701158399	11.5772188564551
chr4	182271973	182271973	*	7.59426845622977e−06	0.0842174212589503	11.747891833895
chr8	142317199	142317199	*	6.71634686800095e−08	0.00376376738969701	11.8573374191352
chr17	1934527	1934527	*	8.04223173595121e−06	0.0871834336493453	12.0394442864687
chr14	70594590	70594590	*	6.16017019652066e−06	0.0759290860445895	12.0613955513625
chr16	1656909	1656909	*	6.96339601592992e−06	0.0814227807271212	12.2328666175387
chr17	34581157	34581157	*	4.03611186194179e−06	0.0586012874071419	12.280701754386
chr2	210000856	210000856	*	3.33071513702361e−07	0.0113868017253078	12.4065640135907
chr9	60546715	60546715	*	5.03211836839035e−06	0.0673730497335436	12.4318026298755
chr2	23381331	23381331	*	5.23000265546139e−06	0.0686693537548019	12.6325799136859
chr8	7355178	7355178	*	1.96532863559155e−07	0.00785520757265003	12.6415175870711
chr5	140113195	140113195	*	1.11816262026038e−06	0.0254312941380053	12.6435781413157
chr10	80315534	80315534	*	1.82075329657621e−06	0.0350964660636285	12.9637817324343
chr11	124953770	124953770	*	4.92000761841497e−06	0.066860123729955	13
chrX	9890823	9890823	*	1.1633577890612e−09	0.000147063954011078	13.0329457364341
chr14	84399673	84399673	*	1.91946182418947e−06	0.0364816806920338	13.3036542141972
chr2	123826928	123826928	*	5.97866405297445e−06	0.075013789476032	13.3259328836137
chr5	28077923	28077923	*	5.30301458359882e−06	0.0987971092962942	13.3566928121384
chr12	125025025	125025025	*	9.18126714338813e−07	0.0225842389002894	13.4502923976608
chr8	7355019	7355019	*	2.65882870076947e−06	0.0452938349135597	13.5951885951886
chrUn_	1964	1964	*	2.3152978161566e−06	0.0418120797258048	13.6407727750384
JTFH01
chr6	3761436	3761436	*	4.42004631218611e−07	0.0138082602428604	13.6571916744331
chr5	1974697	1974697	*	7.57778168074868e−06	0.0842174212589503	13.8895534290271
chr16	67909149	67909149	*	7.95236576760233e−06	0.0866277817184845	13.9528377298161
chr3	184039229	184039229	*	6.83171509345077e−06	0.0805545754455256	14.069581444948
chr20	3697355	3697355	*	3.94920081985913e−07	0.0128544689445323	14.2098487286772
chr8	7355171	7355171	*	4.83087840534921e−08	0.00291772924810652	14.2435063802068
chr16	88777480	88777480	*	2.68307969429307e−06	0.0452938349135597	14.2811780582678
chr1	11858760	11858760	*	2.64990567816795e−06	0.0452938349135597	14.2857142857143
chr12	53945167	53945167	*	5.67778816255337e−06	0.0721102179241014	14.5084201486128
chr3	166976648	166976648	*	4.70085258764814e−06	0.0646905627184063	14.5181525915471
chr1	64723293	64723293	*	7.55515993400678e−06	0.0842174212589503	14.8208556149733
chr3	140562546	140562546	*	3.80868590924468e−06	0.0565902170382831	14.94708994709
chr7	56814596	56814596	*	2.41315458749695e−06	0.0427849084025574	14.9909164943933
chr8	7355160	7355160	*	7.20572670030231e−06	0.0830173843820925	15.0429099349479
chr20	20365677	20365677	*	4.94057119021423e−06	0.0669721407005376	15.0537634408602
chr15	99997598	99997598	*	6.92179009759833e−06	0.0813341226839447	15.0636980592928
chr13	31609131	31609131	*	1.24378725749963e−06	0.0270437857433912	15.1135005973716
chr5	975987	975987	*	2.44683883092541e−06	0.0427849084025574	15.2380952380952
chr2	239939101	239939101	*	2.53035213850558e−06	0.0439438414227893	15.4233608336952
chr13	113453322	113453322	*	4.67067949625201e−06	0.0646024380866661	15.4558800473197
chr16	86734074	86734074	*	2.61208537226382e−06	0.0449326131402911	15.7174969073409
chr12	130437316	130437316	*	4.68840585590966e−07	0.0143172741726291	15.8701092595984
chr1	18938585	18938585	*	9.52484532910695e−06	0.0963911891443587	15.9091952439976
chr13	31339964	31339964	*	3.46548133975747e−06	0.0539758549115726	15.9462915601023
chr19	1625320	1625320	*	9.25421409217928e−06	0.0942489382327313	16.0058617601563
chr1	164576420	164576420	*	4.50915568038441e−06	0.0630098770047527	16.0969568294409
chr7	101245417	101245417	*	8.43570160713426e−06	0.0900875588758716	16.1290322580645
chr1	185649037	185649037	*	5.65890062099215e−06	0.0720386530776885	16.1350112184177
chr7	157441764	157441764	*	8.52564447818141e−08	0.00433116603623758	16.2108033875029
chr5	27120204	27120204	*	7.7351057034027e−06	0.085019457373052	16.2820512820513
chr17	9116020	9116020	*	2.04542042936614e−06	0.037887088288121	16.6060606060606
chr19	12495880	12495880	*	2.3693398810559e−06	0.0423657767191011	16.6666666666667
chr1	145156891	145156891	*	6.75413310027288e−07	0.0188247134508835	16.8543543543544
chr6	27235585	27235585	*	8.69555613449339e−06	0.0923185116238665	16.8707899571341
chr7	81071315	81071315	*	6.50420961415866e−06	0.0780472704441361	16.9085218380993
chr4	153452347	153452347	*	4.82430577592971e−06	0.0658890349368055	17.0630725863284
chr5	95879781	95879781	*	1.52151529640109e−07	0.00672407584453604	17.1367177007789
chr5	1876520	1876520	*	4.00331628585466e−06	0.0583408112075746	17.1789502127947
chr12	42611025	42611025	*	1.18993988524624e−07	0.00588022208162228	17.2442280099135
chrX	69128579	69128579	*	3.26090515374672e−06	0.0521339513270659	17.2504206942433
chr3	161161933	161161933	*	1.58290684073595e−06	0.0320192253929983	17.44.5682257441
chr14	83092425	83092425	*	3.84856319481191e−06	0.056847608020966	17.4449675531238
chr17	62138102	62138102	*	1.66242940637828e−07	0.00722927243522119	17.5211736237145
chr1	187703924	187703924	*	2.98553060428667e−06	0.0494776513261761	17.68144323779
chr9	93259190	93259190	*	3.50462162235937e−06	0.0544295174133111	17.7272727272727
chr8	7355161	7355161	*	3.98595037543776e−06	0.0582438855271366	17.7704396746774
chr8	7355162	7355162	*	7.45369592985245e−07	0.0197641983119014	17.8508173616816
chr2	83181431	83181431	*	9.45177758089121e−07	0.0229364851910676	17.9188712522046
chr3	75444754	75444754	*	6.93435129599055e−06	0.0814115906730275	17.972972972973
chr19	38312833	38312833	*	7.46242227283568e−06	0.0842174212589503	17.9806882102434
chr11	8597404	8597404	*	8.01557190189945e−06	0.0871416746789511	18.111611489776
chr20	31996615	31996615	*	2.02108559825635e−06	0.037887088288121	18.1457431457432
chr7	124033873	124033873	*	4.49103188953225e−06	0.0630098770047527	18.4265876335476
chr16	1129530	1129530	*	8.9506136900616e−06	0.0928502159932842	18.4817226790892
chr1	51847682	51847682	*	3.32580055148821e−06	0.0525531981597828	18.5538096436662
chr16	1552139	1552139	*	1.00722943110529e−05	0.0999525235500815	18.6243894627272
chrX	75523136	75523136	*	9.00356341640296e−06	0.0932215922404464	18.7106918238994
chr3	196170941	196170941	*	8.91496448738382e−06	0.0928502159932842	18.7320231910248
chrX	34656908	34656908	*	4.63214185737443e−06	0.0643971260169504	18.7730911330049
chr6	1410032	1410032	*	6.39716253879425e−10	9.0159288953031e−05	18.7904998227579
chr14	102210659	102210659	*	9.69287010965831e−06	0.0972415423588671	18.8873156529901
chr16	50283745	50283745	*	5.46617803547497e−06	0.0195713318720594	18.9184300958582
chr1	1049815	1049815	*	9.69593584253245e−06	0.0105974382457425	18.9326334208224
chr17	3911398	3911398	*	7.27294592850256e−07	0.0195713318720594	18.9765227739911
chr14	83092419	83092419	*	2.88536804578315e−07	0.0105974382457425	19.1532609326906
chr1	187703949	187703949	*	1.2928834327608e−06	0.0278881831302854	19.1663186942866
chr19	1999311	1999311	*	8.24411713869786e−06	0.0887089768390727	19.1836425021813
chr12	15590179	15590179	*	1.19964810544574e−06	0.0264008744159159	19.2240202768571
chr3	127776110	127776110	*	2.19620718085617e−06	0.0399534971832748	19.3798449612403
chr7	158992589	158992589	*	1.44855505795679e−06	0.0300262709961256	19.6808510638298
chr11	75428589	75428589	*	2.91638004118714e−07	0.0106394515449691	19.7830547341742
chr5	26485521	26485521	*	1.01370811184431e−06	0.0238540614629973	19.8306595365419
chr20	56466014	56466014	*	3.63026718570573e−06	0.0556939703438536	19.886831757202
chr17_	241564	241564	*	5.76537124182039e−08	0.00333396330815266	20.00777000777
GL3835
chr19	41916208	41916208	*	7.8845601894565e−06	0.0860626034115315	20.0155561317086
chr5	73303234	73303234	*	1.09585329697467e−06	0.0250286170567613	20.0578034682081
chr22	40010046	40010046	*	3.72215581649801e−07	0.0124089588893108	20.0913764653141
chr15	79559260	79559260	*	6.82233762843546e−07	0.0188247134508835	20.2613037827485
chr21	8988098	8988098	*	5.88148889293313e−07	0.0170055555348376	20.286252354049
chr12	132283239	132283239	*	9.91702257153166e−06	0.0992756785483333	20.3076923076923
chr10	1210729	1210729	*	9.37593144196478e−06	0.095262512818719	20.3806303849929
chr2	145861437	145861437	*	3.27733497087068e−07	0.0112752229265504	20.4418769636161
chrX	64204264	64204264	*	7.3346745146342e−06	0.083759733446032	20.4679802955665
chr17	17700424	17700424	*	2.32648100362032e−06	0.0418749170562029	20.4968944099379
chr12	131003900	131003900	*	8.36629413960065e−08	0.00429030982645222	20.5016036133
chr12	125048987	125048987	*	1.94629478365732e−07	0.00783675450469636	20.5169003657418
chr8	144588938	144588938	*	6.28197362878975e−06	0.0762540805536126	20.5499025763152
chr2	145881433	145881433	*	9.15449340302414e−07	0.0225842389002894	20.6399509302807
chr2	239283565	239283565	*	2.30772904442503e−07	0.00902467161059745	20.6826855123675
chr22_	143529	143529	*	5.53148029093805e−06	0.0710824509385189	20.7572914889988
KI27073
chrX	6002109	6002109	*	1.23406567416543e−06	0.026940169100084	20.8955223880597
chr4	74099957	74099957	*	9.18777386103428e−06	0.0941242256005261	20.9017923823749
chr1	193918912	193918912	*	7.13310101183261e−06	0.0828502671356697	21.0335917312661
chr11	75428550	75428550	*	2.44787706648499e−06	0.0427849084025574	21.0592114301415
chr12	7639875	7639875	*	3.67124877806243e−06	0.0558327283785535	21.1223735613979
chr8	650032	650032	*	8.9490725133759e−06	0.0928502159932842	21.1396574440053
chr22	5006106	5006106	*	6.48255006428493e−06	0.0780472704441361	21.25900573554
chr19	1999288	1999288	*	8.93878392187845e−06	0.0928502159932842	21.3197535090722
chr12	7639872	7639872	*	2.47253872752689e−06	0.0430774414393711	21.3312224669604
chr7	2508541	2508541	*	1.28858699058934e−06	0.0278881831302854	21.3495440729483
chr14_	23560	23560	*	5.33381728958435e−06	0.0156394701158399	21.3945352554436
KI27072
chr9	41235357	41235357	*	9.50421630768037e−06	0.0963911891443587	21.4494005832163
chr6	163834777	163834777	*	8.55894646117047e−06	0.0912245071392597	21.5409658766034
chr12	916398	916398	*	3.40478209945203e−06	0.053336098016097	21.5830721003135
chr7	94394113	94394113	*	4.97194331814183e−06	0.0672194785204955	21.588628077506
chr12	7639881	7639881	*	1.43892972296886e−06	0.0300026320652593	21.6060425362751
chr1	91949051	91949051	*	6.61331824427898e−06	0.079181725475721	21.6992586912065
chr16	11665637	11665637	*	4.79991926795308e−07	0.014495153532688	21.7257634761222
chr16	73063946	73063946	*	7.07115324323374e−08	0.00388254454036219	21.9392372333549
chr18	5892049	5892049	*	1.95792237866169e−06	0.0370830089196772	21.9876292264988
chr2	231393387	231393387	*	1.00096505921184e−05	0.0999525235500815	22.0759944587374
chr10	49679809	49679809	*	2.04916612734591e−06	0.037887088288121	22.1973586064675
chr2	52859724	52859724	*	3.02898732644796e−06	0.0498881504472339	22.3829449534666
chr17	41825406	41825406	*	1.09006398198915e−06	0.0250014407241902	22.4517374517375
chr14	83092404	83092404	*	1.13685810194342e−08	0.000997750739317906	22.5073336421183
chr13	113472325	113472325	*	9.01828736005716e−07	0.0224868678698749	22.5661375661376
chr11	62433748	62433748	*	1.62887198291174e−06	0.0325593013379642	22.6699203601763
chr20	3752446	3752446	*	2.75225148985962e−06	0.0463176999691646	22.7400732188116
chr12	117155454	117155454	*	1.16700980412723e−06	0.0258309554975365	22.7479320061813
chr3	12882861	12882861	*	6.13496980156831e−06	0.0759290860445895	22.8274428274428
chr12	131003812	131003812	*	4.28713009541067e−06	0.0611650139297923	22.8840454271251
chrX	50469150	50469150	*	1.08239958475546e−06	0.0249308454255277	23.0175652350492
chr7	158996842	158996842	*	4.7281330174762e−08	0.00291084454940201	23.0547453254877
chr11	75428635	75428635	*	8.9348440257961e−06	0.0928502159932842	23.1064187414644
chr16	73063947	73063947	*	1.48923673268711e−07	0.00668118895524527	23.115319679953
chr5	85941	85941	*	3.25399127254312e−07	0.0112662172965075	23.1763285024155
chr1	1296199	1296199	*	1.00738486641878e−06	0.0238083324559683	23.2140226838531
chr1	1426345	1426345	*	6.04666757853652e−06	0.0752135681501139	23.4122807017544
chr2	131658207	131658207	*	5.00031003540981e−06	0.0672194785204955	23.5658042744657
chr15	40069353	40069353	*	3.20644880283443e−06	0.0518442908913507	23.6060161408657
chr17	62701570	62701570	*	7.28005900000395e−06	0.0834868258031381	23.8938053097345
chr7	1169891	1169891	*	1.73669031418151e−11	5.24458669771092e−06	24.1511961722488
chrX	149938294	149938294	*	1.16360309167929e−06	0.0258309554975365	24.1581137309293
chr16	737792	737792	*	3.63726664722779e−06	0.0556939703438536	24.164868233077
chr7	56814628	56814628	*	3.74384449730062e−07	0.0124089588893108	24.2185299915517
chr17	80187604	80187604	*	8.30191286681125e−07	0.0212864726941878	24.243450767841
chr1	237862175	237862175	*	7.51861077966319e−06	0.0842174212589503	24.2988352042791
chr16	22949123	22949123	*	7.22697368386513e−06	0.0830532677577977	24.3350874697767
chrX	8003769	8003769	*	2.4290362673546e−06	0.0427849084025574	24.5106148331955
chr2	238143670	238143670	*	7.32931884194799e−07	0.0196230345936683	24.5531692978719
chr7	138008755	138008755	*	3.30418247278432e−06	0.0525531981597828	24.7375666838754
chr11	75428433	75428433	*	2.83303008784382e−06	0.0475299731256765	24.782824522214
chr17	4656719	4656719	*	2.89013508024201e−08	0.00206712120074571	24.8588709677419
chr3	80770400	80770400	*	7.46791709938012e−06	0.0842174212589503	24.9122807017544
chr2	239939031	239939031	*	2.04549100511889e−06	0.037887088288121	24.9192492361414
chr11	75428554	75428554	*	1.3935170521631e−06	0.0293598595376156	24.9411589822362
chr22_	152007	152007	*	7.89282518458512e−08	0.0041787742269341	25.0632911392405
KI27073
chr10	21500180	21500180	*	2.95144882844735e−06	0.0491844195334005	25.120504263997
chr9	28147464	28147464	*	6.14879791054759e−06	0.0759290860445895	25.1282051282051
chr7	903302	903302	*	7.64296030512978e−06	0.0844418798976758	25.1683501683502
chr11	75428448	75428448	*	1.34074138481072e−06	0.028692784095583	25.3016935402089
chr17	41825404	41825404	*	1.49951976027687e−07	0.00668118895524527	25.4486613709915
chr18	5892057	5892057	*	8.99086690192464e−08	0.00452521486508727	25.4501624567655
chr9	65675170	65675170	*	1.80644306745188e−06	0.0349445418453209	25.7521186440678
chr16	963991	963991	*	4.40906992247558e−11	9.98612818698948e−06	25.7884749825227
chrX	75523150	75523150	*	1.81232645612604e−07	0.00757799771517082	25.8129338691998
chr9	81508657	81508657	*	7.73228766081226e−07	0.0203048155686783	25.904241478012
chrUn_	95	95	*	8.33036425580317e−06	0.0891377508508351	25.9320629660315
KN7075
chr21	8988154	8988154	*	3.58364466696833e−06	0.055183803132989	25.9720837487537
chr21	8988157	8988157	*	4.15234750866201e−08	0.00275258812518899	26.1049131087749
chr18	5892066	5892066	*	8.60242561973568e−07	0.0218508605236883	26.175873192879
chr17	49014937	49014937	*	4.46218606294549e−06	0.0628379144666358	26.1784602169011
chr6	17988707	17988707	*	1.8538964252743e−08	0.00150408387868641	26.3025210084034
chr16	1458835	1458835	*	5.00408209342069e−06	0.0672194785204955	26.3317510548523
chr6	134904851	134904851	*	2.94324128097444e−07	0.0106658529538984	26.3739008792966
chrX	153794709	153794709	*	7.36435888848838e−07	0.0196230345936683	26.5266731006024
chr7	157994921	157994921	*	3.73604051238001e−06	0.0562556007760165	27.2452830188679
chr22_	135426	135426	*	7.48647549334856e−06	0.0842174212589503	27.2573124205172
KI27073
chr9	133918282	133918282	*	8.45093060640189e−07	0.0215668304535604	27.3138626079802
chr19	39891862	39891862	*	1.06649164676383e−06	0.0247743913021197	27.363330529857
chr7	39753899	39753899	*	4.02968560026197e−07	0.0129796760687627	27.4255007327797
chr16	1459220	1459220	*	5.12889046802757e−07	0.0152219475968595	27.6666666666667
chrX	18354020	18354020	*	7.83764850142076e−07	0.0204825413197894	27.7777777777778
chr7	1169927	1169927	*	3.9798887156175e−08	0.00269140230907561	27.8989898989899
chr7	1169866	1169866	*	1.13802644918111e−08	0.000997750739317906	28.0286200628949
chr20	63273801	63273801	*	6.61704195900469e−07	0.0186366552526771	28.2304900181488
chr10	133367962	133367962	*	4.67884402276295e−07	0.0143172741726291	28.2661782661783
chr7	1169868	1169868	*	1.60957645255484e−08	0.00132565058057724	28.4219292158223
chr11	69649130	69649130	*	6.24597690994856e−06	0.0762540805536126	28.6509376890502
chr22_	135525	135525	*	9.69120461941616e−07	0.0232066951977543	28.7601160482516
KI27073
chr7	157999053	157999053	*	1.94415110855558e−08	0.00155411265411141	28.8235294117647
chr2	171875082	171875082	*	4.438847915766E−06	0.0625716210142209	28.8461538461538
chr1	124207129	124207129	*	9.96140853470557e−07	0.0236453997887907	28.8875598086124
chr16	87833135	87833135	*	1.4445373427969e−07	0.00668118895524527	29.0027408812987
chr19	440707	440707	*	8.7413787366665e−06	0.0923419046394581	29.0465631929047
chr7	1169837	1169837	*	3.32174498286576e−06	0.0525531981597828	29.0989145124025
chr22_	135472	135472	*	9.6300489607308e−06	0.0971183112420548	29.1907930228376
KI27073
chr15	25802207	25802207	*	3.67714271408674e−06	0.0558327283785535	29.2625893853906
chr22	49925463	49925463	*	1.16899912549602e−06	0.0258309554975365	29.3992557150452
chr7	1169926	1169926	*	2.60575011184293e−14	2.02346695835041e−08	29.5172629276676
chr13	27425174	27425174	*	7.58145392462663e−06	0.0842174212589503	29.6590909090909
chr14	100440442	100440442	*	9.29761562457193e−07	0.0226635593157138	29.7067669172932
chr14	104159421	104159421	*	7.26336992160453e−07	0.0195713318720594	29.7872340425532
chr22	42692830	42692830	*	3.71006794805588e−06	0.0561757406414359	29.812030075188
chrX	65667643	65667643	*	2.65560720339432e−06	0.0452938349135597	29.8179535467671
chr5	2564529	2564529	*	3.29438357772334e−10	5.42652260539864e−05	29.9133239266277
chr22_	135492	135492	*	3.25156679174846e−06	0.0521339513270659	29.9465131471501
KI27073
chrX	75523159	75523159	*	3.66123615248956e−06	0.0558327283785535	30
chr12	130407617	130407617	*	4.8374970236187e−07	0.0145111594413236	30.2255639097744
chr7	157999054	157999054	*	1.09309148451844e−10	2.20066602635805e−05	30.2298640326809
chr7	1169850	1169850	*	9.29379730743095e−07	0.0226635593157138	30.2420207144367
chr22_	135541	135541	*	1.86822179371111e−06	0.0357942136078405	30.4802442861109
KI27073
chr22	36564600	36564600	*	2.38359128753439e−07	0.00913087271952168	30.6029945601255
chr8	141275243	141275243	*	4.81264100959551e−06	0.0658890349368055	30.6732970242324
chr12	2106958	2106958	*	3.05697247806909e−07	0.0110046433874195	30.7692307692308
chr7	1169849	1169849	*	7.20857883869583e−06	0.0830173843820925	30.9466445816393
chr22_	135506	135506	*	8.88064814109745e−06	0.0928502159932842	31.0674989420229
KI27073
chr22_	135436	135436	*	8.2675165433281e−08	0.00428003347340884	31.0833390165723
KI27073
chrX	17737212	17737212	*	8.71467580601656e−06	0.092341146520027	31.1657414750198
chr13	23570923	23570923	*	9.94118301283143e−07	0.0236453997887907	31.3373312168068
chr12	132174708	132174708	*	1.66855094619531e−06	0.0330656096982751	31.4325314325314
chr17	80182491	80182491	*	1.71280817284708e−07	0.0073892366134521	31.541344735667
chr15	75723716	75723716	*	1.47612094080113e−07	0.00668118895524527	31.5940929070613
chr6	66289559	66289559	*	2.89477259315682e−06	0.0484163987624443	31.8881827209533
chr7	1169859	1169859	*	2.4020770526729e−07	0.00913087271952168	31.8884408602151
chr17	50199609	50199609	*	3.21417447274026e−06	0.0518442908913507	31.9202695115104
chr7	1169867	1169867	*	1.12262169300261e−09	0.000145293281206181	32.0375494071146
chr19	46468105	46468105	*	1.87012128624676e−06	0.0357942136078405	32.1556500450973
chr7	1170095	1170095	*	4.20782744107426e−06	0.0603591447467553	32.2055137844612
chr5	9725687	9725687	*	2.15438836789597e−06	0.0394301962158631	32.2835738068813
chr7	1169892	1169892	*	9.91961448225861e−15	8.98679708211414e−09	32.3045072595159
chr6	88047529	88047529	*	1.25181084625915e−07	0.00613023500091303	32.3684210526316
chr1	55038977	55038977	*	5.1879492291957e−06	0.0682821296207797	32.3809523809524
chr11	39271373	39271373	*	1.14115450774726e−06	0.0257160290067815	32.5301204819277
chr1	236762554	236762554	*	3.81746516099194e−06	0.0565602170382831	32.5496414782129
chr7	1169860	1169860	*	2.90356982453531e−11	7.14415879972436e−06	32.8602570848681
chr20	62228803	62228803	*	3.79264790329375e−06	0.0565902170382831	32.9224447868516
chr7	231461	231461	*	2.58712129926434e−06	0.0447866455203418	32.9752953813104
chr11	17534318	17534318	*	2.95878842413646e−06	0.0491844195334005	33.1380830533033
chr22	20192292	20192292	*	4.2084376317007e−06	0.0603591447467553	33.2174154958965
chr1	201115212	201115212	*	1.03413887580242e−06	0.0242299362649828	33.7993527508091
chr17	79949679	79949679	*	7.12309477177497e−06	0.0828502671356697	34.0579710144928
chr1	19068359	19068359	*	9.15802753255408e−09	0.0008437245102221	34.0660372927528
chr14	54366583	54366583	*	7.61179222371085e−14	4.94046139640635e−08	34.2294289168622
chr4	39530307	39530307	*	1.94501027008325e−07	0.00783675450469636	34.3064489526637
chr14	24316393	24316393	*	5.15260266122569e−07	0.0152219475968595	34.4149207971385
chr22	24551484	24551484	*	2.51821869600452e−08	0.0018251290285407	34.5132743362832
chr22_	135547	135547	*	5.20520235903264e−08	0.00304239824171702	34.5410628019324
KI27073
chr11	8384488	8384488	*	1.07665556751145e−06	0.0249040695354639	34.7738057576716
chr4	8466568	8466568		8.87870673713955e−06	0.0928502159932842	35.0521581012537
chr7	1169916	1169916	*	1.90021721977424e−07	0.00778688637634848	35.1368363437329
chr15	25802206	25802206	*	7.91816851425781e−08	0.0041787742269341	35.2173913043478
chr22_	135441	135441	*	2.13797389688936e−06	0.0392619693290198	35.5576441102757
KI27073
chr8	693717	693717	*	5.30148593625675e−06	0.0687971092962942	35.6578286499775
chr13	36840187	36840187	*	8.21287776817919e−07	0.0211579846138817	35.6624233851957
chr7	1169865	1169865	*	7.51432431054318e−12	2.72307791432124e−06	35.6695266557409
chr11	121964691	121964691	*	9.12389066304506e−06	0.0940664388454195	35.9448988377099
chr3	147245632	147245632	*	1.76882316043239e−09	0.000210081167672482	35.947585348783
chr7	155830996	155830996	*	3.11335275667481e−06	0.0508212671698537	35.958314772402
chr5	54821934	54821934	*	6.27202011073538e−22	3.40932838454125e−15	36.2454415728556
chr22_	135295	135295	*	4.86220251851834e−06	0.0662401855461066	36.2914212445406
KI27073
chr2	239231857	239231857	*	2.04487229693304e−06	0.037887088288121	36.3297150610584
chr3	156099527	156099527	*	9.16963883562384e−06	0.0941242256005261	36.4396973742768
chr22	18080123	18080123	*	4.45835258666235e−07	0.0138483411848068	36.6120218579235
chr12	91924347	91924347	*	4.70477273752853e−08	0.00291084454940201	36.6836498378879
chr9	81508683	81508683	*	7.54923936734874e−06	0.0824174212589503	36.753463927377
chr5	54821933	54821933	*	2.08418167490149e−21	5.66457027985299e−15	36.8225524475524
chr3	164402171	164402171	*	7.75780061073801e−06	0.085019457378052	37.1922065503227
chr22_	135487	135487	*	6.81715173636459e−06	0.0805545754455256	37.516148970801
KI27073
chr7	1169907	1169907	*	5.55429820086667e−10	8.3866415968105e−05	37.6223776223776
chr7	1169908	1169908	*	3.60840314192082e−09	0.000384597332948461	38.0913978494624
chr12	43225358	43225358	*	5.88670366048575e−06	0.0744157923334262	38.3492637561354
chr17	79088193	79088193	*	3.52427394918426e−06	0.0545787940223735	38.3574879227053
chr18	68880349	68880349	*	2.70694587075847e−11	7.17415879972436e−06	38.4508211300172
chr7	1169917	1169917	*	4.69425789360788e−10	7.29055000210528e−08	38.6078685689581
chr8	144355151	144355151	*	3.19764746170857e−07	0.0111421082907189	38.6296056884292
chr5	58069392	58069392	*	8.23419561991435e−09	0.000771710801062833	38.6675552922591
chr15	20148130	20148130	*	9.13707591786956e−06	0.0940664388454195	39.025974025974
chr14	104945443	104945443	*	1.16290174356482e−06	0.0258309554975365	39.1525773499801
chr10	129899628	129899628	*	6.15352324330379e−15	6.68983233086929e−09	39.5831963710473
chr1	3716165	3716165	*	8.09619152365043e−09	0.000771710801062833	39.9122967964478
chr14	107946866	107946866	*	3.08231924412474e−07	0.0110228886887585	40.0182149362477
chr16	48577998	48577998	*	1.29756787051698e−07	0.00624264840344132	40.1739130434783
chr8	693718	693718	*	5.97422882017508e−07	0.0171823056564859	40.1742160278746
chr11	70782840	70782840	*	3.72801252040681e−06	0.0562556007760165	40.1827872888966
chr12	128157922	128157922	*	6.77675141568647e−07	0.0188247134508835	40.3296913935212
chr13	18250663	18250663	*	7.57785879186208e−09	0.000735562996365631	40.688758934373
chr10	1180765	1180765	*	9.62996488757116e−07	0.0231620852906072	40.893277971047
chr9	138175476	138175476	*	1.20630505355594e−06	0.0264403292184999	41.2218733647305
chr14	104944075	104944075	*	4.52076411415441e−06	0.0630098770048527	41.2292470233104
chr4	3471922	3471922	*	2.66957312530285e−06	0.0452938349135597	41.2751996957018
chr6	67825668	67825668	*	2.84475992745567e−11	7.17415879972436e−06	41.6066115702479
chr10	128403039	128403039	*	6.29865814299345e−06	0.0762540805536126	41.6898120194823
chr5	180296704	180296704	*	3.13838321709385e−07	0.0110853389524534	42.2498554077501
chr11	88069244	88069244	*	5.14081990554248e−06	0.0679910831660104	42.7058111380145
chr15	41847967	41847967	*	1.40527755050629e−06	0.0294933249877442	43.010752688172
chr6	148346764	148346764	*	7.16582237651837e−07	0.0195713318720594	43.4297769740808
chr6	33502742	33502742	*	3.89132376285721e−06	0.0570144381016749	43.940586972083
chr14	23580472	23580472	*	3.78802372723624e−11	8.95253951647558e−06	44.0277777777778
chr2	240642627	240642627	*	3.53239100462726e−09	0.000384025583615735	44.1948991244766
chr2	239231873	239231873	*	8.05974095048366e−07	0.0208949057036708	44.2575650950035
chr7	2250267	2250267	*	1.35028756226762e−06	0.028783756954893	44.2706522998494
chr7	155820590	155820590	*	4.04275136855915e−06	0.0586012874071419	44.8245080787074
chr19	4217590	4217590	*	3.384452201028E−07	0.0114981982991192	46.9534050179211
chr3	149283991	149283991	*	7.44387831389826e−08	0.0040463240197852	47.0008302200083
chr7	155820591	155820591	*	8.91796912180136e−06	0.0928502159932842	47.3914098519777
chr2	180232376	180232376	*	1.06477453395238e−13	5.78787372752045e−08	47.7210652099693
chr14	23580473	23580473	*	1.71813177977381e−12	7.78281338089017e−07	48.0059523809524
chr16	86738680	86738680	*	3.14056875557709e−07	0.0110853389524534	48.1323489198797
chr1	9282145	9282145	*	1.40359488987554e−07	0.00663445618166811	48.1624758220503
chr18	514264	514264	*	1.00811463765718e−04	0.0999525235500815	48.4543010752688
chr15	90825733	90825733	*	1.43789090093169e−16	2.60534999137369e−10	48.7622630000508
chr2	3125151	3125151	*	6.28666306926627e−06	0.0762540805536126	49.2557911571996
chr6	150098459	150098459	*	7.2061840532212e−06	0.0830173843820925	49.4505494505495
chr4	8464685	8464685	*	2.43894396005617e−06	0.0427849084025574	49.546485260771
chr4	39530308	39530308	*	2.53346837886689e−10	4.30355048239587e−05	49.8145604395604
chr2	239231852	239231852	*	2.65292963916663e−07	0.00987720963364853	51.1407829919627
chr7	561796	561796	*	1.83706073120751e−07	0.00762278393825859	51.3027557391672
chr5	30014417	30014417	*	3.823936147381E−10	6.11354490811522e−05	51.8518518518518
chr6	1515332	1515332	*	8.51018543688203e−10	0.000115648611832455	51.854527938343
chr14	68875188	68875188	*	2.43028391168425e−06	0.0427849084025574	52.2673594709495
chr1	42736748	42736748	*	1.63011693494085e−06	0.0325593013379642	52.8028933092224
chrX	144230633	144230633	*	5.98920862096141e−06	0.075013789176032	53.154305200341
chr3	164407943	164407943	*	7.47032136289704e−06	0.0842174212589503	53.5544445799293
chr11	63047952	63047952	*	1.71838145006591e−06	0.033721058513901	53.5555555555556
chr19	22925603	22925603	*	4.7138795638219e−07	0.0132686827348675	53.5714285714286
chr17	1934230	1934230	*	4.58837847730844e−08	0.00291084454940201	54.3565147881694
chr1	21384907	21384907	*	6.21719939671562e−06	0.0762540805536126	55.1418439716312
chr1	151852673	151852673	*	7.83454128121666e−06	0.0856877179041533	56.5939278937381
chr1	151852674	151852674	*	9.2588347154018e−06	0.0942489382327313	57.2693383038211
chr20	17699900	17699900	*	6.0308846261762e−07	0.0172539609726149	57.6095829931859
chr16	88646325	88646325	*	4.76592965227729e−08	0.00291084454940201	58.0498866213152
chr16	4756184	4756184	*	4.4860752513821e−07	0.0138552790985831	58.7208180590392
chr17	1934312	1934312	*	3.5584003055642e−07	0.0119399134954185	59.1778810863714
chr19	22925602	22925602	*	6.41324534532623e−07	0.018156745991534	59.3444227005871
chr17	1934311	1934311	*	7.40598841836225e−11	1.54835689572441e−05	59.876395478158
chr10	29586143	29586143	*	3.46369267902865e−07	0.0116943171482325	60.6714951650807
chr10	29586144	29586144	*	1.44058361680097e−06	0.0300026320652593	60.8906525573192
chr7	83749828	83749828	*	2.39873934363701e−07	0.00913087271952168	61.5900743949524
chr20	17699899	17699899	*	1.59150439114324e−07	0.00697665983085667	63.9708252740168
chr14	104944816	104944816	*	1.45430430995669e−06	0.0300262709961256	66.124287552859
chr10	29585703	29585703	*	1.52353443754042e−06	0.0310171868302878	66.4870689655172
chr22	19443961	19443961	*	4.08318044083221e−07	0.0129796760687627	66.9160702667534
chrX	144230634	144230634	*	1.49666439380404e−07	0.00668118895524527	67.1428571428572
chr14	104950312	104950312	*	2.50669658277121e−07	0.0093971283520804	67.2663383396421
chr9	97132113	97132113	*	1.45829012445645e−06	0.0300262709961256	72.1501545906708
indicates data missing or illegible when filed

TABLE 4
Severe Preeclampsia versus Mild Preeclampsia
chr	start	end	strand	pvalue	qvalue	meth.diff
chr13	58885851	688858511	*	4.92448123739017e−08	0.00318907968454339	−72.3861566484517
chr5	2564583	2564583	*	3.35175966322353e−12	1.40425632243613e−06	−66.2912949119846
chr4	6534733	6534733	*	3.13578949268644e−06	0.0532057706716535	−59.1775700934579
chr5	2564423	2564423	*	4.42484745942106e−06	0.0686806731879121	−58.7253205920312
chr11	126741853	126741653	*	1.18515862828247e−06	0.0301337427763311	−57.8643578643579
chr1	34795565	34795565	*	4 12823404811012e−10	6.81353616416951e−05	−55.0250515767757
chr14	95198847	99196847	*	5.12113920067973e−06	0.0734006412826014	−54.9138321195145
chr2	229932965	239932965	*	7 92297576639817e−12	2.47746930670009e−06	−54.7993616051072
chr14	99188646	391386461	*	1.54239463709976e−08	0.0013126000080948	−51.7150165688041
chr5	13244533	13244533	*	1.57986340814749e−06	0.0361306254380479	−49.7589914720059
chr3	174374968	174374968	*	6.20657917700822e−16	5.63410474735697e−10	−49.6864893406376
chr5	2564401	2564401	*	5.95224783573461e−07	0.0195046630119335	−49.6863260869565
chr11	8031611	8031611	*	2.41487944068427e−11	5.48025898110474e−06	−47.8846153846154
chr5	136082613	136082613	*	2.35725340053878e−06	0.0455275240709477	−46.4914772727273
chr5	136082619	136082819	*	8.75539719393383e−06	0.0859654597513623	−46.3641778114679
chr5	13244532	13244532	*	3.35120482949683e−07	0.0137235415745884	−46.1070559610706
chr3	126502264	126502264	*	2.09149986887952e−08	0.0422028306781269	−45.4273297923708
chr2	239932966	239932966	*	4.57260799247884e−10	7.11562081433487e−05	−44.4166382284913
chr3	126542502	126542502	*	4.54232170941777e−10	7.11562081433487e−05	−43.9285714285714
chr5	2564529	25644529	*	4.31565941655994e−24	1.17526822435092e−17	−43.2240348803556
chr13	19390659	19390659	*	6.52490963323898e−09	0.000658108668854797	−43.2237045492378
chr11	8031859	8031859	*	4.86986241988084e−19	5.65837844905373e−13	−42.5545654139222
chr22_KI2	135287	135287	*	8.80504264645363e−12	2.52403128071413e−08	−42.4917734251393
chr16	85319282	85319282	*	8.40300792441353e−06	0.0993896011935138	−42.0016927834363
chr1	96224399	96224399	*	6.68037719914728e−06	0.085610672875619	−41.7702864511375
chr22_KI2	135316	135316	*	1.33407106527642e−06	0.0320086431033939	−41.5205419897109
chr7	132311422	132311422	*	1.29523988473596e−10	2.43259092079145e−05	−41.3749691971308
chr22_KI2	135289	135289	*	5.62986053025345e−09	0.00059583211509241	−40.888906123694
chr7	69645871	69645871	*	2.20166404346001e−10	3.86617598980406e−05	−40.8488063660477
chr22_KI2	135295	135295	*	2.27813051386405e−06	0.0446324446716418	−40.7610787940212
chr7	3987779	3987779	*	8.06746169190422e−07	0.0241425065523422	−40.6579217654136
chr5	107784431	107784431	*	3.53979592069215e−09	0.000426432670304054	−40.561767752809
chr16	87996495	87996495	*	5.93550806535499e−06	0.0805044089124615	−40.3094676108051
chr22_KI2	135291	135291	*	3.68320793314607e−11	8.02422501688664e−06	−40.307328605201
chr1	24599267	24595267	*	1.04025218242041e−24	5.68572518953531e−18	−39.9441212492345
chr11	5441676	5441676	*	6.89214054567867e−09	0.000849866181489538	−30.9146606811145
chr10	132375370	132375370	*	7.01631165267647e−08	0.00409210784865686	−39.8526204977618
chr10	16030596	16030596	*	6.74081377014693e−06	0.08998854597513623	−39.7912685862432
chr22_KI2	135285	136285	*	1.72564062131233e−06	0.0014027892296795	−39.7138013861285
chr22_KI2	135299	136299	*	1.23254023023791e−10	2.39750732273891e−05	−30.3233921687989
chr10	507466	507466	*	7.00810033049266e−06	0.0863554684843166	−39.2650258987955
chr3	191332815	191332815	*	6.88651252420596e−07	0.0213109850082429	−39.0554722638681
chr7	17786743	17788743	*	6.00864383835461e−05	0.0805044069124615	−38.4567630733275
chr2	3059465	3050465	*	3.07734839194474e−06	0.0531842024802153	−38.2179129298569
chr7	3987668	3987668	*	1.11040068873981e−09	0.000151929561318161	−38.2164050801398
chr18	6986228	6986228	*	2.05113183366007e−10	3.72382437433588e−05	−38.1118881118861
chr10	337807	337807	*	6.00704637134146e−05	0.0605044069124615	−38.0769230769231
chr7	3987667	3987667	*	7.4742522431422e−07	0.0227421536582436	−37.6096308400307
cnr22_KI2	135264	135284	*	4.61660991055785e−10	0.0168828992811911	−36.975649682832
chr11	8031860	8021360	*	6.94997205283417e−10	0.000128276818910468	−38.9689632401497
chr22	25357353	25357353	*	1.74524969943093e−06	0.0383209636441827	−36.7647058823529
chr13	97774924	97774924	*	5.98051361144579e−07	0.0195046830119335	−36.6323417238749
chr22_KI2	135293	135293	*	2.78209747912251e−07	0.0119312375301267	−30.4374112396024
chr11	890198	890108	*	2.464399089143e−06	0.0470492486041918	−36.2764556072715
chr5	136082560	136082560	*	2.62585999305225e−06	0.048774456437286	−38.231884057971
chr22_KI2	135232	135263	*	1 64713863102e−07	0.00793905077587731	−36.0171635515738
chr4	1027464	1027464	*	8.64024183408533e−08	0.00480193959462267	−35.8465808485608
chr13	111424321	111424321	*	2.56225580563267e−06	0.0481217439563169	−35.8437566249735
chr5	3366831	3388831	*	7.76032003656057e−06	0.00440276260380905	−35.7737653557589
chr4	8691677	8691677	*	4.48814647635642e−06	0.0686581799695674	−35.6140729548888
chr1	24595268	24695268	*	8.18773763380201e−12	2.47746930670009e−06	−36.508229163011
chr7	904312	904312	*	2.82356615985535e−06	0.0505870922243516	−36.4920100925147
chr9	95680618	95680616	*	4.98560458056026e−07	0.0175117800967409	−35.4768786127168
chr10	114489621	114489821	*	3.37242998390943e−06	0.055829522577273	−35.4368932038835
chr3	75792481	75792481	*	8.45615169039445e−06	0.0993698011935136	−35.4193341869398
chr22_KI2	135281	135281	*	2.46360164963479e−06	0.0470492486041918	−35.3036138332256
chr10	132311421	132311421	*	1.25312114276786e−19	2.27503809303867e−13	−35.2910208894879
chr15	25802206	25802206	*	4.68220148803728e−09	0.000531282769728816	−35.2173913043478
chr10	132375369	132375369	*	9.17165642026229e−07	0.0254884047550052	−34.8463356973995
chr8	23263546	23263546	*	6.45430024317848e−07	0.0203198234913697	−34.8381328816111
chr2	312992	312991	*	4.25324025626474e−06	0.0663760430653978	−34.1097480106101
chr1	58910511	56510611	*	2.1559377456398e−07	0.90954856348302861	−33.2446669796067
chr22_KI2	135218	135218	*	4.6556930481352e−06	0.0698771984183716	−33.0687713290467
chr11	5441675	5441075	*	7.83599690933602e−09	0.000762119544262917	−32.6813559322034
chrUn_JTF	692	692	*	2.38165501919187e−09	0.000301666627604682	−32.3000764144988
chr11	69641427	69841427	*	6.83625937953498e−07	0.0247250346151797	−32.2670250896057
chr18	6988227	6986227	*	9.83311674288533e−07	0.026644771977707	−31.6245954692557
chr4	147741995	147741995	*	2.7309153965415e−06	0.00209491674083664	−31.5789473684211
chr22_KI2	135214	135214	*	1.81370521147475e−06	0.0362298238733574	−31.1381194618577
chr2	240900827	240900827	*	1.00272860602628e−07	0.00546135333089328	−31.09825620389
chr7	82335540	82335540	*	4.66000575787923e−11	9.7618015693243e−06	−31.0788904579903
chr3	126540525	126540525	*	5.59460969374949e−08	0.003385668326681	−30.9444613094448
chr5	2564154	2564154	*	3.31561199169442e−10	5.64326693370866e−05	−30.327868852459
chr22_KI2	135184	135184	*	2.15115023860617e−07	0.00954658348302861	−29.8768020869656
chr13	19390621	19390621	*	6.22500411692352e−06	0.0824322876531253	−29.8389129468325
chr13	114199991	114199991	*	7.36305132975948e−06	0.0909360547916652	−29.7918886668861
chr14	104159421	104159421	*	8.97817613899101e−07	0.0252059810905183	−29.7872340425532
chr22	36810296	36610298	*	9.7925578165256e−07	0.026644771977707	−29.7376093294461
chr1	1174524	1174524	*	3.80036405270521e−08	0.00254668935196708	−296628475424125
chr17	76888520	76888529	*	2.69924920290801e−08	0.00209491674083664	−29.457437742638
chr11	82736713	82736713	*	3.833979211525490-08	0.002671458253075	−29.4373162256081
chr2	67894619	87694619	*	1.24857832641374e−06	0.0310519261803968	−29.3589743589744
chr15	25802207	25602207	*	9.0939310259263e−07	0.0254000115885122	−29.0611783956948
chr13	111424320	111424320	*	1.46313496403555e−07	0.00737866007087029	−29.012526652452
chr16	5861285	5661285	*	3.06745002204275e−06	0.0531842024802153	−28.8888888888888
chr2	240900828	240900828	*	1.93972594653999e−07	0.00880391820835207	−28.8602941176471
chr5	3386320	3366320	*	9.77734786435600e−09	0.000892455196703998	−28.7586440350114
chr2	21043838	21043838	*	7.41918042410687e−06	0.091010150510033	−28.5491071428571
chr19	536536	536536	*	1.18952799286425e−06	0.0301337427763311	−28.3422459893048
chr2	71157442	71157442	*	3.28828585961074e−06	0.0547694881592753	−28.3006257974361
chr15	25177929	25177929	*	4.5248663580141e−08	0.0689077210559564	−28.2614881206431
chr2	21043849	21043849	*	2.93556654246345e−06	0.0520882886068996	−28.1921083253993
chr10	133079171	133079171	*	1.59209375268922e−06	0.0364305254380479	−27.8610694652674
chr10	43735564	43735564	*	1.63131665990732e−06	0.0384137432602935	−27.7803379416283
chr22_KI2	135337	135337	*	1.11579755422875e−09	0.000151929561318161	−27.5953389830508
chr22_KI2	151971	151971	*	1.59124062873556e−06	0.0361305254380479	−27.3683176100629
ch311	1310405	1310405	*	5.33609782299179e−06	0.0756849325628703	−27.2635914889336
chr1	1183524	1183524	*	7.71139187586468e−06	0.0935412787544808	−27.1919879082736
chr6	107986042	107986042	*	5.89672830903689e−08	0.08049244300033657	−27.0833323333333
chr12	34803294	34603294	*	4.74602141870145e−06	0.00311560998057664	−26.742994723631
chr22	50221241	50221241	*	6.37996028043536e−36	0 0833664425943691	−26.666666666667
chr11	73669807	75668807	*	3.22001077021541e−07	0.0133854753477995	−26.5592841183311
chr5	179558764	179559764	*	2.73908703612642e−06	0.0504675755135091	−26.4172335600907
chr11	66317171	65317171	*	5.06888594491374e−06	0.0730044536391841	−25.9856188461891
chr10	129392359	129692359	*	6 78295134636805e−06	0.0861148957126395	−25.8333333333333
chr1	143660428	143660428	*	1.66382915759149e−06	0.0372803334504422	−25.8642916191001
chr13	26721718	26721718	*	1.30995278279394e−06	0.0317095433019338	−25.4052684903749
chr21	24209067	24209067	*	1.65875894177013e−06	0.0368758395260794	−25.0793650793651
chr1	37761722	37761722	*	2.72052738051602e−06	0.00209491574083884	−26.0102847739635
chr12	34596867	34595867	*	3.24407480247751e−07	0.0133854753477995	−24.5849939351932
chr2	239124912	239124912	*	1.27843694629087e−07	0.00663142533378825	−24.4055944055944
chr17	50508777	59508777	*	1.22381168237542e−06	0 0307185051929208	−24.0464461646728
chr14	99644817	99644817	*	1.77708954393715e−06	0.0383239636441827	−23.9861060852437
chr1	1014255	10142555	*	1.05779778214713e−06	0.0276674136531402	−23.7411347517731
chr16	1813014	1813014	*	3.79009066345539e−06	0.060822916883671	−23 6842105263158
chr12	131003860	131003860	*	4.92083237806295e−07	0.0174034247925778	−23.6567185250021
chr21	45924346	45924346	*	1.27148608905986e−06	0.0311141973151349	−23.5351117191062
chr22_KI2	135190	135190	*	2.45948166748495e−07	0.0106313886873837	−23.1272260173994
chr17	541517	541517	*	1.84086759655569e−07	0.0084959183392371	−22.9565363881402
chr17	6657726	6657726	*	4.23135730054393e−08	0.0662242922027417	−22.9131853266402
chr17	76125688	78125668	*	5.46217618665255e−08	0 076326089849964	−22.8026533996683
chr12	131003874	131003874	*	3.95608413352707e−07	0.015958753482099	−22 5504599887583
chr17	82602604	82602804	*	6.60019338701713e−06	0.0847827841529265	−22.548393898251
chr5	179559762	179559782	*	9.93107668605606e−07	0.0267769949116706	−22.479420684533
chr2	21043749	21043749	*	6.78820398741178e−06	0 00401750062258783	−22 4768502155934
chr13	26721718	26721718	*	4.51847364258328e−06	0.0689877210559584	−22.1915430893842
chr12	131003900	131003900	*	9.83152307985395e−09	0.000892455196703998	−21.8302743900014
chr15	23562122	23562122	*	9.53224069739073e−07	0.0263539455332046	−21.8302175020004
chr13	26721799	26721799	*	5.68840647469251e−09	0.00059583211509241	−21.7505731530436
chr11	134123845	134123845	*	7.40755475669141e−08	0.091010150510033	−21.6851829489192
chr1	1174573	1174573	*	8.670S3493169424e−07	0.0247256346151797	−21.6071022080583
chr12	131003845	131003845	*	3.32142098906435e−08	0.0551527220901558	−21.3448439727548
chr5	179559730	179559730	*	4.4493731793013e−08	0.00302918442825989	−21.2669683257919
chr10	54197246	64197246	*	6.39350151946486e−07	0.0202454383824147	−21.1297440423654
chr6	168274189	168274189	*	7.62683233953893e−06	0.0927220565237502	−21.0458671379227
chr14	45254118	452541185	*	1.7675763034448e−06	0.0383260638441827	−21.0438883523241
chr5	179559727	17955972	*	5.18514010684151e−08	0.00323355106686028	−20.8166344476402
chr15	26670544	20670544	*	0.57081633245934e−06	0.0846049612014401	−20.7447642952388
chr5	1873843	1873848	*	5.29821095587927e−08	0.0755410041505466	−20.6499856197872
chr1 4	47626570	47626570	*	4.4855641752075e−06	0.0686581789695674	−20.5314854996384
chr11	32587024	32587024	*	4.02438809105929e−07	0.0154357729174998	−20.4469388219758
chr22_KI2	135172	135172	*	2.96363688962771e−06	0.0524072227605916	−20.3497023809524
chr4	991538	991538	*	5.33854116705892e−08	0.00326706364126484	−20.3204539990626
chr12	43224372	43224372	*	2.11953882648195e−06	0.0422859020989609	−20.1728148886673
chr8	59626782	69626762	*	5.55504783871103e−07	0.018876249143924	−20.1510152624661
chr13	26721860	26721860	*	8.17251545891727e−11	1.6485575975412e−05	−20.0897163804746
chr2	43068231	43068231	*	4.54198259339927e−08	0.0689077210558564	−20.0117508813161
chr3	128076766	128076786	*	3.20533786537605e−06	0.0537164527692233	−20
chr12	131003954	131003954	*	5.61128333821062e−10	6.48939161424928e−05	−19.9837734802737
chr18	87996503	87996503	*	1.58225402892991e−06	0.0361305254380479	−1S.8601232811759
chr10	75406332	75406332	*	5.85664856383476e−06	0.0804262434607438	−19.7413793103448
chr14	96204295	96204295	*	1.04834808342838e−06	0.0279299980797102	−19.8880875141545
chr7	1664582	1664582	*	5.95823912985958e−07	0.0805044069124615	−19.8078431372549
chr11	358649	358649	*	2.31581423988576e−06	0.0448824712900888	−19.4798628556983
chr1	143850348	143650348	*	1.3165886821617e−11	3.5853863813919e−08	−19.3658173178357
chr5	179658725	179559725	*	3.04205651851537e−07	0.0127450280704937	−19.3580053008737
chrX	119895971	119895971	*	4.88691500817437e−08	0.0714098888353116	−19.3543205910697
chr1	57300518	57300518	*	3.83573318221001e−06	0 0612647800081921	−19.3141254462009
chr7	47193273	47183273	*	6.26906520762471e−06	0.0626741246992889	−13.3043884220355
chr17	2564522	2564522	*	6.01856848104745e−07	0.019505471487784	−19.1745823779889
chr1	152300237	152309237	*	5.22317519281422e−06	0.0746666191660396	−18.146957214988
chr7	155491719	155491719	*	1.95916917726278e−06	0.0407935481402735	−19.1120689656172
chr17	4856719	4856719	*	3.48983540689618e−08	0.0570565231594122	−18.928284604811
chr1	4482715	4482715	*	1.55786243171406e−06	0.0361056947720476	−18.9009661835749
chr17	65080533	68080533	*	1.27315850007177e−06	0.0311141973151249	−18.8689217758985
chr13	26721781	26721761	*	4.42911525026881e−12	1.67058223393983e−06	−15.845090571655
chr1	143850313	143650313	*	1.14052091864076e−06	0.029439990801941	−15.679233544855
chr4	18246094	18246094	*	7.20843667867115e−06	0.0900474803277271	−18.5714285714286
chr1	3297266	3297268	*	5.58034558886507e−06	0.0781318961619251	−18.5441616766467
chr7	2734909	2734909	*	1.21597197481903e−06	0.0306610260790559	−18.534252179112
chr5	1594618	1594818	*	2.05158815175506e−06	0.0420537840569778	−18.3320220298977
chr10	131357947	131357947	*	1.93734979274014e−06	0.0407935481402735	−18.2949920339883
chr3	80770636	80770636	*	7.58258032655166e−07	0.0229435906044005	−18.2008484983811
chr16	830459	830459	*	4.17496376388107e−06	0.0655200906056461	−18.1907862286673
chrl	161765500	161765500	*	3,30954960805627e−06	0.06243586965581149	−18.1046387096774
chr11	5226562	5226562	*	4.53915870642454e−06	0.0689077210559564	−17.9129298588829
chr13	28721798	26721798	*	4.4231700015475e−07	0.0163882721279377	−17.7473780209758
chr1	56536130	56536130	*	7.51249604815342e−06	0.0919477519692117	−17.7083333333333
chr13	26721855	26721855	*	5.47002433996104e−07	0.0185599690146348	−17.5969869691375
chr12	131003956	131002858	*	4.01623544629303e−07	0.0154357729174998	−17.5598080124402
chr14	45254151	45254151	*	5.65838055051808e−06	0.0787913846991108	−17.3858221620178
chr3	26721846	26721846	*	2.66006932982005e−06	0.0488461024469941	−17.3768519961052
chr22_KI2	130050	130050	*	5.87711225819775e−06	0.0804262434607436	−17.3670977011494
chr5	44975082	44975082	*	5.37876922683487e−06	0.0760920092566294	−17.3486088379706
chr22_KI2	130072	130072	*	5.01577473186701e−07	00175117800967409	−17.1248373776797
chr18	41866148	41866148	*	2.20483911205559e−06	0.0436677786730833	−17.0957189588363
chr12	131003958	131003958	*	1.71412988543137e−06	0.0377975495131531	−17.0848357184087
chr7	157656310	157656310	*	3 8200747790465e−08	0.0612647800061921	−17.0212785957447
chr6	144085281	144085281	*	6.10346448657239e−06	0.0814766433784329	−17.0193855245142
chr10	485209	485209	*	1.96083731196005e−06	0.0407935481402735	−17.0169234312968
chr16	1259559	1259559	*	1.047808558358e−07	0.00584628788062724	−17.0055345439589
chr12	131003930	131003930	*	1.35928521793845e−08	090117521916101005	−16.8487742757084
chr9	68617416	68617416	*	2.8099492607405e−06	0.0505348061207153	−16.7318697662388
chr5	2564519	2564519	*	1.87750879951218e−07	0.00659314004745804	−16.7184593500383
chr15	93109787	93109787	*	5.61434744676122e−06	9.0764064052923216	−16.6718218373028
chr12	131003937	131003937	*	5.30496413706434e−08	0.00326706384126484	−16.5249359290018
chr14	96204284	90204284	*	2.29045450318785e−06	0.0447130542217479	−16.3123890171037
chr12	132406949	132406949	*	7.07980111491958e−03	0.0890507871491927	−16.018132631425
chr11	71791438	71791438	*	5.82850642299545e−06	0.080136415709074	−16.0022871452862
chr13	113453322	113453322	*	8.1774906695496e−07	0.0242057812564003	−15.9806806801203
chr3	87740644	87740644	*	2.60145167832789e−06	0.0466898854103074	−15.9473564783299
chr13	26721792	26721792	*	3.9940376849237e−09	0.000472902049970336	−15.8225882773086
chr1	161765471	1717658471	*	1.6712203451711e−07	0.00798446336860869	−15.5247963707599
chr8	8889239	3889239	*	8.1012273145316e−06	0.0971878186520002	−15.4873164218959
chr5	1594626	1594626	*	3.50097645817091e−06	0.0572613821970457	−15.4499854910252
chr8	83403144	83403144	*	4.60577226629857e−06	0.0692964138182791	−15.3224839057939
chr1	1901639	1901639	*	6.53811013845974e−06	0.0845756135183432	−15.3160408277929
chr17	82255105	82555105	*	4.08389779638427e−08	0.00311107462044008	−15.2852233676976
chr1	143650108	143650108	*	4.07686705382738e−07	0.0154900489905263	−15.2726613925962
chr10	26905429	26905429	*	6.46447760981082e−06	0.0642314035294549	−15.2222222222222
chr1	143650354	143650354	*	1.09106622091776e−07	0.00582596416048903	−15.1225664563142
chr1	161765462	161765462	*	3.16195968986384e−06	0.0534831930284655	−14.6486486486486
chr5	26455521	26455521	*	3.66253721183468e−06	0.059546207832716	−14.6148129799356
chr13	26721761	26721761	*	6.35885294614058e−07	0.0202454388824147	−14.5494747238933
chr2	237374812	237374812	*	6.01583863791704e−06	0.0605044069124615	−14.5002166450022
chr11	794471	794471	*	2.6771698710741e−06	0.0490948965839937	−14.4927536231884
chr8	69626749	69626749	*	1.2439216864157e−06	0.0310398119604081	−14.4447664589673
chr14	86905800	88905800	*	4.67246155974962e−08	0.069721984946531	−14.3822314292778
chr7	151395933	151385933	*	8.57553878334847e−07	0.0205825398261493	−14.261530652089
chr3	11094200	11094200	*	8.44860879825346e−06	0.0993898011935136	−14.1490765171504
chr13	26721885	26721885	*	4.6020058423205e−06	0.0692964138182791	−14.1257827634365
chr4	4862758	4862758	*	3.18455825658399e−06	0.0535329353951193	−14.0882194728349
chr22	16423589	16423589	*	8.40540642183114e−07	0.0243650192427174	−14 0361872581612
chr17	82254249	82254249	*	2.98732669855532e−06	0.0524853239544041	−14.0114068441065
chr7	103348856	103348858	*	2.9995918694905e−07	0.0126645373026706	−13.9090178673911
chr4	52076911	52076911	*	4.71292611684663e−07	0 0169628567466589	−13.834243988657
chr4	2057497	2057497	*	4.48799459056866e−06	0.0688581799695674	−13.7688591846184
chr2	241209373	241209373	*	3.12140531562756e−06	0.0532057706716535	−13.719533733663
chr16	86504405	86504405	*	2.81135946965069e−06	0.0505348081207153	−13.6753731343284
chrX	49204268	49204268	*	1.52834466908663e−06	0.0357258240919441	−13.6101272464909
chr19	18309702	18309702	*	2.96076626534099e−06	0.0524853239544041	−13.5828267477204
chr5	141728980	141728980	*	1.84054810269114e−07	0.0084959163592371	−13.2978723404255
chr7	4313888	4313888	*	6.55300859796376e−08	0.0845456135183432	−13.2412661326213
chr21	46391285	46391285	*	3.41633507937604e−08	0.0563849747852756	−13.2048872180451
chr6	40392378	40392378	*	7.16643972135821e−06	0.0897286358870338	−13.1810341200724
chr15	43592915	43552915	*	2.36424825066912e−07	0.0103014873466267	−13.1208813434035
chr17	88060764	63080764	*	2.50668574277778e−06	0.0474056133491432	−13.0354844640559
chr7	4783206	4783206	*	6.38653554754939e−06	0.0993896011935136	−13.0193236714976
chr8	11661684	11681684	*	3.96060168080359e−06	0.0628414933614009	−12.9044834307992
chr2	160217	160217	*	8.22738121183723e−07	0.0242218194330929	−12.7600966658475
chr19	55851218	55851218	*	2.10380518109225e−06	0.0422026356781289	−12.7328565078595
chr8	97277742	97277742	*	5.96115325576567e−08	0.0805044069124615	−12.6268365938794
chr19	17300931	17300931	*	8.46083165374636e−06	0.0993896011935136	−12.5751524069792
chr17	50537161	50537161	*	2.7140603178109e−06	0.0492900321709151	−12.5628140703518
chr16	57799576	57799576	*	7.27338729961737e−06	0.0902310306896508	−12.5
chr7	103348635	103348635	*	1.38193720164451e−06	0.0330118855844703	−12.43666393133
chr3	140651528	140651528	*	7.56449190844073e−05	0.0923766575412269	−12.3973727422003
chr1	181765468	161765468	*	2.14718100668988e−06	0.0426610371368189	−12.3321511556806
chr4	991509	991509	*	8.00221764880927e−06	0.0962889791844913	−12.3015873015873
chr1	16352354	16352354	*	6.28717146735685e−06	0.098982484567516	−12.0481927710843
chr7	132898870	132898870	*	2.09872777363957e−06	0.0422028306781269	−11.9698994303397
chr1	4734937	4734937	*	1.32677290581186e−06	0.0319745929969958	−11.7240033997966
chr11	77102985	77102985	*	7.23333211567592e−06	0.0901516830695011	−11.6711856441718
chr6	48984943	48984943	*	5.44744353303574e−08	0.076839904282006	−11.6263034953112
chr2	174597724	174597724	*	5.66182894733513e−07	0.0188024492491642	−11.62411141738
chr15	93109798	93109798	*	4.10145035343818e−06	0.0645621885271625	−11.5681801457061
chr2	208227882	208227882	*	9.96282827695804e−08	0.00546135333089328	−11.5132827324478
chr2	43899527	43665527	*	1.4593044288804e−06	0.0342590081787141	−11.4503816793893
chr2	208227863	208227863	*	8.40677562297141e−07	0.0243550192427174	−11.3602805463271
chr8	141311868	141311868	*	8.46248937588812e−06	0.0993896011935138	−11.1888111888112
chr2	208227880	208227880	*	9.59222417137886e−07	0.0283858844513235	−11.1607142857143
chr18	79831604	79831604	*	1.55233074871489e−06	0.0361058947720478	−11.1084462017948
chr8	23174099	23174099	*	3.24401256614072e−06	0.0541978174521009	−11.0568147686052
chr16	87644974	87644974	*	8.46724367791053e−06	0.0993896011935136	−10.9837193304288
chr5	73909679	73909679	*	1.78711816075786e−06	0.0383209636441827	−10.9370403059723
chr16	87644983	87644983	*	1.96183037358965e−06	0.0407935481402735	−10.7316563869777
chr8	31244455	31244455	*	6.41383947352033e−06	0.0839707598158959	−10.6145884058299
chr13	20988513	20988513	*	2.10782632983772e−08	0.0422028306781269	−10.5820105820108
chr5	174727258	174727258	*	8.74800733169197e−07	0.0247276251033757	−10.5327868852459
chr13	26721874	20721874	*	6.34193931008917e−09	0.000651723051262004	−10.4363028276072
chr7	112790652	11290652	*	1.1827481226228e−07	0.00625420212415542	−10.3943407585792
chr18	79420768	79420768	*	1.43210284086288e−06	0.03386889550114558	−10.2611206615656
chrX	9680823	9890823	*	5.11450989461697e−08	0.00323355106688028	−10.1117886178862
chr7	5213420	5213420	*	1.43647114850058e−06	0.0386669550114556	−10.0653741632931
chr8	144424694	144424694	*	7.95511286855291e−06	0.0960697529881829	−10.0591715976331
chr5	1594768	1594768	*	7.71097114265812e−07	0.0232031727296786	−9.97294372294372
chr19	11674350	11674350	*	8.14925655519208e−06	0.0975491442501126	−9.76600985221675
chr8	51859968	51859968	*	1.78596448155091e−06	0.0383209636441827	−9.47368421052631
chr1	171512085	171512085	*	1.83932226412483e−06	0.0392857019489325	−9.39844449032968
chr1	222473534	222473534	*	2.2551123433035e−06	0.0445016352786368	−9.22771123206127
chr8	142317199	142317199	*	6.86345119172967e−08	0.0867325588634461	−5.59760394644116
chr1	143665074	143655074	*	4.64356953105479e−08	0.0695473390748895	−8.34165077443195
chr6	27139558	27139558	*	3.00733618704668e−06	0.0524961810386534	−7.89521778860798
chr10	117134164	117134164	*	6.39179744992982e−06	0.0833864425943691	−7.69230769230789
chr3	49588913	49686913	*	4.7651168533535e−07	0.0169626587456569	−7.1856284251497
chr7	19143956	19143956	*	4.75311863477424e−06	0 0703473440743174	−5.09573982799097
chr20	9507030	9507830	*	9.74842783691994e−07	0.026644771977707	6.60146899268504
chr4	173530158	175530158	*	7.41632016665465e−06	0.091010150510033	6.78705922629237
chr6	160766958	160766958	*	7.60809065355089e−06	0 0927511294851	10.3838824583793
chrX	102769283	102769283	*	1.07147891551537e−06	0.0280567372188611	11.6205012765257
chr3	2562330	2962330	*	5.00810922871852e−06	0.072737615982152	12.7913279132791
chr6	19143512	19143912	*	3.93570692954429e−06	0.0628777699789228	13.1667315203136
chr18	14999781	14999781	*	8.61006335702661e−07	0.0247250346151797	13 9360070730185
chrX	101078809	101078809	*	3.4779754911831e−06	0.0570565231594122	14.1369357187435
chr11	134524258	134524258	*	1.29339543137739e−06	0.0314485172760425	14.7020348837209
chr22	49495230	49495230	*	3.95472048588311e−06	0.0627969489463513	14.7742818057456
chr3	58634151	58634151	*	1.77836418642227e−06	0.0383209636441827	15.0747756729811
chr14	105072092	105072092	*	4.1535117364589e−07	0.0156014285134611	15.6081445523193
chr6	170244465	170244465	*	4.80259132750802e−06	0.0706953384987618	15.8248296319728
chrX	131082922	131082922	*	5.32128227621392e−06	0.0758718573032399	15.9548648073238
chr16	34134797	34134797	*	1.26797707427921e−06	0.0311141973151349	18.0759326833878
chr6	33908181	33908181	*	4.68612236406862e−06	0.0169626567458589	16.2448100131944
chr5	141340409	141340409	*	4.51018579728723e−07	0.0165977640969179	16.3002602793177
chrX	87517786	87517786	*	4.095419664166616-07	0.0154900469905263	16.304347628067
chr9	89282539	89282539	*	4.59889238463388e−06	0.0892964138182791	16.5402223675605
chr15	70377264	70377264	*	2.05385531809394e−06	0.0420537840569776	16.6666866668667
chr8	41287058	41287058	*	5.2933562089499e−07	0.0163631295002267	16 8895372396467
chr16	87555699	87555699	*	4.4115326422306e−07	0.0163882721279377	17.1391702185899
chr13	20191903	20191903	*	5.75225863995006e−06	0.0793155206694152	17.5
chrX	38220439	38220439	*	6.04365590629082e−06	000361722238751272	17.8014454165082
chr2	3595039	3595039	*	8.00862418793134e−08	0.0962889791844913	17.808744321558
chr4	8975138	8975138	*	3.06569405476921e−08	0.0531842024802153	18.1483614582733
chr6	26575634	26575634	*	1.71994185373697e−08	0.0014027892296795	18.5060140491895
chr15	39976519	39976519	*	5.41642540434772e−07	0.0185599600146348	16.7257398055011
chr12	10184240	10184240	*	3.04233383534199e−06	0.0529394489505411	18.8100961538462
chrX	102518010	102518010	*	7.14652520671625e−08	0.004097209722755862	18.9068393261252
chr10	130045285	130045285	*	3.60989003609393e−07	0.0143512680308773	19.0101743749812
chr14	56991860	56991860	*	1.94098410171215e−06	0.0407935481402735	19.047619047619
chr10	131002291	131002291	*	1.27393301987318e−06	0.0311141973151349	19.1568521184934
chr11	2335408	2335406	*	5.11338829067549e−06	0.0734006412826014	18.3214285714256
chr6	28323335	26322335	*	2.4705966868042Se−06	0.0470492486041918	19.3381813013715
chr10	1939365	1939365	*	4.78729152727838e−06	0.0706610975744969	19.4548228059341
chrX	9786687	9786687	*	2.8528009405353e−06	0.0509434672156153	19.5652173913043
chr9	128123005	136123005	*	6.38307648440232e−10	9.3860472993744e−05	19.640315190927
chr5	141376448	141376448	*	1.72097120390688e−07	0.00815065729982553	20.0885559219437
chr3	13378377	133783777	*	4.31398536965116e−06	0.0671316765826345	20.2777777777778
chr21	44391581	44391581	*	3.87145880576543e−07	0.0152796155173413	20.513845611147
chr2	144996436	144996436	*	3.52539420708921e−07	0.0142228861820428	20.5388137042813
chr17	16600593	16890593	*	3.10825504189033e−06	0.0532057706716535	20.5541833899352
chr3	58931710	58931710	*	5.69682803373614e−06	0.0791523862100819	20.6349206349206
chr7	928901	928901	*	4.855420099465e−06	0.0712803415859173	20.8764814741899
chr17	82444953	82444953	*	3.62680894294201e−06	0.0591416739319224	20.9060867940188
chr6	13848066	13848066	*	2.87204146383313e−07	0.0122207428565902	21.0434425387696
chr3	43524113	45524113	*	1.96409627568811e−06	0.0407935481402735	21.1669041830832
chr3	133783816	133783816	*	3.17835747878176e−06	0.0535328353951163	21.6176359032502
chr1	227569821	227559821	*	1.35670287958669e−07	0.00697101073589234	21.6293573894643
chr18	14459067	14459067	*	5.64397650463842e−07	0 0186024492491642	21.6940639269406
chr2	7671793	7671793	*	5.52319620914389e−06	0.0775310411534318	22.0228420621436
chr13	91399621	91399621	*	6.01585270177058e−06	0.0805044069124615	22.1772379667117
chr9	138175007	138175007	*	1.54560809554006e−06	0.00758054895275172	22.2551928783383
chr3	156801089	156801089	*	2.06138677178192e−06	0.0422028306781269	22.4215998658393
chr20	29409561	29409561	*	2.06836443110187e−06	0.0421922484160333	22.5413223140498
chr5	1857306	1857306	*	1.61542972410629e−06	0.0362298238733574	22.8949161782745
chr4	187077799	187077799	*	6.69844507674789e−05	0.0856409096782738	23.1278619232124
chr7	928941	928941	*	6.7623954004734e−07	0.0247276251033757	23.282707961088
chrX	64205954	64205954	*	7.19774324394541e−07	0.0220238488742712	23.304941417225
chr14	72024078	72024078	*	1.11963392176793e−06	0 029036462846846	23.4848464848485
chr20	29327605	29327605	*	2.21688948158047e−07	0.00974609984702593	23.4852499558382
chr12	131682651	131682651	*	1.06654392288074e−06	0.0280567372188611	23.507228158391
chr5	141340332	141340332	*	5.45661064546844e−07	0.0185599690146348	23.920298950251
chr17	80444289	80444289	*	1.56639682506849e−06	0.0361305254380479	24.0074557315937
chr13	26742990	26742990	*	3.76167682506849e−08	0.0607950822976882	24.4664958157725
chr7	48848154	48848154	*	3.25247853927017e−06	0.00242665581524988	24.4950978876071
chr5	140850727	140850727	*	6.16727589384563e−06	0.0821271853731615	24.6682015167031
chr1	34794600	34794600	*	1.05125449488232e−06	0.0279299960797102	24.6813933729822
chr8	142700204	142700204	*	6.20362094554192e−07	0.0199928827520275	24.7757646132225
chr18	14999751	14999751	*	3.95745105060171e−07	0.0153958753482099	25
chr18	14459028	14459028	*	3.42402841145458e−06	0.00245380S90233896	25.0255195041925
chr6	10488257	10488257	*	6.20507731296274e−06	0.0824290340108123	25.0724637881159
chr3	149098135	149098135	*	3.00012805S09595e−05	0.0524981810388534	25.2777394084101
chr18	144590001	144590001	*	1.90462950722191e−06	0.0403640053464906	25.4508687218852
chr20	29327570	25327570	*	2.97162559695699e−06	0.00224605693600296	25.4985915492958
chr11	104365610	104365610	*	2.91276278423939e−06	0.0518442457590117	25.6756756756757
chr19	312027	312027	*	1.70996790180858e−06	0.0014027892296795	25.6893615750007
chr17	38967385	38967385	*	4.71624653174529e−06	0.0699918229023937	25.8905099894848
chr3_K127	100778	100778	*	3.78783209546821e−06	0.060892915663671	25.9943769122633
chr18	34163380	34163380	*	1.18258929007038e−06	0.0301337427763311	27.0676691729323
chr20	29327582	29327582	*	1.75874955641314e−06	0.0383209636441827	27.2700962303741
chr6	44300399	44300399	*	6.23559210462214e−06	0.0824322676531253	27.2727272727273
chr2	131830102	131830102	*	5.75002517798597e−06	0.0793155206694152	27.7456239412761
chr2	3595133	3595133	*	3.49083836034092e−07	0.0141886740320074	28.1515151515151
chr7	6847529	6847529	*	1.02368761539681e−06	0.0274655488067424	28.2588530583292
chr16	33516978	33516978	*	4.64798836079939e−06	0 0695473390746895	28.4582494969819
chr12	132306911	13236911	*	2.18847746776791e−06	0 0017512694171144	28.5575902947553
chr5	180136675	180136675	*	8.37460211822385e−06	0.0993896011935138	28.5858585858588
chr1	181508966	181508966	*	2.3469478792325e−09	0 000301666627804682	28.6469200943359
chr3	195066314	195066314	*	7.15776572770304e−07	0.0220238488742712	28.8111888111888
chr2	32886786	32886786	*	8.01885771230533e−06	0.00450254066045791	28.9589540623471
chr3	31453095	31453095	*	5.71325563683518e−06	0.0791788288040146	29.3018682399213
chr6	57856838	57856838	*	6.54950554375731e−06	0 0645756135183432	29.4098143236074
chr15	55300605	55300605	*	2.0095006012122e−06	0.0414573066230963	29.4919655900016
chr11	44973084	44973084	*	4.48398099646763e−06	0.0688581799695674	29.9889462048637
chr17	44210851	44210851	*	5.05528238089629e−06	0.0730044538391841	30.0059417706476
chr20	29316291	29319291	*	2.31249539612642e−06	0.0448824712900888	30.1051401889159
chr18	84583062	84583062	*	4.89558226964777e−06	0.0714098868353116	30.1058201058201
chr18	14459005	14459005	*	1.85571362882208e−09	0.000246515352040742	30.1441127729647
chr11	67732143	67732143	*	2.71358467207263e−06	0.0492900321709151	30.2325581395349
chr15	63459143	63459143	*	6.4598735838223e−06	0.0842314035294549	30.4066402680475
chr18	14458946	14458946	*	1.31534148752621e−06	0.00115601043150803	30.6574219783925
chr17	7589975	7589975	*	3.10161402351736e−06	0.0532057706716535	30.752320468979
chr2	217151607	217151607	*	1.48555255261793e−07	0.00742298173707828	36.8823529411765
chr12	132845363	132849383	*	2.69479562996917e−06	0.0492622950656391	31.879950226155
chr16	100279068	100279068	*	7.06249339511167e−08	0.00408210784856686	32
chr20	30898792	30896792	*	4.45025009354426e−08	0.000515707479414639	32.1693121693122
chr20	29327613	29327613	*	2.02907500000052e−11	4.80493076300122e−06	32.5396825396825
chr3	153369305	153369305	*	5.48638465154488e−07	0.0185599690146346	32.632598351335
chr2	32886723	32886723	*	7.52619510213316e−12	2.47746930570009e−06	32.8019855591285
chr22	45507251	45507251	*	2.22369779641187e−13	1,1010320235066e−07	32.9369696625706
chr7	89228935	89228935	*	1.41335375242921e−07	0.0071942242110053	33.0267825619126
chr7	928993	928993	*	1.92475388130971e−11	4.76557118932831e−06	33.726233299075
chr11	397198	397198	*	7.1495193569201e−06	0.0897230414315909	34.4482255712202
chr3	195115685	195115685	*	3.37192524891591e−08	0.00244868851904246	34.6089381374228
chrX	46851858	45851858	*	2.41048204553009e−06	0.0463910642301176	34.6503495503497
chr8	55098353	55098953	*	4.97997754605695e−08	0.0031890796454339	34.7368421052632
chr6	156156	1681567	*	4.17447598831296e−13	1.89466750621155e−07	34.8264960422164
chr14	102870466	102870466	*	3.47830651579176e−09	0.000428432670304054	34.8574359530309
chr5	1959094	1959094	*	2.49918016551001e−06	0.0474050133491432	35.627381098685
chr1	4299834	4288834	*	1.03979784054783e−08	0.000928401741010009	35.6285970748192
chr3	166224276	166224276	*	4.11858264893907e−08	0.002838092970226	35.7338562501034
chr18	14459993	14458996	*	6.50993779989158e−06	0.0845756135183432	35.7354509896883
chr8	142832873	142832875	*	5.9883304409567e−06	0.0805044089124615	35.910176779742
chr6	1681586	1681566	*	1.85285395089447e−06	0.0394201336746685	35.9133126934984
chr16	54456992	54458992	*	1.49755367782113e−11	3.88400872606384e−06	38.3604254965512
chr3	112932828	172932828	*	8.12538588446888e−06	0.0241918488849359	38.4271760364662
chr6	189657426	189657426	*	1.98887519801073e−07	0.00899515250327591	36.9354560966828
chr15	98002095	96002005	*	3.68701724074763e−06	0.0597656032904584	37.5
chr16	93137709	33137709	*	7.28939903032013e−06	0 0902310306896508	37.8393865158371
chr6	169662531	169659531	*	7.79719153734424e−06	0.9943718696225847	33.0064173015483
chr11	133930630	133930860	*	4 66153749256147e−06	0.00311107462044008	38.0324634561923
chr1	42998349	42888349	*	8.56454997755985e−14	4.67556827763799e−08	38.172892793094
chr22	19587101	19587101	*	3 96768850381144e−06	0062819720042154	36.5690789473684
chr6	35611746	35511745	*	2.61256790394456e−06	0.0487305828366124	39.0104273154774
chr11	27861394	27861394	*	2.54336454548864e−06	0.0479322306796731	39.0323711581162
chr13	152170275	152170275	*	5.02270297767131e−06	0.0727556159209122	39.1206977840641
chr11	44973691	44973891	*	3.57594818000015e−07	0.0143202484337346	39.3649721330856
chr3	166224278	156224278	*	5.68886623855735e−09	0.00059583211509241	39.624746857426
chr2	134130068	134130066	*	4.90357801395954e−06	0.0714088888353116	40 2011073115128
chr7	99229534	99228934	*	5.19451644200866e−19	5.65837844905373e−13	40.771144278807
chr11	104369609	104365609	*	4.78408342782841e−07	0.0169628567456589	40.8163286306122
chr6	188742910	168742610	*	2.27787950946004e−08	0.0446324446716418	40.9059875742473
chr6	65058854	55098854	*	1.83374033615836e−07	0.0084959183892371	41.3793103448276
chr14	24474225	24474225	*	1.61642525166116e−06	0.0362298238733574	42.0780616865802
chr17	235124	235124	*	9.65271563943995e−09	0.000892455196703998	42.2969996841654
chr11	44973890	44973580	*	4.60111453557583e−12	1.67066223393963e−06	42.5012094823416
chr11	1414150	1414150	*	2.63289414882991e−06	0.0487744456437286	42.5505050505051
chr1	227632806	227632806	*	1.55884055040968e−07	0.00758054695275172	42.6892551724138
chr20	29689780	20689780	*	2.6417856941678e−06	0.048774456437280	42.7924528301887
chr19	7554442	7554442	*	6.78637687224499e−07	0.0210682039883814	44.4444444444444
chr6	148879582	148879582	*	8 35300056008424e−07	0.0243550192427174	44.6864116806899
chr3	156224224	156224224	*	2.98779548730793e−14	1.80811157991764e−08	44.9087273689789
chr14	90550043	90556043	*	1.54503168335909e−07	0.00758054695275172	45.2725786689043
chr1	16230895	16230895	*	7.11401434897332e−09	0.0007044480404355789	46.5004274722143
chr13	113647978	113647879	*	6.36498701987876e−07	0.0202454388324147	47.9166668666667
chr1	107575707	107570707	*	3.12930693512951e−06	0 0532057700616535	49.2538911894902
chr19	1010407	1010407	*	7.25046348903469e−06	0.0901558944505037	49.2941515413426
chr1	153310526	153310526	*	1.41583773179048e−06	0.0336740125742141	49.5238095238095
chr18	88226977	85226977	*	5.14488367336249e−09	0.000571868729957131	53.6044644924824
chr7	92143674	92143674	*	6.83017303311138e−06	0.086512750559202	53.7322393438062
chr16	12390340	12390340	*	8.26122365062025e−15	5.64233608541426e−09	53.9408824838257
chr2	26814238	26814238	*	5-72897625338545e−07	0.0189108020195478	54.2857142857143
chr17	74543706	74543706	*	1.15153442203967e−05	0.0295840705366211	60.9452736318408
chr3	16735280	16738280	*	7.79770257209085e−16	6.06716066816175e−10	62.4486746786861
chr1	110464247	110464247	*	1.08671595058176e−06	0.0283195680914636	64.1176470588235
indicates data missing or illegible when filed

Discussion of this Example

This Example relates to a high read depth comprehensive bisulfite sequencing analysis of methylation in cfDNA from the plasma of pregnant and non-pregnant women. This Example also shows the global impact of pregnancy on the plasma DNA methylome and provide methylated cfDNA biomarkers of complex gestational diseases such as preeclampsia and preterm birth, and non-invasive biomarkers that are able to capture both fetal and maternal influences on the molecular phenotype during early gestation.

This Example discovers that pregnancy has a profound impact on methylation signatures in plasma cfDNA. This Example observed an overall reduction in DNA methylation that was influenced significantly by the location of the CpG sites of interest. This Example also discovered that, in general, exonic and intronic sequences in cfDNA displayed significant reductions in DNA methylation levels in pregnant versus non-pregnant women, whereas CpG islands displayed relatively fewer differences. Given that the early gestational CV is known to be generally hypomethylated compared to gestational age-matched MBCs, the pregnancy-specific reductions in global DNA methylation levels in maternal plasma are highly influenced by methylated cfDNA (mcfDNA) fragments originating in the chorionic villus. Comparing the presently disclosed plasma data with similar data from a comparison of gestational age-matched CV and maternal leukocyte samples (MBC) provided further support for this finding. It was found that when CpGs were differentially methylated between CV and MBC AND between pregnant and non-pregnant plasma, the direction of change in plasma was, in every case, predicted by that in CV/MBC. In other words, if a CpG site was hypomethylated in CV compared to MBC, its methylation levels were reduced in the plasma of pregnant women compared to non-pregnant women. The opposite was also true.

High throughput genomic approaches including the ones disclosed herein have demonstrated enormous potential both with regard to unraveling complex heterogeneous phenotypes at the molecular level and generating novel hypotheses to improve understanding of complex pathophysiology including in the context of gestational disease (Hong et al., Epigenetics. 2018; 13:163-172; Strauss et al., Am J Obstet Gynecol. 2018; 218:294-314 e292; Heng et al., PLoS One. 2016; 11:e0155191). The high throughput genomic approaches have also been used with great success to develop biomarkers, including those consisting of cell-free methylated DNA signatures (Ngo, Science. 2018; 360:1133-1136; Karlas, J Transl Med. 2017; 15:106; Hardy et a., Gut. 2017; 66:1321-1328; Liggett et al., J Neurol Sci. 2010; 290:16-21). These approaches have been used in oncology, including a particularly promising application involves plasma-based detection of DNA methylation (Shen et al., Nature. 2018; 563:579-583).

Example Embodiment B of Example 1: Method for DNA-Methylation-Based Liquid Biopsy for Non-Invasive Pregnancy Phenotyping and Disease Diagnosis

Provided below is an algorithm that can be used to diagnose a subject with a pregnancy-associated disorder and/or pregnancy abnormality.

The methylomes of either maternal tissues or the placenta could be affected by certain types of pregnancy abnormality, and the changes of these methylomes can lead to changes in the methylation patterns of the DNA fragments found in maternal plasma, which are released by the maternal tissues and the placenta. An algorithm was developed to identify the changes of methylation patterns in the methylome of maternal plasma caused by the abnormal pregnancies. The main insight behind this algorithm is that the methylome of the DNA fragments in a maternal plasma sample is a mixture of a variety of component methylomes with either maternal or fetal origin, and that the proportion of these different component methylomes in the mixture varies from subject to subject, even among the population with normal pregnancies. By constructing a model of maternal plasma methylome as a linear combination of various component methylomes of fetal and maternal origins, the algorithm can accurately predict the methylation patterns of a new maternal plasma sample under the hypothesis that it is from a normal pregnancy. Consequently, the algorithm exhibits high sensitivity in detecting abnormal methylation patterns of a maternal plasma sample caused by changes of the methylomes of some fetal/maternal tissues when the sample is from an abnormal pregnancy.

Let i be any CpG site in human genome, z_i,j be the methylation level of CpG site i in a maternal plasma sample j, p_i,x,j be the proportion of a component methylome m_r,j of either fetal or maternal origin in maternal plasma sample j at site i, m_i,r,j be the methylation level of CpG i in methylome m_r,j. Our hypothesis is:

z_i,j=Σ_r=1^Rp_i,r,jm_i,r,j  (1)

where p_i,r,j, m_i,r,j>=0, m_i,r,j<=1, p_i,1,j+ . . . +p_i,N,c=1.

It is assumed that there is a set of CpG sites S such that, for any CpG site i in S, and any maternal plasma j from a normal pregnancy, we have m_i,r,j=m_i,r and p_i,r,j=p_r,j.

That is, it is assumed that in any maternal plasma sample from a normal pregnancy, the proportions of different component methylomes in the mixture are the same for all CpG sites in S. It is also assumed that, by restricting to the set of CpG sites S, maternal plasma samples from all normal pregnancies have the same set of component methylomes. We call them restricted reference component methylomes (RRCM), and label them as m₁^S, . . . m_R^S or simply m₁, . . . , m_R when there is no confusion. For any maternal plasma sample j from a normal pregnancy, its methylome restricted to set of CpG sites in S can be expressed as a weighted average of the restricted reference component methylomes. More precisely, let z_j^S be the methylome of maternal plasma sample C restricted to S, then for some mixture vector p_j=[p_j,1, . . . , p_k,R]^T, we have:

z_j^S=[m₁^S, . . . ,m_R^S]p_j  (2)

Finally, it is assumed that the set S is the union of two disjoint subsets R and T, where T is a union of K non-empty sets T_k such that T=∪_k=1^KT_k where the index k represents the k^th type of abnormal pregnancy. T_k's do not need to be disjoint. Moreover, T_k itself is the union of two disjoint sets D_k and V_k. Either D_k or V_k could be empty, but not both. It is assumed that for any maternal plasma sample, including one from an abnormal pregnancy, when restricted to CpG sites in C, its methylome can always be expressed as a weighted average of the restricted reference component methylomes. Therefore: z_j^C[m₁^C, . . . , m_l^C]p_j regardless whether j is from an abnormal pregnancy, where C refers to the set of reference CpG sites. On the other hand, for a maternal plasma sample l form an abnormal pregnancy, when restricted to CpG sites in S=C∪T, its methylome can no longer be expressed as a weighted average of the restricted reference component methylomes. Therefore, we have: w_l^S≠[m₁^S, . . . , m_R^S]p_l for any mixture vector P. More specifically, for a maternal plasma sample l from the k^th type of abnormal pregnancy, we have: 1), w_j^C=[m₁^C, . . . , m_R^C]p_l, 2), if D_K is non-empty, then w_lDk=[m_1,k^Dk, . . . , m_R,k^Dk]p_l such that [m₁^Dk, . . . , m_R^Dk]≠[m_1,k^Dk, . . . , m_R,k^Dk], and 3), if V_K is non-empty, then w_l^Vk=[m₁^Vk, . . . , m_R^Vk]q_l such that p_l≠q_l. In other words, in a maternal plasma sample from the k^th type of abnormal pregnancy, if the set D_k is not empty, the component methylomes of the sample restricted to D_k are no longer the same as the reference component methylome restricted to D_k. If the set V_k is not empty, in this maternal plasma sample, the proportion of the reference component methylomes restricted to V_k is no longer the same as the proportion of the reference component methylome restricted to R.

T is the target set of CpG sites, D_k is the differential methylation target set, V_k is the copy number variation target set, and T_k is the target set for the k^th type of abnormal pregnancy.

The main steps of the algorithm of this Example are as follows:

• • 1) Identify the sets of CpG sites R, and T₁, . . . , T_K for the list of K types of abnormal pregnancies.
  • 2) Estimate the restricted reference component methylomes m₁, . . . , m_R, or R predictor methylomes n₁, . . . , n_R that are independent linear combinations of the reference component methylomes such that n_r=[m₁, . . . , m_R]q_r for R linearly independent mixture vectors q₁, . . . , q_R.
  • 3) (Optional) If the reference component methylomes are available, estimate the proportions of these components at the reference CpG sites C for the test maternal plasma samples.
  • 4) Predict the methylation level of the test maternal plasma samples at the target set T_k of CpG sites, under the hypothesis that the sample is from a normal pregnancy.
  • 5) Compare the predicted methylation levels at D_k and V_k against the observed methylation levels, and reject the null hypothesis that a test sample is from a normal pregnancy if the observed methylation levels are significantly different form the predicted levels.

This algorithm can be implemented in a variety of ways. For example, given the methyl-seq data for a set of maternal plasma samples from normal pregnancies, the EM algorithm or the data augmentation method can be applied to estimate the component methylomes, then use the maximum likelihood method to estimate the proportion of these component methylomes in the test sample. Below we will present some simple implementations that use linear regression.

In the first simple implementation of the algorithm of this Example, it is assumed that the restricted methylome of a maternal plasma sample from normal pregnancy can be approximated by a mixture of two restricted reference methylomes, one representing the DNA fragments from the fetus, another representing the DNA fragments from maternal tissues. It is further assumed that the estimations of these two reference component methylomes are available. For example, in the implementation below, the methylome of chorionic villi samples (CVS) will be used as an approximation to the fetal methylome in the maternal plasma sample, the methylome of plasma samples from healthy non-pregnant women (NPP) as an approximation to the maternal methylome in the maternal plasma sample. The implementation of the algorithm includes the following steps:

1. Identify the reference set C, and the target sets T_l, . . . , T_K.

• • 1.1. Collect the methylation data for a set of CVS samples, a set of NPP samples, and a set of maternal plasma samples from normal pregnancies. For each type of abnormal pregnancy, collect a set of maternal plasma samples from that type of abnormal pregnancy. All these samples should have matched age, race, gestation age (if applicable) and other relevant parameters. These are the training data.
  • 1.2. Let x_i,j be the observed methylation level of CpG i site in a CVS sample j, and y_i,l the observed methylation level of CpG site l in a NPP sample l, s_x,i² the sample variance of x_i,j over all CVS samples, s_y,i² the sample variance of y_i,j over all NPP samples. Identify the CpG sites S₀, such that for any i∈S₀, we have both s_x,i²<c₀ and s_y,i²<c₀ for some constant c₀.
  • 1.3. Let x_i be the sample mean of x_i,j over all CVS samples, y_i the sample mean of y_l,j over all NPP samples. Identify the subset C₀ of S₀ such that for any i∈C₀, we have for some constant
  • 1.4. Let x^R0 be the vector of x_i for all i∈C₀, and y^C0 be the vector of y_l for all i∈C₀. Let z_j^C0 be the observed methylation levels of CpG sites in C₀ for a normal maternal plasma sample j, and w_l^C0 the observed methylation level of CpG sites in C₀ for a maternal sample l from some abnormal pregnancy. For each j regress z_j^C0 against x^C0 and y^C0, with the constraints that the intercept must be 0, and the coefficients must be non-negative and add to 1, and get the residual e_j^C0. Similarly, for each l regress w_l^C0 against x^C0 and y^C0, with the constraints that the intercept must be 0, and the coefficients must be non-negative and add to 1, and get the residual e_l^C0. Identify the subset C of C₀ such that for any l∈C, we have

$\frac{e_{i,z}}{s_{i,z}}<c_2,\frac{e_{i,w}}{s_{i,w}}<c_2,s_{i,w}^2<c_3,$

• • and s_i,z²<c₃ for some constants c₂ and c₃, where e_i,z and e_i,w are means of methylation levels of CpG site i in all the normal samples and all the abnormal pregnancy respectively, s_i,x² and s_i,w² are sample variances of methylation levels of CpG site in all the normal samples and all the abnormal pregnancy respectively. C is the reference set of CpG sites.
  • 1.5. Let T₀=S₀\C. Let x^C and x^T0 be the vectors of x_i and x_h, for all l∈C and h∈T₀ respectively, and y^C and y^T0 be the vectors of y_i and y_h for all i∈C and h∈T₀ respectively. Let z_j^C and z_j^T0 and be the observed methylation levels of CpG sites in C and T₀ respectively for a normal maternal plasma sample j, w_lk^C and w_lk^T0 the observed methylation level of CpG sites in C and T₀ respectively for a maternal sample l_k from a pregnancy with the k^th type of abnormality, w_lg^C and w_lg^T0 the observed methylation level of CpG sites in C and T₀ respectively for a maternal sample l_g from a pregnancy with the g^th type of abnormality, where g≠k. For each j, l_k, and l_g, regress z_j^C, w_lk^C, and w_lg^C respectively against x^C and y^C, with the constraints that the intercept must be 0, and the coefficients must be non-negative and add to 1. Apply the fitted models respectively to x^T0 and Y^T0 to predict z_l^T0, w_lk^T0, and w_lg^T0 respectively, and get the differences e_j^T0, e_lk^T0 and e_lg^T0 between the predicted values and observed values. Let e_i, e_i,g, and e_i,k be sample means, and s_i², s_i,k², and s_i,g² be the sample variances, of entries in e_j^T0, e_lk^T0, e_lg^T0 and for CpG site i respectively over the normal samples, the samples from the k^th type of abnormal pregnancy, and samples from the type of abnormal pregnancy. Identify the subset T_k of T₀ such that for any i∈T_k, we have

$\frac{s_i}{s_i}<c_2,\frac{s_{i_k}}{s_i}>c_{2,k},\frac{s_{i_k}-s_i}{\sqrt{s_{i_k}^2+s_i^2}}>c_k,\mathrm{and}\frac{\text{?}-\text{?}}{\sqrt{\text{?}+\text{?}}}>\text{?}$ $\text{?}\text{indicates text missing or illegible when filed}$

for some constants c₂, c_2,k, c_k, and c_k,g. T_k is the target set for the k^th type of abnormal pregnancy.

• • 2. Estimate fetal fraction of the new maternal plasma samples to be tested. Recall that x^C and y^C are mean vectors of the methylation levels of the training CVS and training NPP data for the CpG sites in the reference set C. For any new maternal plasma sample t to be tested, let z_t^C be the observed methylation levels of CpG sites in C. Regress z_t^C against x^C and y^C, with the constraints that the intercept must be 0 and the coefficients must be non-negative and add to 1. The estimated coefficient for x^C is the estimated fetal fraction for the maternal plasma sample t.
  • 3. Test if the new maternal plasma samples are from the k^th type of abnormal pregnancy. For the new maternal plasma sample t, let x^Tk and y^Tk be mean vectors of the methylation levels of the training CVS and training NPP data for the CpG sites in the target set T_k identified in step 1 of this algorithm, apply the fitted regression models obtained from the step 2 of this algorithm to x^Tk and y^Tk to predict the methylation levels of CpG sites in T_k for sample t under the hypothesis that sample t is from a normal pregnancy. Let n_k be the number of CpG sites in T_k. Define functions

$f_k\left(x_1,\dots ,x_{n_k}\right)={{\Sigma }_i\left(-1\right)}^{I\_\left(e_{i_k}-e_i\right)}{x}_i$

• • and

$f_{k,g}\left(x_1,\dots ,x_{n_k}\right)={{\Sigma }_i\left(-1\right)}^{I\_\left(e_{i_k}-e_{i_g}\right)}{x}_i,$

• • where I_(x)=I_(−∞,0)(x), that is, the indicator function for the interval (∞,0), e_i and e_ik are estimations obtained from step 1.5 of the algorithm. We will say the sample is from the k^th type of abnormal pregnancy if f_k(e_1r, . . . , e_nk,r)>c_k, and f_k,g(e_1r, . . . , e_nk,r)>c_k,g for all g≠k, where e_ir is the difference between the observed methylation level of the CpG site i∈T_k for sample t and the predicted value by the fitted model obtained from step 2, and g is any type of abnormal pregnancy that is different form the k^th type of abnormal pregnancy.

Other ways of implementing the algorithm of this Example can be developed by modifying the simple implementation presented above. Specifically, we do not need to assume that there are only two component reference methylomes that make up the maternal plasma methylomes, nor do we need to approximate them by the CVS and NPP methylomes. Instead, we can collect a set of predictor methylomes that are mixtures of component reference genomes, as long as the number of the predictor methylomes is the same as the number of the reference component methylomes, and the mixture vectors of the predictor methylomes are linearly independent. For example, they can be methylomes of maternal mononuclear cells (MBC), or even normal maternal plasma samples. The only shortcoming of using the predictor methylomes instead of the reference component methylomes is that a straightforward estimation of the fraction of fetal DNA fragments in a maternal plasma sample cannot always be provided.

In this algorithm, we choose the difference between observed methylation levels in certain target regions and the predicted methylation levels as the test statistic to determine if in a maternal plasma sample the methylome has been affect by some type of pregnancy abnormality. To illustrate the advantage of this approach, let us assume that the mixture vector p_l for the methylome of a normal maternal plasma sample i follows a Dirichlet's distribution with parameters α₁= . . . =α_R. Furthermore, for CpG site i, its methylation levels in the R reference vector p_i for component methylomes are m_i,r=(r−1)/(R−1). It can be shown that the methylation level of i in sample j then has a mean of 0.5, and a variance of

$\frac{R+1}{12\left(R-1\right)\left(R{\alpha }_1+1\right)}.$

If we have a methyl-seq library of in sample j with a coverage of N for CpG site i, the variance of the measured methylation level z_i,j is

${\sigma }_1^2=\frac{1}{4N}+\frac{N-1}{N}\frac{R+1}{12\left(R-1\right)\left(R{\alpha }_1+1\right)}.$

In other words, if we use z_i,j as a test statistic to detect abnormal pregnancy using maternal plasma sample, under the null hypothesis, the test statistic has a variance of σ₁². However, in our algorithm, we first estimate the mixture vector p_i, then predict z_i,j by Σ_rm_i,rp_r,j. Note that in a methyl-seq data, we can get millions of CpG sites covered in each library, and that the variance of the coefficients in a linear regression model is inversely proportional to sample size. Thus it is possible to get highly accurate estimation of the mixture vector p_j, even if we take into account that adjacent CpG sites tend to have correlated methylation levels. Assuming we can get an accurate estimate of Σ_rm_i,rp_r,j, the variance of the difference z_i,j−Σ_rm_i,rp_r,j between the observed methylation level and our prediction will be

$\frac{1}{4N}-\frac{1}{N}\frac{R+1}{12\left(R-1\right)\left(R{\alpha }_1+1\right)}.$

In other words, under the null hypothesis, the test static z_i,j−Σ_rm_i,rp_r,j used in our algorithm has a much smaller variance than the other candidate test statistic z_i,j. This in turns means that our test will achieve a higher power at the same level of type I error.

Example Embodiment C of Example 1: Non-Invasive Analysis of Methylated Cell-Free DNA in Preterm Birth

Preterm birth (PTB) is one of the most important problem in modern obstetrics. In 2010, more than 1 million infants born preterm (at less than 37 weeks of gestation) died worldwide, making it the second leading cause of death in children under the age of 5 years. Preterm infants who survive are at risk of chronic lung disease, deafness blindness or other visual impairment, and learning and cognitive disabilities. The 12% rate of preterm birth in the United States ranks 131st of 184 countries, behind many developing nations. The past 3 decades in the United States have seen little decline in preterm births, including the earliest deliveries, which cause the most morbidity and mortality. Identifying potential targets for preterm birth prevention is a public health priority. One reason for the lack of predictive biomarkers for PTB is the difficulty associated with analyzing tissue samples from ongoing pregnancies. Therefore, relevant fetal tissues (cord blood and placenta) have largely been characterized at birth such that many studies are confounded by comparisons between samples collected at premature birth with controls that are collected at normal term. Thus, it is challenging, if not impossible, to separate gestational age differences from true markers of disease pathology.

Spontaneous preterm birth is a physiologically heterogeneous syndrome. The cascade of events that culminate in spontaneous preterm birth has several possible underlying pathways. Four of these pathways are supported by a considerable body of clinical and experimental evidence: excessive myometrial and fetal membrane overdistention, decidual hemorrhage, precocious fetal endocrine activation, and intrauterine infection or inflammation. These pathways may be initiated weeks to months before clinically apparent preterm labor. The processes leading to preterm parturition may originate from one or more of these pathways; for example, intrauterine infection or inflammation and placental abruption often coexist in preterm births. Decidual hemorrhage and intrauterine infection share several inflammatory molecular mechanisms that contribute to parturition. The understanding of the nature of the molecular cross-talk among these pathways is in its infancy.

The etiologic heterogeneity of preterm birth adds complexity to therapeutic approaches. Although the ultimate clinical presentation of women with preterm labor may appear to be homogeneous, the antecedent contributing factors probably differ considerably from woman to woman. A key feature of pathologic parturition is the dynamic interplay of the mother and the fetus/placenta. A critical element of the present disclosure is the notion that the approach disclosed herein permits both cross-sectional and longitudinal assessment of maternal and placental “signatures” from the readily accessible maternal blood compartment.

Delineating the factors predictive of PTB, the present disclosure obtains a better understanding of the mechanisms and biologic pathways that lead to spontaneous preterm parturition. Moreover, the use of predictors of sPTB permits the identification of a group of women at the highest risk for whom an intervention may be tested and for whom intervention is most critically needed. Third, identifying women at low risk of PTB, unnecessary, costly, and sometime hazardous interventions might be avoided.

Two fundamental compartments, the maternal and the fetal, provide the sources of biomarkers. The maternal compartment may be subdivided into blood, saliva, urine, cervix, and vagina. The fetal compartment may be subdivided into placenta, cord blood and amniotic fluid. The maternal plasma compartment contains biomarkers derived from both mother and fetus and offers several distinct theoretical advantages. Obtaining maternal plasma is less invasive than obtaining amniotic fluid, placental biopsy or fetal blood. Maternal blood is drawn routinely at several points in time during prenatal care and, thus, a plasma biomarker could be incorporated into the usual provision of care. Using maternal plasma for biomarkers of PTB also facilitates central processing and analysis. After minimal processing at the point of care, blood specimens may be transported to a central facility for processing and analysis.

It is known that plasma contains fragmented “cell-free” DNA that exists outside of intact cells. In addition to containing cell-free maternal (self) DNA, plasma from pregnant mothers contains significant amounts of DNA from the developing fetus. Methods of non-invasive detection of fetal aneuploidy have been developed using maternal plasma DNA in early gestation. This was achieved via the DNA sequencing of circulating cell-free fetal DNA in maternal plasma. Recent advances in the genomic analysis of circulating cell-free DNA (cfDNA) have enabled the development of sophisticated methods for detecting fetal aneuploidy. A small number of groups have extended these methods to enable the quantification of DNA methylation levels in cfDNA from maternal plasma. This is an attractive avenue of research because DNA methylation patterns are associated with cellular phenotype, are altered in complex gestational disease states, and are cell lineage-specific. Thus, DNA methylation signatures identified in plasma potentially contain information relating to both pathobiology and the cell lineage-specific origins of the signal. However, previous analyses of methylated cell-free DNA (mcfDNA) in maternal plasma have consisted only of proof of concept studies performed at low-resolution. None have generated detailed insights into normal temporal or pathobiological changes in DNA methylation that may occur during pregnancy.

Thus, despite its biological importance and potential clinical utility, this field is in its infancy. Critical knowledge gaps include an understanding of how methylated cell-free DNA signatures are modulated during gestation in normal pregnancies, how these are influenced by both maternal and fetal cell lineages and how they are influenced in the context of disease.

Because DNA methylation is intimately linked to cellular phenotype and modified in the context of gestational diseases and environmental exposures, the characterization of methylated cell-free DNA (mcfDNA) provides an opportunity to identify altered epigenetic signatures in early gestation that are associated with gestational disease and other negative pregnancy outcomes. This is particularly relevant given that PTB is frequently associated with abnormal placental phenotypes including malperfusion and inflammation.

This Example Embodiment C relates to novel methodology addresses these gaps. The presently disclosed novel methodology assesses methylated cell-free DNA (mcfDNA) that signatures in maternal plasma at high resolution and have generated experimental and computational tools to dissect these signatures such that their maternal and fetal inputs may be distinguished and quantified. The present disclosed methodology has identified putative DNA methylation signatures in early gestation (11-13 weeks) that are associated with an eventual diagnosis of preeclampsia as disclosed in Example Embodiment A.

The present disclosure of this Example establishes a paradigm for non-invasive pregnancy monitoring in which phenotypic information relating to both the mother and the fetus is ascertained both in the context of normal development and spontaneous preterm birth (sPTB). Importantly, novel molecular and computational methods have been developed to enable these goals. This creates an opportunity in which epigenomic signatures in maternal plasma can be exploited to define biomarkers for sPTB without the need for risky and invasive tissue biopsy. The present disclosure of this Example discloses applying the present disclosed methodology in the area of sPTB. The present disclosure of this Example has the ability to distinguish the cell-lineage specific contributions to plasma DNA methylation signals. This requires DNA methylation data for key reference cell lineages, for example cytothrophoblasts and syncytiotrophoblasts. Therefore, the present disclosure of this Example relates to optimized methods to assess genome-wide DNA methylation signatures in specific cell lineages, in which dilution effects caused by heterogeneous samples are avoided. The present disclosure of this Example can be used in combination to perform high-resolution analysis of DNA methylation in the context of normal development and spontaneous preterm birth.

The present disclosure of this Example relates to novel approaches for non-invasive pregnancy monitoring and early detection (or exclusion) of sPTB. A major barrier to progress has been that methylation profiling of maternal plasma DNA is technically challenging and expensive. Previous genome-wide bisulfite sequencing approaches in pregnancy have resulted in low coverage whole-genome data from low numbers of samples. However, the presently disclosed molecular and computational methods of this Example allow accurate bisulfite sequencing of plasma cell-free DNA at high read depth. The present disclosure of this Example also relates to methods for whole-genome bisulfite DNA sequencing of ultra-low input DNA samples obtained via laser capture microdissection. The present disclosure of this Example further relates to reference data from homogenous populations of cytotrophoblasts and syncytiotrophoblasts and other placental cell types. The present disclosure of this Example determines DNA methylation differences between mother and fetus and to identify biomarkers for the non-invasive prediction of sPTB.

The present disclosure of this Example further relates to clinical liquid biopsy assays for non-invasive pregnancy monitoring and risk assessment for sPTB.

Study 1

To characterize temporal changes in cfDNA methylation signatures in maternal plasma across the gestational age span, dynamic changes in methylated cell-free DNA signatures in maternal plasma across the gestational age range in primigravids who experience normal pregnancy are profiled. Novel computational tools and genetic approaches are used to further characterize these signatures with respect to their maternal and/or fetal origins. To conduct parallel analysis of DNA methylation in maternal leukocytes, dynamic changes in DNA methylation patterns of maternal leukocytes (from the same blood samples processed above) are profiled to understand how these are modulated throughout gestation. These data illuminate the leukocyte contribution to methylated cell-free signatures identified in maternal plasma.

A high-resolution temporal analysis of the circulating cell-free DNA methylome during gestation in maternal plasma from normal outcome pregnancies is performed. Such analysis improves the understanding of the biology of cell-free nucleic acids throughout gestation and provides a framework from which to explore pathogenesis of PTB.

The temporal dynamic changes in mcfDNA in all three trimesters of pregnancy in matched primigravid women who undergo normal pregnancy and deliver at term were characterized. Genotypic differences between mother and fetus, resulting from the fetal inheritance of uniquely paternal variants not possessed by the mother are explored to identify the origin (fetal, maternal or both) of mcfDNA in maternal plasma and determine the fraction of fetal DNA fragments (Koh et al., Proc Natl Acad Sci USA. 2014; 111:7361-7366). This allows the identification of both paternally- and maternally-inherited sequence variants in circulating methylated DNA fragments.

Identification of Fetoplacental Methylation Patterns in Maternal Plasma:

mcfDNA patterns in plasma samples from n=6 pregnant women who had normal pregnancy outcomes (gestational age 11-13 weeks) were compared with those in plasma of n=12 non-pregnant controls. mcfDNA patterns that are associated with pregnancy were identified. This is important with respect to the understanding of the epigenomic changes that occur during early gestation and is a first step towards the characterization of the early gestational methylome in maternal plasma. Such pregnancy-specific signals can originate in maternal hematopoietic cells, can be fetoplacental, and/or can be derived from other (non-hematopoietic) adult organ systems. As shown in FIG. 10, hierarchical clustering of data clearly separates pregnant from non-pregnant individuals on the basis of mcfDNA profile. Previous experiments using DNA from chorionic villus (CV) samples and matched maternal leukocytes indicated that CpG methylation levels in the latter are distributed in a biphasic manner with two peaks reflecting largely hypomethylated and hypermethylated states respectively. This is not the case in the early gestational placenta which contains relatively fewer highly methylated CpG sites (>80%), more partially methylated sites (20-80%) and a slight increase in hypomethylated sites (<20%) (FIG. 13A). The CpG methylation distributions of circulating cell-free plasma DNA from non-pregnant women (FIG. 9A) largely reflected those of maternal leukocytes (FIG. 13A). Although this distribution is preserved in the plasma of pregnant women, there is a notable reduction in high methylation sites and an increase in partially methylated sites (FIG. 11A). Given that the early gestational placenta displays a significant reduction in highly methylated (>80%) CpG sites, the distribution of CpG methylation in the plasma of pregnant, compared to non-pregnant women reflects the presence in maternal plasma of significant numbers of placentally-derived circulating cell-free DNA fragments. Thus, patterns of CpG methylation in the CV genome are broadly visible via the window of maternal plasma.

These CpG methylation distributions are influenced by the location of the CpG sites of interest examining, in turn, those sites that are present in defined genomic elements, specifically; exons, introns, promoters, CpG island, CpG island shores and enhancers. To reduce potential bias created by the presence, within other elements, of CGIs, regions of interest were filtered to exclude those that overlap with CGIs (with the exception of course of CGIs themselves). As shown in FIGS. 9B-9G, methylation distributions were highly influenced by CpG site location. Those located in promoters were less frequently hypermethylated compared to all sites (FIG. 9B). CpGs in introns and exons were infrequently hypomethylated compared to all sites (FIGS. 9C and 9D).

Distributions in CGIs appeared similar to those promoters except even few HM sites exist (FIG. 9E). CpGs in CGI shores and enhancers were more frequently methylated to intermediate levels and the former display reduced numbers of HM sites (FIGS. 9F and 9G). Pregnancy resulted in a clear relative reduction in highly methylated sites (>80%) in every case (FIG. 9B-9G), reflecting the presence of maternal plasma of a significant number of relatively hypomethylated CV genome equivalents. Additionally, a small reduction was observed in hypomethylated sites in all regions except promoters and a slight increase in partially methylated sites, particularly in CGI shores and enhancers.

To examine the spatial differences between CpG methylation between cell-free plasma DNA from pregnant and non-pregnant women, CpG methylation levels were plotted relative to genomic coordinates both in a genome-wide fashion and with respect to each autosome. As shown in FIG. 17, methylation levels were broadly reduced in DNA extracted from maternal plasma relative to non-pregnant female plasma, further indicating that the hypomethylated placenta is a significant contributor to the CpG methylation signature of cell free plasma DNA in the first trimester of pregnancy.

The impact of genomic location on the present disclosure's ability to detect pregnancy-specific DNA methylation signals in maternal plasma. The methylation rates of CpG sites present in each of the same structural and regulatory genomic elements (see above) was compared between pregnant and non-pregnant control plasma DNA. As before, regions of interest were filtered to exclude those that overlapped with CGIs. As shown in FIGS. 12A-12F, CpGs within exons (FIG. 12B) and introns (FIG. 12C) display the most significant pregnancy-specific differences in methylation, followed by those in promoters (FIG. 12A). In contrast, sites within CGI shores, enhancers and CGIs (FIGS. 12D-12F) showed almost no difference in methylation level between pregnant and non-pregnant plasma. This indicates that the impact of the first trimester placental methylome on CpG methylation levels in maternal plasma is influenced significantly by genomic location.

Bisulfite DNA sequencing data is generated from serial samples of maternal plasma and paired leukocytes obtained from 6 normal term primigravid pregnancies within gestational age windows centered at each of weeks 12, 26, and 39 weeks of gestation. All samples are obtained from women with normal healthy pregnancies who went on to deliver at term. Six matched non-pregnant controls are analyzed. Individuals are matched for gestational age, race/ethnicity, fetal sex, smoking history, and BMI. DNA is extracted from maternal plasma as previously described (Chu et al., PLoS One. 2017; 12:e0171882). Bisulfite-converted DNA libraries undergo targeted capture using the SeqCap Epi CpGiant Enrichment Kit Roche, Pleasanton, Calif.) and sequenced to a read depth of −150×. This approach targets 80.4 Mb of the human genome and 5.5 million individual CpG sites. The Bismark Bisulfite Read Mapper (Krueger et al., Bioinformatics. 2011; 27:1571-1572) is used to align the libraries against the GRCh38 reference genome and determine the methylation status for each CpG site. The beta-binomial test implemented in the R packages methylSig (Park et al., Bioinformatics. 2014; 30:2414-2422) and DSS (Park et al., Bioinformatics. 2016; 32:1446-1453) are used to compare the methylation status of groups of samples. Specifically, for each pair of sample groups and each CpG site with sufficient coverage, it is tested whether the methylation rate of that CpG site is the same in cases and controls. The origin (maternal or fetal) of mcfDNA species was determined via the detection of polymorphic sequence variants that are uniquely paternal and therefore inherited by the fetus. Informative variants are single nucleotide polymorphisms (SNPs) for which the mother is a homozygote and the fetus is a heterozygote (Koh et al., Proc Natl Acad Sci USA. 2014; 111:7361-7366). Reference genotypes are obtained for mothers and fathers using oligonucleotide microarrays. The p values of the beta-binomial tests for methylation states are adjusted using the Benjamini and Hochberg method to control the false discovery rate (Benjamini et al., J Roy Stat Soc B Met. 1995; 57:289-300).

Besides characterizing the temporal dynamic changes in levels of mcfDNA, pregnancy specific circulating mcfDNA is defined and determined in many cases whether they are maternally or fetally-derived. This insight improves the understanding of the biology and epigenomic landscape of circulating cell-free nucleic acids in pregnancy.

Study 2

To identify DNA methylation signatures in placenta associated with cell lineage, gestational age and spontaneous preterm birth, cell-type-specific DNA methylation signatures are characterized in both early gestational chorionic villus biopsies (12 weeks) and placental villus samples obtained at normal delivery (39 weeks). Placental and paired plasma and leukocyte samples obtained at delivery are also investigated in a group of women whose pregnancies resulted in preterm birth with confirmed placental malperfusion and inflammation. Cell type-specific analysis of these placentas improves the understanding of DNA methylation signatures associated with defined placental pathology in sPTB. Furthermore, plasma and leukocytes from these sPTB individuals are compared with plasma and leukocytes from gestational age-matched controls whose pregnancies are progressing normally and later deliver at term.

There are limited publicly available data describing the genome-wide DNA methylation architecture of specific human cell types in reproductive systems, particularly in early gestation. A number of studies (Chu et al., PLoS One. 2011; 6:e14723; Bunce, Prenat Diagn. 2012; 32:542-549; Chu et al., Prenat Diagn. 2009; 29:1020-1030) have provided insight but these are generally restricted to analyses of heterogenous tissue biopsies containing multiple cell types. Given that previous evidence points towards trophoblasts as the primary contributors to fetal DNA fragments in maternal plasma (Alberry, Prenatal diagnosis. 2007; 27:415-418; Lo et al., Pediatr Res. 2003; 53:16-17), it is essential to definitively characterize DNA methylation in purified cytrophoblasts and syncytiotrophoblasts. This generates critical information relating to the cell-type specificity of these epigenetic profiles and their key differences compared to maternal leukocyte. It also provides fundamental insight into their epigenomic architecture both in early gestation, at term, and in the context of sPTB pathobiology with placental malperfusion and inflammation. Plasma and leukocytes are collected from gestational age-matched controls with ongoing normal pregnancies that later deliver at term. These are compared with plasma and leukocytes obtained at delivery from mothers who experience sPTB. The presently disclosed subject matter of this Example can characterize placental and non-hematopoietic maternal DNA methylation signals in plasma, which enables insights into the sPTB phenotype in a developmentally appropriate gestational age-matched context. This has not been achieved in prior studies using conventional approaches because of the need to compare sPTB samples with normal term controls (at term).

Comprehensive comparisons of DNA methylation between first trimester chorionic villus (CV) tissue and maternal leukocytes were performed previously. These experiments were first performed using a novel custom microarray assay that was developed to provide a relatively unbiased assessment of methylation across chromosomes 13, 18 and 21 (Chu et al., PLoS One. 2011; 6:e14723; Bunce, Prenat Diagn. 2012; 32:542-549). In these experiments, it was not specifically focus on regions of perceived biological significance and so the data reveal unbiased insight into the epigenetic architecture of the first trimester chorionic villus. The results confirmed that CV tissue is generally hypomethylated relative to maternal leukocytes (FIG. 19). They also revealed a richly detailed picture of how placental methylation is organized. For example, numerous partially methylated regions were identified in which multiple CpG sites were methylated at between 20 and 80% in CV tissue. This was illustrated on a genome-wide level in FIGS. 20A-20B in which the frequency of methylation was plotted at specific CpG sites in maternal leukocytes (FIG. 20A) and CV (FIG. 20B). There was a clear bimodal distribution in the former, with large numbers of CpG sites that were either completely hypermethylated or completely hypomethylated. This distribution was not evident in CV genomes, which displayed significantly fewer fully hypermethylated sites and significantly more partially methylated sites. It was further identified a considerable number of regions in which the chorionic villus genome was totally unmethylated relative to its maternal leukocyte counterpart. These were termed tissue-specific differentially methylated regions (T-DMRs). Notably, it was also found numerous sites at which this situation was reversed. To investigate whether there were significant differences in the spatial location of these T-DMRs and whether they are clustered together or dispersed randomly throughout the genome, analysis was focused on broad differentially methylated genomic regions rather than individual T-DMRs.

In order to identify broad regions of interest, a “sliding windows” approach was used (Chu et al., PLoS One. 6:e14723). It was observed that the CV and maternal leukocyte genomes shared common features but that T-DMRs tend to cluster together in distinct chromosomal locations. It was notable that regions dense in T-DMRs were also those that encode the fewest numbers of structural genes (compare the top and middle panels of FIGS. 21A-21C with the corresponding bottom panel). Thus, placental T-DMRs are more likely to be found outside coding regions of the genome.

Bisulfite Sequencing for Comprehensive Analysis of DNA Methylation

To further develop the understanding of DNA methylation in the early gestational placenta and demonstrate the proficiency of relevant methodology, a detailed analysis of first trimester CV and maternal leukocyte methylation profiles was performed using targeted bisulfite sequencing. A commercially available oligonucleotide panel (Agilent Methylseq) was used to target an 80.4 Mb region covering all human chromosomes, with emphasis on the capture of the exome, promoters, and known CpG islands. For each sample, an average sequencing depth of 29× coverage was achieved. The distributions of single CpG sites were examined that were identified as differentially methylated (DM) between CV and MBC. A multiple testing significance filter of q=<0.1 and a % methylation filter of >+/−10% was used for these analyses. FIG. 22 showed patterns of differential methylation between CV and maternal leukocytes within a representative region of chromosome 1. Only statistically significant differences in DNA methylation (identified using the software Methylsig) were included (Park et al., Bioinformatics. 2014; 30:2414-2422). Differential methylation of a sub-set of discrete CpG sites was confirmed using targeted bisulfite next generation DNA sequencing (data not shown) as previously described (Chu et al., PLoS One. 2014; 9:e107318). The distribution of DM sites was examined in the context of a variety of genomic elements and sought to determine whether these distributions change among those that are hypermethylated in CV relative to MBC. As shown in FIG. 23, the methylation differences between CV and MBC at DM sites were dependent on genomic location. Specifically, exons, introns and promoters contained fewer DM sites hypermethylated in CV than MBC. In contrast, enhancers, CGIs and CGI contained more DM sites hypermethylated in CV than MBC. These results suggest that CpGs in genomic elements associated with the control and initiation of transcription are more likely to display higher levels of CpG methylation in CV than in MBC.

Laser Capture Microdissection and Whole Genome Bisulfite DNA Sequencing

Laser capture microdissection (LCM) followed by whole genome bisulfite sequencing (WGBS) in limited tissue samples were performed as shown in FIGS. 24A-24D and FIG. 25. Specifically, 39 WGBS were generated from individual tissue samples following laser capture. These libraries were prepared from gut epithelial cells harvested by LCM from infants born prematurely who were treated surgically for necrotizing enterocolitis (NEC). FIGS. 24A-24D shows representative before and after images of LCM. FIG. 25 shows hierarchical clustering of resulting WGBS data.

Placental tissues are obtained in early gestation (week 12) via chorionic villus sampling and also, from the same women, after normal term delivery (week 39). Blood samples (plasma and leukocytes) are collected at the same time points from the same individuals. These individuals will overlap with those sampled in Study 1. Placental and paired plasma and leukocyte samples obtained at delivery are further investigated from a group of women whose pregnancies resulted in preterm birth with confirmed placental malperfusion and inflammation. Cell type-specific analysis of these placentas improves the understanding of DNA methylation signatures associated with defined placental pathology in sPTB. Furthermore, plasma and leukocytes from these sPTB individuals are compared with plasma and leukocytes from gestational age-matched normal controls whose pregnancies are ongoing.

DNA methylation analysis methods for leukocytes and plasma are disclosed in the above Study 1. Placental tissue samples are snap frozen on dry ice and stored at −80° C. Blood samples are centrifuged and buffy coat and plasma snap frozen and stored at −80° C. Laser capture is performed using protocols for microdissection, DNA extraction and WGBS as shown in FIGS. 24A-24D and FIG. 25. WGBS libraries are generated from cytotrophoblasts and syncytiotrophoblasts. These trophoblast cell types are the focus of the study because of their significant contribution to the fetoplacental component of cfDNA in maternal plasma. Individuals are matched for age, race and BMI and will be non-smokers.

Standardized performance and classification of placental pathologic examination and diagnostic classification was performed as disclosed in Catov et al., Placenta. 2015; 36:687-692. Placental malperfusion and inflammation are among the most common lesions associated with preterm birth, and standardized and validated approaches have been incorporated to classification and sub classification of placental pathologic findings into the presently disclosed perinatal research. Using this validated system, it is straightforward to define the eligible cases of preterm birth with placental malperfusion and/or inflammation.

The sizes of tissue samples can be limited when targeting gestational week 12, but they are sufficient to generate enough DNA for effective bisulfite conversion and sequencing (see FIGS. 24A-24D and FIG. 25). Samples are also pooled from multiple individuals in addition to processing biological replicate samples of multiple pools.

Study 3

To identify non-invasive biomarkers for the prediction of spontaneous preterm birth, existing banked plasma samples were used to compare methylation signatures in cell-free DNA obtained in early gestational (11-13 weeks) from women whose pregnancies later ended in sPTB birth with those who later delivered at term.

Study 3 identifies associations between an eventual outcome of sPTB and early gestational DNA methylation signatures in maternal plasma. This is achieved using a large cohort of plasma samples obtained from mothers who are enrolled in the ongoing NIPT research program. These samples, obtained between 11 and 13 weeks of gestation, have all been collected with IRB approval for use in the context of the current proposal and have been handled and stored in a manner that optimizes their use for non-invasive prenatal testing.

Bisulfite sequencing was performed following solution-phase hybridization capture of cell-free DNA obtained from maternal plasma obtained between 11-13 weeks gestation from pregnant women who later developed either preterm severe preeclampsia <32 weeks (SPE<32) (n=5) and term preeclampsia without severe symptoms (MPE) (n=5). A normal control (NC) group (n=6) was also analyzed. Individuals were matched for gestational age and fetal gender (male) and were all non-smokers. Methylation signatures present in mcfDNA was identified in early gestation, before the onset of preeclampsia symptoms. Using the logistic regression test as implemented in the R package methylKit (Akalin et al., Genome Biol. 2012; 13:R87), after adjusting the p values to control the false discovery rate, n=552 significantly differentially methylated CpG sites were identified that distinguish SPE<32 plasma from NC and n=549 that distinguish MPE plasma from NC. The most significant of these are shown in FIGS. 15A-15B respectively. Some overlap exists between the differentially methylated sites that characterize each of the two preeclampsia phenotypes, indicating the potential for common mechanisms of disease pathogenesis (not shown). These results indicate the potential of DNA methylation analysis of mcfDNA for the molecular phenotyping of pregnant women in early gestation for the identification of individuals at risk of complex gestational disease.

DNA is extracted and genome-wide bisulfite sequencing carried out as disclosed in Study 1. Plasma DNA methylation is analyzed in a minimum of n=50 sPTB<34 weeks gestation with confirmed placental malperfusion and inflammation, and n=50 normal controls (NC). n=3750 plasma samples have been collected and stored, and sample acquisition continues at a rate of n=1150 per year with total >8000. The bisulfite sequencing libraries are prepared and processed as disclosed in Study 1. The beta-binomial test is used, implemented in the R packages methylSig and DSS, to identify CpG sites differentially methylated in plasma cfDNA samples collected between weeks 11-13 gestation from women who later experienced sPTB compared to normal controls. Cases and controls are matched with the following considerations: gestational age, ethnicity, fetal gender, smoking history, BMI, gravidity. Exclusion criteria will include multiple gestation, maternal autoimmune diseases, anti-phospholipid antibody syndrome, IVF Conception, maternal pre-pregnancy diabetes and maternal pre-pregnancy chronic hypertension. The differentially methylated CpG sites are intersected with those whose methylation states have changed during the pregnancy, as identified in Study 1 disclosed herein. Those differentially methylated CpG sites that are maternal or fetal specific are identified, as determined in Study 1 disclosed herein. A minimum of 50 differentially methylated sites is selected for multiplex PCR-targeted bisulfite sequencing using methods described in Chu et al., PLoS One. 2014; 9:e107318. Both the elastic net and the support vector machine algorithms are employed to predict the risk of sPTB using the plasma DNA methylation levels at those selected sites. Similarly, leave-one-out cross validation is used to evaluate the performance of the elastic net models and support vector machine models. The false positive rate and the false negative rate, as well as their 95% confidence intervals, are estimated.

Power Analysis of Study 3

Using logit transformation, the methylation levels of 0.375 and 0.625 are transformed to about −0.51 and 0.51. Assuming the within group variance of the logit transformed methylation levels is 1, then with 50 samples per group, the difference between the methylation levels of 0.375 and 0.625 for a CpG site can be detected at the significance level of 0.0004 with a power of 0.9. Assuming that for only 1% of the targeted CpG sites the difference in methylation levels are about 0.25 or more between the two groups, by setting the significance level at 0.0004, the false discovery rate can be controlled at <0.05 when detecting these differentially methylated CpG sites.

One potential challenge is that phenotypic heterogeneity may reduce the sensitivity of the methods disclosed herein in identifying differentially methylated CpG sites with the false discovery rate controlled at 0.05. To address this problem, large numbers of appropriate maternal plasma samples and associated clinical outcome data are obtained. This allows the application of stringent inclusion/exclusion criteria when selecting the study cohorts. To further mitigate this possibility, the logit transformed preeclamptic methylation data is clustered using the non-negative matrix factorization method as implemented in the Python module Nimfa, then a beta-binomial test is performed between each subgroup of preeclamptic plasma samples and the normal plasma samples to identify differentially methylated CpG sites.

REFERENCES

• 1. Chu T, Shaw P A, Yeniterzi S, Dunkel M, Rajkovic A, Hogge W A, et al. Comparative evaluation of the minimally-invasive karyotyping (mink) algorithm for non-invasive prenatal testing. PLoS One. 2017; 12:e0171882
• 2. Rabinowitz M, Valenti E, Pettersen B, Sigurjonsson S, Hill M, Zimmermann B. Noninvasive aneuploidy detection by multiplexed amplification and sequencing of polymorphic loci. Obstet Gynecol. 123 Suppl 1:167S
• 3. Jiang P, Tong Y K, Sun K, Cheng S H, Leung T Y, Chan K C, et al. Gestational age assessment by methylation and size profiling of maternal plasma DNA: A feasibility study. Clin Chem. 2017; 63:606-608
• 4. Sun K, Lun F M F, Leung T Y, Chiu R W K, Lo Y M D, Sun H. Noninvasive reconstruction of placental methylome from maternal plasma DNA: Potential for prenatal testing and monitoring. Prenat Diagn. 2018; 38:196-203
• 5. Chu T, Bunce K, Shaw P, Shridhar V, Althouse A, Hubel C, et al. Comprehensive analysis of preeclampsia-associated DNA methylation in the placenta. PLoS One. 2014; 9:e107318
• 6. Han Y, Yang Z, Ding X, Yu H, Yi Y. Variation of long-chain 3-hydroxyacyl-coa dehydrogenase DNA methylation in placenta of different preeclampsia-like mouse models. Zhonghua Fu Chan Ke Za Zhi. 2015; 50:740-746
• 7. Mayne B T, Leemaqz S Y, Smith A K, Breen J, Roberts C T, Bianco-Miotto T. Accelerated placental aging in early onset preeclampsia pregnancies identified by DNA methylation. Epigenomics. 2017; 9:279-289
• 8. van den Berg C B, Chaves I, Herzog E M, Willemsen S P, van der Horst G T J, Steegers-Theunissen R P M. Early- and late-onset preeclampsia and the DNA methylation of circadian clock and clock-controlled genes in placental and newborn tissues. Chronobiol Int. 2017; 34:921-932
• 9. Xuan J, Jing Z, Yuanfang Z, Xiaoju H, Pei L, Guiyin J, et al. Comprehensive analysis of DNA methylation and gene expression of placental tissue in preeclampsia patients. Hypertens Pregnancy. 2016; 35:129-138
• 10. Ye Y, Tang Y, Xiong Y, Feng L, Li X. Bisphenol a exposure alters placentation and causes preeclampsia-like features in pregnant mice involved in reprogramming of DNA methylation of wnt2. FASEB J. 2019; 33:2732-2742
• 11. Yeung K R, Chiu C L, Pidsley R, Makris A, Hennessy A, Lind J M. DNA methylation profiles in preeclampsia and healthy control placentas. Am J Physiol Heart Circ Physiol. 2016; 310:H1295-1303
• 12. Zhang L, Leng M, Li Y, Yuan Y, Yang B, Li Y, et al. Altered DNA methylation and transcription of wnt2 and dkk1 genes in placentas associated with early-onset preeclampsia. Clin Chim Acta. 2019; 490:154-160
• 13. Barcelona de Mendoza V, Wright M L, Agaba C, Prescott L, Desir A, Crusto C A, et al. A systematic review of DNA methylation and preterm birth in african american women. Biol Res Nurs. 2017; 19:308-317
• 14. Behnia F, Parets S E, Kechichian T, Yin H, Dutta E H, Saade G R, et al. Fetal DNA methylation of autism spectrum disorders candidate genes: Association with spontaneous preterm birth. Am J Obstet Gynecol. 2015; 212:533 e531-539
• 15. Burris H H, Rifas-Shiman S L, Baccarelli A, Tarantini L, Boeke C E, Kleinman K, et al. Associations of line-1 DNA methylation with preterm birth in a prospective cohort study. J Dev Orig Health Dis. 2012; 3:173-181
• 16. Hong X, Sherwood B, Ladd-Acosta C, Peng S, Ji H, Hao K, et al. Genome-wide DNA methylation associations with spontaneous preterm birth in us blacks: Findings in maternal and cord blood samples. Epigenetics. 2018; 13:163-172
• 17. Liu Y, Hoyo C, Murphy S, Huang Z, Overcash F, Thompson J, et al. DNA methylation at imprint regulatory regions in preterm birth and infection. Am J Obstet Gynecol. 2013; 208:395 e391-397
• 18. Menon R, Conneely K N, Smith A K. DNA methylation: An epigenetic risk factor in preterm birth. Reprod Sci. 2012; 19:6-13
• 19. Parets S E, Conneely K N, Kilaru V, Fortunato S J, Syed T A, Saade G, et al. Fetal DNA methylation associates with early spontaneous preterm birth and gestational age. PLoS One. 2013; 8:e67489
• 20. Parets S E, Conneely K N, Kilaru V, Menon R, Smith A K. DNA methylation provides insight into intergenerational risk for preterm birth in african americans. Epigenetics. 2015; 10:784-792
• 21. Vidal A C, Benjamin Neelon S E, Liu Y, Tuli A M, Fuemmeler B F, Hoyo C, et al. Maternal stress, preterm birth, and DNA methylation at imprint regulatory sequences in humans. Genet Epigenet. 2014; 6:37-44
• 22. Wang X M, Tian F Y, Fan L J, Xie C B, Niu Z Z, Chen W Q. Comparison of DNA methylation profiles associated with spontaneous preterm birth in placenta and cord blood. BMC Med Genomics. 2019; 12:1
• 23. Chu T, Handley D, Bunce K, Surti U, Hogge W A, Peters D G. Structural and regulatory characterization of the placental epigenome at its maternal interface. PLoS One. 2011; 6:e14723
• 24. Strauss J F, 3rd, Romero R, Gomez-Lopez N, Haymond-Thornburg H, Modi B P, Teves M E, et al. Spontaneous preterm birth: Advances toward the discovery of genetic predisposition. Am J Obstet Gynecol. 2018; 218:294-314 e292
• 25. Heng Y J, Pennell C E, McDonald S W, Vinturache A E, Xu J, Lee M W, et al. Maternal whole blood gene expression at 18 and 28 weeks of gestation associated with spontaneous preterm birth in asymptomatic women. PLoS One. 2016; 11:e0155191
• 26. Ngo T T M, Moufarrej M N, Rasmussen M H, Camunas-Soler J, Pan W, Okamoto J, et al. Noninvasive blood tests for fetal development predict gestational age and preterm delivery. Science. 2018; 360:1133-1136
• 27. Karlas T, Weise L, Kuhn S, Krenzien F, Mehdorn M, Petroff D, et al. Correlation of cell-free DNA plasma concentration with severity of non-alcoholic fatty liver disease. J Transl Med. 2017; 15:106
• 28. Hardy T, Zeybel M, Day C P, Dipper C, Masson S, McPherson S, et al. Plasma DNA methylation: A potential biomarker for stratification of liver fibrosis in non-alcoholic fatty liver disease. Gut. 2017; 66:1321-1328
• 29. Liggett T, Melnikov A, Tilwalli S, Yi Q, Chen H, Replogle C, et al. Methylation patterns of cell-free plasma DNA in relapsing-remitting multiple sclerosis. J Neurol Sci. 2010; 290:16-21
• 30. Ramaswamy S, Tamayo P, Rifkin R, Mukherjee S, Yeang C H, Angelo M, et al. Multiclass cancer diagnosis using tumor gene expression signatures. Proc Natl Acad Sci USA. 2001; 98:15149-15154
• 31. DeRisi J, Penland L, Brown P O, Bittner M L, Meltzer P S, Ray M, et al. Use of a cdna microarray to analyse gene expression patterns in human cancer. Nat Genet. 1996; 14:457-460
• 32. Chen B, Dias P, Jenkins J J, 3rd, Savell V H, Parham D M. Methylation alterations of the myod1 upstream region are predictive of subclassification of human rhabdomyosarcomas. Am J Pathol. 1998; 152:1071-1079
• 33. Ghosh R K, Pandey T, Dey P. Liquid biopsy: A new avenue in pathology.

Cytopathology. 2018

• 34. Shen S Y, Singhania R, Fehringer G, Chakravarthy A, Roehrl M H A, Chadwick D, et al. Sensitive tumour detection and classification using plasma cell-free DNA methylomes. Nature. 2018; 563:579-583
• 35. Koh W, Pan W, Gawad C, Fan H C, Kerchner G A, Wyss-Coray T, et al. Noninvasive in vivo monitoring of tissue-specific global gene expression in humans. Proc Natl Acad Sci USA. 2014; 111:7361-7366.
• 36. Lapaire O, Grill S, Lalevee S, Kolla V, Hosli I, Hahn S. Microarray screening for novel preeclampsia biomarker candidates. Fetal Diagn Ther. 2012; 31:147-153.
• 37. Winn V D, Gormley M, Fisher S J. The impact of preeclampsia on gene expression at the maternal-fetal interface. Pregnancy Hypertens. 2011; 1:100-108.
• 38. Kang J H, Song H, Yoon J A, Park D Y, Kim S H, Lee K J, et al. Preeclampsia leads to dysregulation of various signaling pathways in placenta. J Hypertens. 2011; 29:928-936.
• 39. Chaouat G, Rodde N, Petitbarat M, Bulla R, Rahmati M, Dubanchet S, et al. An insight into normal and pathological pregnancies using large-scale microarrays: Lessons from microarrays. J Reprod Immunol. 2011; 89:163-172.
• 40. Varkonyi T, Nagy B, Fule T, Tarca A L, Karaszi K, Schonleber J, et al. Microarray profiling reveals that placental transcriptomes of early-onset hellp syndrome and preeclampsia are similar. Placenta. 2011; 32 Suppl:S21-29.
• 41 Yuen R K, Penaherrera M S, von Dadelszen P, McFadden D E, Robinson W P. DNA methylation profiling of human placentas reveals promoter hypomethylation of multiple genes in early-onset preeclampsia. Eur J Hum Genet. 2010; 18:1006-1012.
• 42. Blair J D, Yuen R K, Lim B K, McFadden D E, von Dadelszen P, Robinson W P. Widespread DNA hypomethylation at gene enhancer regions in placentas associated with early-onset pre-eclampsia. Mol Hum Reprod. 2013; 19:697-708.
• 43. Fan H C, Blumenfeld Y J, Chitkara U, Hudgins L, Quake S R. Noninvasive diagnosis of fetal aneuploidy by shotgun sequencing DNA from maternal blood. Proc Natl Acad Sci USA. 2008; 105:16266-16271.
• 44. Lo Y M, Lun F M, Chan K C, Tsui N B, Chong K C, Lau T K, et al. Digital per for the molecular detection of fetal chromosomal aneuploidy. Proc Natl Acad Sci USA. 2007; 104:13116-13121.
• 45. Lun F M, Chiu R W, Allen Chan K C, Yeung Leung T, Kin Lau T, Dennis Lo Y M. Microfluidics digital per reveals a higher than expected fraction of fetal DNA in maternal plasma. Clin Chem. 2008; 54:1664-1672.
• 46. Chu T, Yeniterzi S, Rajkovic A, Hogge W A, Dunkel M, Shaw P, et al. High resolution non-invasive detection of a fetal microdeletion using the gcrem algorithm. Prenat Diagn. 2014; 34:469-477
• 47. Peters D, Chu T, Yatsenko S A, Hendrix N, Hogge W A, Surti U, et al. Noninvasive prenatal diagnosis of a fetal microdeletion syndrome. N Engl J Med. 2011; 365:1847-1848
• 48. Yatsenko S A, Peters D G, Saller D N, Chu T, Clemens M, Rajkovic A. Maternal cell-free DNA-based screening for fetal microdeletion and the importance of careful diagnostic follow-up. Genetics in medicine: official journal of the American College of Medical Genetics. 2015
• 49. Lau T K, Cheung S W, Lo P S, Pursley A N, Chan M K, Jiang F, et al. Non-invasive prenatal testing for fetal chromosomal abnormalities by low-coverage whole-genome sequencing of maternal plasma DNA: Review of 1982 consecutive cases in a single center. Ultrasound Obstet Gynecol. 2014; 43:254-264
• 50. Catov J M, Peng Y, Scifres C M, Parks W T. Placental pathology measures: Can they be rapidly and reliably integrated into large-scale perinatal studies? Placenta. 2015; 36:687-692
• 51. Catov J M, Scifres C M, Caritis S N, Bertolet M, Larkin J, Parks W T. Neonatal outcomes following preterm birth classified according to placental features. Am J Obstet Gynecol. 2017; 216:411 e411-411 e414
• 52. Krueger F, Andrews S R. Bismark: A flexible aligner and methylation caller for bisulfite-seq applications. Bioinformatics. 2011; 27:1571-1572
• 53. Park Y, Figueroa M E, Rozek L S, Sartor M A. Methylsig: A whole genome DNA methylation analysis pipeline. Bioinformatics. 2014; 30:2414-2422
• 54. Park Y, Wu H. Differential methylation analysis for bs-seq data under general experimental design. Bioinformatics. 2016; 32:1446-1453
• 55. Benjamini Y, Hochberg Y. Controlling the false discovery rate—a practical and powerful approach to multiple testing. J Roy Stat Soc B Met. 1995; 57:289-300
• 56. Bunce K, Chu T, Surti U, Hogge W A, Peters D G. Discovery of epigenetic biomarkers for the noninvasive diagnosis of fetal disease. Prenat Diagn. 2012; 32:542-549
• 57. Chu T, Burke B, Bunce K, Surti U, Allen Hogge W, Peters D G. A microarray-based approach for the identification of epigenetic biomarkers for the noninvasive diagnosis of fetal disease. Prenat Diagn. 2009; 29:1020-1030
• 58. Alberry M, Maddocks D, Jones M, Abdel Hadi M, Abdel-Fattah S, Avent N, et al. Free fetal DNA in maternal plasma in anembryonic pregnancies: Confirmation that the origin is the trophoblast. Prenatal diagnosis. 2007; 27:415-418
• 59. Lo Y M. Fetal DNA in maternal plasma/serum: The first 5 years. Pediatr Res. 2003; 53:16-17
• 60. Akalin A, Kormaksson M, Li S, Garrett-Bakelman F E, Figueroa M E, Melnick A, et al. Methylkit: A comprehensive r package for the analysis of genome-wide DNA methylation profiles. Genome Biol. 2012; 13:R87

Examples of Embodiments for Example 1

A1. A method for diagnosing, prognosing, classifying and/or monitoring a pregnancy-associated disorder in a pregnant subject comprising:

(a) obtaining a biological sample from the subject;

(b) determining the methylation status and/or level of one or more genomic loci in the biological sample;

(c) comparing the methylation status and/or level of the one or more genomic loci to a reference; and

(d) diagnosing a pregnancy-associated disorder in the subject, wherein the difference in the methylation status and/or level of the one or more genomic loci in the biological sample compared to the reference indicates the presence of the pregnancy-associated disorder in the subject.

A2. The method of embodiment A1, wherein an increase in the level of methylation of the one or more genomic loci in the biological sample indicates the presence of the pregnancy-associated disorder in the subject or a decrease in the level of methylation of the one or more genomic loci in the biological sample indicates the presence of the pregnancy-associated disorder in the subject.
A3. The method of embodiment A1, wherein a decrease in the level of methylation of at least one of the one or more genomic loci in the biological sample and an increase in the level of methylation of at least one of the one or more genomic loci in the biological sample indicates the presence of the pregnancy-associated disorder in the subject.
A4. The method of any one of embodiments A1-A3, wherein the biological sample is selected from the group consisting of a blood sample, stool sample, saliva sample and urine sample.
A5. The method of any one of embodiments A1-A4, wherein the biological sample is obtained from the pregnant subject between week 10 and week 13 of gestation.
A6. The method of any one of embodiments A1-A5, wherein the one or more genomic loci comprise one or more CpG sites.
A7. The method of any one of embodiments A1-A6, wherein the pregnancy-associated disorder is selected from the group consisting of preeclampsia, preterm labor, preterm birth, hyperemesis gravidarum, ectopic pregnancy and intrauterine growth retardation.
A8. The method of any one of embodiments A1-A7, wherein the pregnancy-associated disorder is preeclampsia.
A9. A method for diagnosing, prognosing and/or monitoring preeclampsia in a pregnant subject comprising:

(a) obtaining a biological sample from the subject;

(b) determining the methylation status and/or level of one or more genomic loci in the biological sample;

(c) comparing the methylation status and/or level of the one or more genomic loci to a reference; and

(d) diagnosing preeclampsia in the subject,

wherein the difference in the methylation status and/or level of the one or more genomic loci in the biological sample compared to the reference indicates preeclampsia in the subject.

A10. The method of embodiment A9, wherein an increase in the level of methylation of at least one of the one or more genomic loci in the biological sample indicates the presence of the pregnancy-associated disorder in the subject or a decrease in the level of methylation of at least one of the one or more genomic loci in the biological sample indicates the presence of the pregnancy-associated disorder in the subject.
A11. The method of embodiment A9, wherein a decrease in the level of methylation of at least one of the one or more genomic loci in the biological sample and an increase in the level of methylation of at least one of the one or more genomic loci in the biological sample.
A12. The method of any one of embodiments A9-A11, wherein the biological sample is a blood sample, stool sample, saliva sample and urine sample.
A13. The method of any one of embodiments A9-A12, wherein the biological sample is obtained from the pregnant subject between week 10 and week 13 of gestation.
A14. The method of any one of embodiments A9-A13, wherein the one or more genomic loci comprise one or more CpG sites.
A15. A method for determining if a pregnant subject is at increased risk of having a preterm birth comprising:

(a) obtaining a biological sample from the subject;

(b) determining the methylation status and/or level of one or more genomic loci in the biological sample;

(c) comparing the methylation status and/or level of the one or more genomic loci to a reference; and

(d) determining that the subject is at an increased risk of preterm birth,

wherein the difference in the methylation status and/or level of the one or more genomic loci in the biological sample compared to the reference indicates that the subject is at an increased risk of preterm birth.

A16. The method of embodiment A15, wherein an increase in the level of methylation of the one or more genomic loci in the biological sample indicates that the subject is at an increased risk of preterm birth or a decrease in the level of methylation of the one or more genomic loci in the biological sample indicates that the subject is at an increased risk of preterm birth.
A17. The method of embodiment A15, wherein an increase in the level of methylation of at least one of the one or more genomic loci in the biological sample indicates that the subject is at an increased risk of preterm birth and a decrease in the level of methylation of at least one of the one or more genomic loci in the biological sample indicates that the subject is at an increased risk of preterm birth.
A18. The method of any one of embodiments A15-A17, wherein the biological sample is a blood sample, stool sample, saliva sample and urine sample.
A19. The method of any one of embodiments A15-A18, wherein the biological sample is obtained from the pregnant subject between week 10 and week 13 of gestation.
A20. The method of any one of embodiments A15-A20, wherein the one or more genomic loci comprise one or more CpG sites.
A21. A method of treating a pregnancy-associated disorder in a pregnant subject comprising:

(a) obtaining a biological sample from the subject;

(b) determining the methylation status and/or level of one or more genomic loci present in the biological sample;

(c) comparing the methylation status and/or level of the one or more genomic loci to a reference;

(d) diagnosing a pregnancy-associated disorder in the subject, wherein the difference in the methylation status and/or level of the one or more genomic loci in the biological sample compared to the reference indicates the presence of the pregnancy-associated disorder in the subject; and

(e) treating the subject diagnosed with the pregnancy-associated disorder.

A22. A method of treating a pregnancy-associated disorder in a pregnant subject comprising;

(a) diagnosing a pregnancy-associated disorder in the subject by utilization of the algorithm disclosed in Example Embodiment C; and

(b) treating the subject diagnosed with the pregnancy-associated disorder.

A23. The method of embodiment A21 or A22, wherein the pregnancy-associated disorder is selected from the group consisting of preeclampsia, preterm labor, preterm birth, hyperemesis gravidarum, ectopic pregnancy and intrauterine growth retardation.
A24. The method of embodiment A21, A22 or A23, wherein the pregnancy-associated disorder is preeclampsia.
A25. The method of embodiment A24, wherein treating the subject diagnosed with preeclampsia comprises one or more of the following:

(a) administration of an anti-hypertensive medication;

(b) delivery;

(c) administration of a corticosteroid;

(d) administration of an HMG-CoA reductase inhibitor; and/or

(e) administration of an anti-convulsant medication.

A26. The method of any one of embodiments A8-A14, A24 and A25, wherein the reference is the methylation status and/or level of the one or more genomic loci in a biological sample obtained from a pregnant subject that does not have preeclampsia or from a non-pregnant subject.
A27. The method of any one of embodiments A1-A7 and A21, wherein the reference is the methylation status and/or level of the one or more genomic loci in a biological sample obtained from a pregnant subject that does not have the pregnancy-associated disorder or from a non-pregnant subject.
A28. The method of any one of embodiments A1-A27, wherein the subject is human.
A29. The method of any one of embodiments A1-A28, wherein the one or more genomic loci are present within maternal nucleic acids isolated from the biological sample.
A30. The method of embodiment A29, wherein the maternal nucleic acids are obtained from leukocytes in the biological sample.
A31. The method of embodiment A29, wherein the maternal nucleic acids are cell-free nucleic acids in the biological sample.
A32. The method of embodiment A31, wherein the maternal nucleic acids are placental nucleic acids.
A33. The method of any one of embodiments A1-A28, wherein the one or more genomic loci are present within fetal nucleic acids isolated from the biological sample.
A34. The method of embodiment A33, wherein the fetal nucleic acids are cell-free nucleic acids in the biological sample.
A35. A method for diagnosing, prognosing and/or monitoring a pregnancy-associated disorder in a pregnant subject comprising:

(a) obtaining a biological sample from the subject;

(b) determining the fraction of fetal nucleic acid (fetal fraction) in the biological sample;

(c) determining the methylation status of one or more genomic loci in placental nucleic acids present in the biological sample; and

(d) diagnosing the subject with the pregnancy-associated disorder by analyzing the fetal fraction and the methylation status of the genomic loci in the placental nucleic acid.

A36. The method of embodiment A35, wherein the pregnancy-associated disorder is selected from the group consisting of preeclampsia, preterm labor, preterm birth, hyperemesis gravidarum, ectopic pregnancy and intrauterine growth retardation.
A37. The method of embodiment A35 or A36, wherein the pregnancy-associated disorder is preeclampsia.
A38. The method of any one of embodiments A35-A37, wherein the biological sample is a blood sample, stool sample, saliva sample and urine sample.
A39. The method of any one of embodiments A35-A37, wherein the methylation status is the methylation rate of the one or more genomic loci.
A40. The method of any one of embodiments A35-A39, wherein the fetal fraction is determined by:

(a) analyzing the methylation status of one or more reference genomic loci in the biological sample; and

(b) analyzing the methylation status of the one or more reference genomic loci in a reference sample of maternal blood cells;

(c) analyzing the methylation status of the one or more reference genomic loci in a reference sample of placental nucleic acids.

A41. The method of any one of embodiments A35-A32, wherein the methylation status of the one or more genomic loci is determined by:

(a) analyzing the methylation status of the one or more genomic loci in the biological sample; and

(b) analyzing the methylation status of the one or more genomic loci in a reference sample of maternal blood cells.

A42. A kit for diagnosing, prognosing and/or monitoring a pregnancy-associated disorder in a pregnant subject comprising a means for determining and/or detecting the methylation status of one or more genomic loci.
A43. The kit of embodiment A42, wherein the means comprises one or more primers and/or probes for determining and/or detecting the methylation status of the one or more genomic loci.

Abstract of this Example (Example 1)

This Example provides methods for diagnosing, prognosing, monitoring and/or treating pregnancy-associated disorders, e.g., preeclampsia, in a pregnant subject. For example, but not by way of limitation, the methods disclosed in this Example include determining the methylation status of one or more genomic loci in a biological sample of a pregnant subject. This Example further provides algorithms and kits for performing such methods.

Example 2—Methods of Assessing DNA Methylation in Cerebrospinal Fluid for Phenotyping and Diagnosis of Central Nervous System Disorders

Introduction of this Example

This Example provides methods, algorithms and kits for diagnosing, prognosing, monitoring, classifying and/or treating central nervous system (“CNS”) disorders.

Background of this Example

Brain and spinal cord biopsies have been used to diagnose CNS tumors, infections, inflammation, and other CNS disorders. This procedure involves the removal of a small piece of brain or spinal cord tissue for the diagnosis of abnormalities of the CNS. The procedure can also be used for identifying brain- or spinal-cord specific molecular phenotypes. However, brain and spinal cord biopsy is an invasive procedure that carries the risk associated with anesthesia and surgery, for example, CNS injury, seizure, death, and complications.

Liquid biopsy is a non-invasive method for diagnosis and phenotyping via the analysis of circulating cell-free nucleic acids. Liquid biopsy has been used in reproductive genetics for detecting fetal aneuploidy (Chu T et al., Bioinformatics. 2009; 25(10):1244-50; Fan H C et al., Proc Natl Acad Sci USA. 2008; 105(42):16266-71; Chiu R W et al., Proc Natl Acad Sci USA. 2008; 105(51):20458-63), fetal microdeletions and duplications (Chu T et al., Bioinformatics. 2009; 25(10):1244-50; Peters D et al., N Engl J Med. 365(19):1847-8; Chu T et al., Prenat Diagn. 2014; 34(5):469-77; Chu T et al., PLoS One. 2016; 11(6):e0153182), and single nucleotide variants (Camunas-Soler J et al., Clin Chem. 2018; 64(2):336-45) by sequencing DNA obtained from maternal plasma. Liquid biopsy has also been used for detecting and quantifying mutations in tumor-derived DNA obtained from plasma, and other fluid reservoirs including cerebrospinal fluid (“CSF”) (De Rubis G et al., Trends Pharmacol Sci. 2019; Muinelo-Romay L et al., Int J Mol Sci. 2018; 19(8); Seoane J et al., Ann Oncol. 2018; Hiemcke-Jiwa L et al., Crit Rev Oncol Hematol. 2018; 127:56-65; Hiemcke-Jiwa L S et al., Hematol Oncol. 2018; 36(2):429-35).

DNA methylation is involved in the regulation of gene expression (Ball M P et al., Nature biotechnology. 2009; 27(4):361-8) and DNA methylation patterns can be altered as a consequence of environmental exposure and are a central component of disease pathogenesis (van Vliet J et al., CMLS. 2007; 64(12):1531-8; Abdolmaleky H M et al., Human molecular genetics. 2006; 15(21):3132-45). However, very little is understood regarding the potential of DNA methylation in cerebrospinal fluid to provide non-invasive diagnostic and phenotypic information on CNS.

Therefore, there remains a need for non-invasive methods for diagnosing CNS disorders.

Summary of this Example

This Example provides methods for diagnosing, prognosing, monitoring, classifying and/or treating CNS disorders, e.g., brain and spinal cord tumors, brain and spinal cord infections, brain and spinal cord inflammations, neuropsychiatric disorders, and neurodegenerative diseases. For example, but not by way of limitation, the methods of this Example include determining the methylation status of one or more genomic loci in a cerebrospinal fluid sample of a subject. This Example further provides algorithms and kits for diagnosing, prognosing, monitoring, classifying and/or treating CNS disorders.

In one aspect, this Example provides a method for diagnosing, prognosing, classifying and/or monitoring a CNS disorder in a subject comprising: (a) obtaining a cerebrospinal fluid sample from the subject; (b) determining the methylation status and/or level of one or more genomic loci in the cerebrospinal fluid sample; (c) comparing the methylation status and/or level of the one or more genomic loci to a reference; and (d) diagnosing the CNS disorder in the subject, wherein the difference in the methylation status and/or level of the one or more genomic loci in the cerebrospinal fluid sample compared to the reference indicates the presence of the CNS disorder in the subject.

In another aspect, this Example provides a method of treating a CNS disorder in a subject comprising: (a) obtaining a cerebrospinal fluid sample from the subject; (b) determining the methylation status and/or level of one or more genomic loci present in the cerebrospinal fluid sample; (c) comparing the methylation status and/or level of the one or more genomic loci to a reference; (d) diagnosing a CNS disorder in the subject, wherein the difference in the methylation status and/or level of the one or more genomic loci in the cerebrospinal fluid sample compared to the reference indicates the presence of the CNS disorder in the subject; and (e) treating the subject diagnosed with the CNS disorder.

In certain embodiments, an increase in the level of methylation of the one or more genomic loci in the cerebrospinal fluid sample indicates the presence of the CNS disorder in the subject or a decrease in the level of methylation of the one or more genomic loci in the cerebrospinal fluid sample indicates the presence of the CNS disorder in the subject. In certain embodiments, a decrease in the level of methylation of at least one of the one or more genomic loci in the cerebrospinal fluid sample and an increase in the level of methylation of at least one of the one or more genomic loci in the cerebrospinal fluid sample indicates the presence of the CNS disorder in the subject.

In certain embodiments, the reference is the methylation status and/or level of the one or more genomic loci in a cerebrospinal fluid sample obtained from a subject that does not have the CNS disorder.

In another aspect, this Example provides a method of treating a CNS disorder in a subject comprising: (a) measuring the methylation status and/or level of one or more genomic loci present in a cerebrospinal fluid sample from the subject prior to a treatment of the CNS disorder; (b) measuring the methylation status and/or level of one or more genomic loci present in a cerebrospinal fluid sample from the subject during the treatment of the CNS disorder; and (c) continuing the treatment if the difference in the methylation status and/or level of the one or more genomic loci between the cerebrospinal fluid samples from prior to and during the treatment of the CNS disorder indicates the subject is responsive to the treatment. In certain embodiments, the method further comprises (d) administering a different treatment to the subject if the difference in the methylation status and/or level of the one or more genomic loci between the cerebrospinal fluid samples from prior to and during the treatment of the CNS disorder indicates the subject is not responsive to the treatment.

In certain embodiments, an increase in the level of methylation of the one or more genomic loci in the cerebrospinal fluid sample indicates the subject is responsive to the treatment, or a decrease in the level of methylation of the one or more genomic loci in the cerebrospinal fluid sample indicates the subject is responsive to the treatment. In certain embodiments, a decrease in the level of methylation of at least one of the one or more genomic loci in the cerebrospinal fluid sample and an increase in the level of methylation of at least one of the one or more genomic loci in the cerebrospinal fluid sample indicates the subject is responsive to the treatment. In certain embodiments, an increase in the level of methylation of the one or more genomic loci in the cerebrospinal fluid sample indicates the subject is responsive to the treatment, or a decrease in the level of methylation of the one or more genomic loci in the cerebrospinal fluid sample indicates the subject is not responsive to the treatment. In certain embodiments, a decrease in the level of methylation of at least one of the one or more genomic loci in the cerebrospinal fluid sample and an increase in the level of methylation of at least one of the one or more genomic loci in the cerebrospinal fluid sample indicates the subject is not responsive to the treatment.

This Example further provides for algorithms for diagnosing and/or monitoring a subject with a CNS disorder. In certain embodiments, the algorithm can be used to classify a CNS disorder of a subject.

In another aspect, this Example provides a kit for diagnosing, prognosing and/or monitoring a CNS disorder in a subject comprising a means for determining and/or detecting the methylation status of one or more genomic loci. In certain embodiments, the means comprises one or more primers and/or probes for determining and/or detecting the methylation status of the one or more genomic loci.

In certain embodiments, the one or more genomic loci are present within nucleic acids isolated from the cerebrospinal fluid sample. In certain embodiments, the one or more genomic loci are present within cell-free nucleic acids isolated from the cerebrospinal fluid sample.

In certain embodiments, the CNS disorder is selected from the group consisting of brain and spinal cord tumors, brain and spinal cord infections, brain and spinal cord inflammations, neuropsychiatric disorders, and neurodegenerative diseases. In certain embodiments, the subject is human.

Description of this Example

This Example provides methods for diagnosing, prognosing, monitoring, classifying and/or treating CNS disorders, e.g., brain and spinal cord tumors, brain and spinal cord infections, brain and spinal cord inflammations, neuropsychiatric disorders, and neurodegenerative diseases. For example, but not by way of limitation, the methods of this Example include determining the methylation status of one or more genomic loci in a cerebrospinal fluid sample of a subject. In certain embodiments, the methods of this Example include the use of an algorithm to diagnose, prognose, monitor, classifying and/or treat CNS disorders.

Definitions of this Example

Unless defined otherwise, all technical and scientific terms used in this Example generally have their ordinary meanings in the art, within the context of this invention and in the specific context where each term is used. The following references provide one of skill with a general definition of many of the terms used in this invention: Singleton et al., Dictionary of Microbiology and Molecular Biology (2nd ed. 1994); The Cambridge Dictionary of Science and Technology (Walker ed., 1988); The Glossary of Genetics, 5th Ed., R. Rieger et al. (eds.), Springer Verlag (1991); and Hale & Marham, The Harper Collins Dictionary of Biology (1991). Certain terms are discussed below, or elsewhere in the specification, to provide additional guidance to the practitioner in describing the compositions and methods of the invention and how to make and use them.

The terms “comprise(s),” “include(s),” “having,” “has,” “can,” “contain(s),” and variants thereof, as used herein, are intended to be open-ended transitional phrases, terms or words that do not preclude the possibility of additional acts or structures. The singular forms “a,” “an” and “the” include plural references unless the context clearly dictates otherwise. The present disclosure also contemplates other embodiments “comprising,” “consisting of” and “consisting essentially of,” the embodiments or elements presented herein, whether explicitly set forth or not.

As used herein, the use of the word “a” or “an” when used in conjunction with the term “comprising” in the claims and/or the specification may mean “one,” but it is also consistent with the meaning of “one or more,” “at least one,” and “one or more than one.” Still further, the terms “having,” “including,” “containing” and “comprising” are interchangeable and one of skill in the art is cognizant that these terms are open ended terms.

The term “about” or “approximately” means within an acceptable error range for the particular value as determined by one of ordinary skill in the art, which will depend in part on how the value is measured or determined, i.e., the limitations of the measurement system. For example, “about” can mean within 3 or more than 3 standard deviations, per the practice in the art. Alternatively, “about” can mean a range of up to 20%, preferably up to 10%, more preferably up to 5%, and more preferably still up to 1% of a given value. Alternatively, particularly with respect to biological systems or processes, the term can mean within an order of magnitude, preferably within 5-fold, and more preferably within 2-fold, of a value.

In certain embodiments, the term “biomarker” refers to a marker (e.g., DNA methylation status) that allows detection of a disease and/or disorder in an individual, including detection of the disease or the disorder in its early stages. In certain embodiments, the term “biomarker” refers to a marker (e.g., DNA methylation status) that allows the characterization of a phenotype of a disease and/or a disorder in an individual. Early stage of a disease, as used herein, refers to the time period between the onset of the disease and the time point that signs or symptoms of the disease emerge. In certain non-limiting embodiments, the presence, absence and/or level of a biomarker in a cerebrospinal fluid sample of a subject is determined by comparing to a reference control.

The terms “reference sample,” “reference control,” “control” or “reference,” as used interchangeably herein, refers to a control for a methylation status of a genomic locus that is to be detected in a cerebrospinal fluid sample of a subject. In certain embodiments, a reference sample can be a sample from a healthy individual, e.g., an individual that does not have a CNS disorder (e.g., brain and spinal cord tumors, brain and spinal cord infections, brain and spinal cord inflammations, neuropsychiatric disorders, and neurodegenerative diseases). In certain embodiments, a reference sample can be a sample from a control individual that does not have the disease or phenotype to be detected by a biomarker disclosed herein. In certain embodiments, a control or reference can be the presence, absence and/or a particular level of a methylation state of a genomic locus in a healthy individual. In certain embodiments, a reference can be a predetermined presence, absence and/or particular level of a methylation state of a genomic locus that indicates a subject does not have a CNS disorder. In certain embodiments, a reference can be the methylation status of a locus in an individual having a disease or a phenotype, e.g., an individual that has a CNS disorder (e.g., brain and spinal cord tumors, brain and spinal cord infections, brain and spinal cord inflammations, neuropsychiatric disorders, and neurodegenerative diseases), where the methylation status of the locus is known to be not associated with the disease or the phenotype.

The term “CNS disorder” or “central nervous system disorder” as used herein, refers to any condition or disease that are caused by damage or disruption to the CNS, including the brain and spinal cord, and results in alternations of the function or structure of the CNS. Non-limiting exemplary CNS disorders include brain and spinal cord tumors, brain and spinal cord infections (e.g., meningitis, brain abscesses and encephalitis), neurodegenerative diseases (e.g., Parkinson's disease, Alzheimer's disease, Huntington's disease, spinocerebellar ataxia, spinal muscular atrophy, motor neuron diseases and prion disease), CNS inflammations (e.g., CNS vasculitis, antibody-mediated inflammatory brain diseases, demyelinating conditions, rasmussen's encephalitis and neurosarcoidosis, and secondary inflammation that occurs second to another disease, e.g., meningitis, in the body), and neuropsychiatric disorders (e.g., depression, schizophrenia and autism spectrum disorders).

The term “slightly invasive or non-invasive method” refers to a method that does not involve the removal of tissues by biopsy from brain or spinal cord. In certain embodiments, slightly invasive or non-invasive methods, as disclosed herein, include obtaining cerebrospinal fluid from a subject by lumbar puncture or spinal tap, ventricular puncture, cisternal puncture or from a shunt or ventricular drain. In certain embodiments, slightly invasive or non-invasive methods, as disclosed herein, include obtaining cerebrospinal fluid from the lymphatic system, e.g., the lymphatic system around nose and pharynx, of a subject.

The term “patient” or “subject,” as used interchangeably herein, refers to any warm-blooded animal, e.g., human or non-human. Non-limiting examples of non-human subjects include non-human primates, dogs, cats, mice, rats, guinea pigs, rabbits, fowl, pigs, horses, cows, goats, sheep, etc. In certain embodiments, the subject is human.

The term “nucleic acid,” “nucleic acid molecule” or “polynucleotide” includes any compound and/or substance that comprises a polymer of nucleotides. Each nucleotide is composed of a base, specifically a purine- or pyrimidine base (i.e., cytosine (C), guanine (G), adenine (A), thymine (T) or uracil (U)), a sugar (i.e., deoxyribose or ribose) and a phosphate group. In certain embodiments, the nucleic acid molecule is described by the sequence of bases, whereby said bases represent the primary structure (linear structure) of a nucleic acid molecule. The sequence of bases is typically represented from 5′ to 3′. These terms encompass deoxyribonucleic acid (DNA) including, e.g., complementary DNA (cDNA) and genomic DNA, ribonucleic acid (RNA), in particular messenger RNA (mRNA), synthetic forms of DNA or RNA, and mixed polymers comprising two or more of these molecules. The herein described nucleic acid molecule can contain naturally occurring or non-naturally occurring nucleotides. Examples of non-naturally occurring nucleotides include modified nucleotide bases with derivatized sugars or phosphate backbone linkages or chemically modified residues.

The term “isolated” (e.g., isolated genomic DNA) refers to a biological component that has been substantially separated or purified away from other biological components in the cell of the organism in which the component naturally occurs, e.g., other chromosomal and extra-chromosomal DNA and RNA, proteins and organelles. Nucleic acids, e.g., DNA, that have been “isolated” include nucleic acids purified by standard purification methods.

The term “genomic locus” or “genomic DNA locus,” as used herein, refers to any fixed position in a genome. For example, but not by way of limitation, a genomic locus can refer to a genomic element, a chromosomal region, a gene, a region of a gene, e.g., an exon or intron, a regulatory region of a gene, e.g., a promoter or enhancer, a CpG site, a CpG island or a CpG island shore. For example, but not by way of limitation, a genomic locus can include one or more CpG sites, e.g., between about 1 to about 100 CpG sites. In certain embodiments, a genomic locus can be of any particular length, e.g., between about 1 to about 10,000 nucleotides in length.

As used interchangeably herein, “methylation state,” “methylation profile,” “methylation status” and “methylation level” refer to the presence, absence, percentage and/or quantity of methylation at a particular nucleotide, or nucleotides, within a DNA region, e.g., a genomic locus. The methylation status of a particular DNA sequence (e.g., a genomic locus) can indicate the methylation state of every nucleotide in the sequence, indicate the methylation state of any of the nucleotides (e.g., cytosines) in the sequence, can indicate the methylation state of a subset of the nucleotides (e.g., of cytosines), can indicate the percentage or fraction of methylated cytosines at any particular stretch of nucleotides within the sequence or can indicate the average rate of methylation of all the cytosines (or a subset of the cytosines) present in a nucleic acid.

As used herein, a “methylated nucleic acid molecule” refers to a nucleic acid molecule that contains one or more methylated nucleotides that is/are methylated.

As used herein, a “CpG site” or “methylation site” is a nucleotide within a nucleic acid that is susceptible to methylation either by natural occurring events in vivo or by an event instituted to chemically methylate the nucleotide in vitro.

A “CpG island,” as used herein, describes a segment of a nucleic acid, e.g., DNA sequence, that have a high frequency of CpG dinucleotide repeats. See, e.g., Illingworth and Bird, FEBS Letters 2009; 583:1713-1720. For example, but not by way of limitation, Yamada et al. (Genome Research 2004; 14:247-266) have described a set of standards for determining a CpG island: it must be at least 400 nucleotides in length, has a GC content greater than 50% and an OCF/ECF ratio greater than 0.6. Others (Takai et al., Proc. Natl. Acad. Sci. U.S.A. 2002; 99:3740-3745) have defined a CpG island less stringently as a sequence of at least 200 nucleotides in length, having a greater than 50% GC content and an OCF/ECF ratio greater than 0.6.

A “CpG island shore,” as used herein, refers to methylation hotspots that are present a short distance, e.g., less than 2 kb, from CpG islands.

The term “methylome,” as used herein, refers to the amount or pattern of methylation at different sites or regions within a population of cells. The methylome can correspond to all of the genome, a subset of the genome (e.g., repeat elements in the genome) or a portion of the subset (e.g., those areas found to be associated with a CNS disorder). A methylome from cerebrospinal fluid can be referred to a “cerebrospinal fluid methylome,” or a “cerebrospinal fluid DNA methylome.” The cerebrospinal fluid methylome is an example of a cell-free methylome since cerebrospinal fluid includes cell-free DNA (cfDNA).

As used herein, the term “increase” refers to alter positively by at least about 2%, including, but not limited to, alter positively by about 5%, by about 10%, by about 15%, by about 20%, by about 25%, by about 30%, by about 35%, by about 40%, by about 45%, by about 50%, by about 55%, by about 60%, by about 65%, by about 70%, by about 75%, by about 80%, by about 85%, by about 90%, by about 95% or by about 100%.

As used herein, the terms “reduce,” “reduction” or “decrease” refers to alter negatively by at least about 2%, including, but not limited to, alter negatively by about 5%, by about 10%, by about 15%, by about 20%, by about 25%, by about 30%, by about 35%, by about 40%, by about 45%, by about 50%, by about 55%, by about 60%, by about 65%, by about 70%, by about 75%, by about 80%, by about 85%, by about 90%, by about 95% or by about 100%.

Methods of this Example

This Example provides methods for diagnosing, monitoring, classifying and/or treating CNS disorders in a subject, e.g., brain and spinal cord tumors, brain and spinal cord infections, brain and spinal cord inflammations, neuropsychiatric disorders, and neurodegenerative diseases, by analyzing the methylation status of one or more genomic loci in a cerebrospinal fluid sample of the subject. In certain embodiments, the methods can utilize the algorithm disclosed herein. In certain embodiments, methods of this Example allow the early diagnosis or screening of a subject with a CNS disorder, e.g., the subject does not have any symptoms, or only have early symptoms of the CNS disorder, e.g., headache, vomiting, and nausea.

In certain embodiments, the cerebrospinal fluid samples obtained for use in this Example comprise cfDNA, which carries DNA methylation information from the cell of origin. cfDNA can arise from cellular apoptosis and necrosis, and can be generated from active secretory processes, with the formation of extracellular vesicles. DNA signatures are highly tissue-specific, and include in vivo information relating to the tissue source of cfDNA. In certain embodiments, this Example comprises analyzing cfDNA in cerebrospinal fluid sample, to identify genetic phenotypes that are drivers and/or consequences of CNS disorders.

The cerebrospinal fluid sample from the subject can be collected using any suitable methods known in the art. In certain embodiments, the cerebrospinal fluid is collected by a lumbar puncture or a spinal tap, where a hollow needle is inserted into the spinal canal to collect the cerebrospinal fluid. In certain embodiments, the cerebrospinal fluid is collected by a cisternal puncture, where a needle is placed below the occipital bone to collect the cerebrospinal fluid. In certain embodiments, the cerebrospinal fluid is collected by a ventricular puncture, where a needle is inserted directly into one of the brain's ventricles. In certain embodiments, a fluoroscopy is used to assist in guiding the insertion of the needle. In certain embodiments, the cerebrospinal fluid is collected from a tube that is part of a ventriculoperitoneal shunt or an external ventricular drain.

In certain embodiments, the cerebrospinal fluid sample is collected from the subject before the subject has any symptom of the CNS disorder, i.e., a non-symptomatic subject. In certain embodiments, the cerebrospinal fluid sample is collected from the non-symptomatic subject who is at high risk of the CNS disorder, e.g., multiple members of the subject's family have a history of cancer or the subject suffered a brain trauma before. In certain embodiments, the cerebrospinal fluid sample is collected from the subject who has at least one early symptom of the CNS disorder, e.g., headache, vomiting, and nausea. In certain embodiments, the cerebrospinal fluid sample is collected from the subject who has at least one symptom of the CNS disorder, e.g., persistent headache; pain in the face, back, arms, or legs; an inability to concentrate; loss of feeling; memory loss; loss of muscle strength; tremors; seizures; increased reflexes, spasticity, tics; paralysis; and slurred speech. In certain embodiments, the cerebrospinal fluid sample is collected from the subject who has previously received or is currently receiving a treatment for a CNS disorder. In certain embodiments, two or more cerebrospinal fluid samples (e.g., two or more, three or more, four or more, five or more, six or more or seven or more cerebrospinal fluid samples) can be obtained before and during the subject is receiving the treatment of the CNS disorder (e.g., serially obtained samples).

Diagnostic, Prognostic, Classification, and Monitoring Methods of this Example

This Example provides diagnostic and prognostic methods for diseases and/or disorders that are characterized by differential methylation of genomic loci. For example, but not by way of limitation, this Example provides methods for diagnosing, prognosing, classifying and/or monitoring a CNS disorder in a subject that includes analyzing the methylation status of certain genomic loci.

Non-limiting examples of CNS disorders that can be diagnosed, monitored and/or treated by the presently disclosed subject matter include brain and spinal cord tumors, brain and spinal cord infections (e.g., meningitis, brain abscesses and encephalitis), neurodegenerative diseases (e.g., Parkinson's disease, Alzheimer's disease, Huntington's disease, spinocerebellar ataxia, spinal muscular atrophy, motor neuron diseases and prion disease), CNS inflammations (e.g., CNS vasculitis, antibody-mediated inflammatory brain diseases, demyelinating conditions, Rasmussen's encephalitis and neurosarcoidosis and secondary inflammation that occurs second to another disease, e.g., meningitis, in the body), and neuropsychiatric disorders (e.g., depression, schizophrenia and autism spectrum disorders).

In certain embodiments, the analyzed genomic loci can include one or more genomic loci that exhibit differential methylation in a cerebrospinal fluid sample from a subject that has a CNS disorder compared to a reference sample. For example, but not by way of limitation, the methods of this Example include assessing the methylation status of one or more genomic loci, e.g., about 5 or more, about 10 or more, about 50 or more, about 100 or more, about 500 or more, about 1,000 or more, about 5,000 or more, about 10,000 or more, about 25,000 or more, about 50,000 or more or about 100,000 or more genomic loci in a cerebrospinal fluid sample of a subject. In certain embodiments, the genomic loci can be selected from the genes, or a region within the genes, provided in Tables 8 & 9, FIGS. 29-32, and Example D. In certain embodiments, the one or more genomic loci can be one or more promoter regions of one or more genes, one or more exons of one or more genes, one or more introns of one or more genes, one or more CpG sites, one or more CpG islands, one or more CpG island shores, one or more enhancers of one or more genes or a combination thereof. In certain embodiments, the genomic loci are present on a particular chromosome.

In certain embodiments, this Example provides for diagnosing, prognosing and/or monitoring a CNS disorder in a subject by detecting the DNA methylation profiles associated with the CNS disorder. In certain embodiments, the methods of this Example include (a) obtaining a cerebrospinal fluid sample from the subject, (b) determining the methylation status of one or more genomic loci present in the cerebrospinal fluid sample, e.g., present within cfDNA in the cerebrospinal fluid sample, (c) comparing the methylation status of the one or more genomic loci to a reference and (d) diagnosing a CNS disorder in the subject. In certain embodiments, the difference in the methylation status of the one or more genomic loci in the cerebrospinal fluid sample compared to the reference indicates the presence of the CNS disorder in the subject. In certain embodiments, the difference in the methylation status can also indicate the severity of the CNS disorder.

In certain embodiments, the methods disclosed herein for diagnosing, prognosing and/or monitoring a CNS disorder in a subject can include (a) obtaining a cerebrospinal fluid sample from the subject, (b) determining the level of methylation of one or more genomic loci present in the cerebrospinal fluid sample, (c) comparing the level of methylation of the one or more genomic loci to a reference; and (d) diagnosing a CNS disorder in the subject. In certain embodiments, the difference in the level of methylation of the one or more genomic loci in the cerebrospinal fluid sample compared to the reference indicates the presence of the CNS disorder in the subject. In certain embodiments, the difference in the methylation level can also indicate the severity of the CNS disorder.

In certain embodiments, diagnosing a CNS disorder in the subject includes characterizing a phenotype of the CNS disorder, wherein the difference in the methylation status of the one or more genomic loci in the cerebrospinal fluid sample compared to the reference indicates the phenotype of the CNS disorder. In certain embodiment, the phenotype of the CNS disorder includes the severity of the CNS disorder, prognosis of the CNS disorder, molecular expression profile of the CNS disorder, responsiveness of the CNS disorder to certain treatments, or any combinations thereof.

In certain embodiments, the methods disclosed herein for determining if a subject is at risk of developing a CNS disorder in a subject can include (a) obtaining a cerebrospinal fluid sample from the subject, (b) determining the level of methylation of one or more genomic loci present in the cerebrospinal fluid sample, (c) comparing the level of methylation of the one or more genomic loci to a reference; and (d) determining that the subject is at risk of developing a CNS disorder, wherein the difference in the level of methylation of the one or more genomic loci in the cerebrospinal fluid sample compared to the reference indicates that the subject is at risk.

In certain embodiments, diagnosing, prognosing and/or monitoring of a subject with a CNS disorder can be based on a higher or lower methylation level of the genomic locus in the cerebrospinal fluid sample of the subject relative to the methylation level in a reference sample, e.g., a cerebrospinal fluid sample from a subject that does not have the CNS disorder. In certain embodiments, a difference of greater than about 5%, greater than about 10%, greater than about 15%, greater than about 20%, greater than about 25%, greater than about 30%, greater than about 35%, greater than about 40%, greater than about 45%, greater than about 50%, greater than about 55%, greater than about 60%, greater than about 65%, greater than about 70%, greater than about 75%, greater than about 80%, greater than about 85%, greater than about 90% or greater than about 95% in the methylation (e.g., level, percentage and/or fraction) of the one or more genomic loci in a cerebrospinal fluid sample obtained from a subject compared to a control is indicative that the subject has the CNS disorder or is at risk of developing the CNS disorder. In certain embodiments, the difference can be a decrease in methylation (e.g., level, percentage and/or fraction) of the genomic loci in the cerebrospinal fluid sample of the subject. Alternatively, the difference can be an increase in methylation (e.g., level, percentage and/or fraction) of the genomic loci in the cerebrospinal fluid sample of the subject. In certain embodiments, the difference can be a decrease in the methylation of a genomic locus and an increase in the methylation of a different genomic locus in the sample obtained from the subject. In certain embodiments, a decrease in the level of methylation of one or more genomic loci in the cerebrospinal fluid sample and the increase in the level of methylation of one or more different genomic loci in the cerebrospinal fluid sample indicates the presence of the CNS disorder.

In certain embodiments, diagnosis of a subject with a CNS disorder can be based on the methylated or unmethylated state of a genomic locus, e.g., a CpG site. In certain embodiments, a genomic locus, e.g., a CpG site, in a sample from a subject diagnosed with a CNS disorder can be methylated and the genomic locus, e.g., the CpG site, in a reference sample can be unmethylated. In certain embodiments, a genomic locus in a sample from a subject diagnosed with a CNS disorder can be unmethylated and the genomic locus in a reference sample can be methylated.

Diagnostic, Prognostic, Classification and Monitoring Methods Using an Algorithm of this Example

This Example further provides diagnostic and prognostic methods for diseases and/or disorders that are characterized by differential methylation of genomic loci by using an algorithm, as disclosed in Example Embodiment E. For example, but not by way of limitation, this Example provides methods for diagnosing, prognosing, classifying and/or monitoring a CNS disorder in a subject that includes analyzing the methylation status of certain genomic loci and/or genomic fractions.

Methods of Treatment and/or Prevention of this Example

This Example further provides methods of treating a subject with a CNS disorder. Non-limiting examples of CNS disorders that can be treated by the presently disclosed methods include brain and spinal cord tumors, brain and spinal cord infections (e.g., meningitis, brain abscesses and encephalitis), neurodegenerative diseases (e.g., Parkinson's disease, Alzheimer's disease, Huntington's disease, spinocerebellar ataxia, spinal muscular atrophy, motor neuron diseases and prion disease), CNS inflammations (e.g., CNS vasculitis, antibody-mediated inflammatory brain diseases, demyelinating conditions, Rasmussen's encephalitis and neurosarcoidosis and secondary inflammation that occurs second to another disease, e.g., meningitis, in the body), and neuropsychiatric disorders (e.g., depression, schizophrenia and autism spectrum disorders). This Example also provides methods of preventing the development of a CNS disorder in a subject.

In certain embodiments, the methods disclosed herein can include diagnosing a subject with a CNS disorder as disclosed herein, followed by treatment of the subject. Any diagnosis methods of this Example can be used with the methods for treating a subject. In certain embodiments, the methods of this Example can include determining that a subject is at risk of developing a CNS disorder as disclosed herein, followed by a method for preventing the development of the CNS disorder in the subject.

In certain embodiments, the treatment methods can include (a) obtaining a cerebrospinal fluid sample from the subject, (b) determining the methylation status and/or level of one or more genomic loci present in the cerebrospinal fluid sample, (c) comparing the methylation status and/or level of the one or more genomic loci to a reference, (d) diagnosing a CNS disorder in the subject, where the difference in the methylation status and/or level of the one or more genomic loci in the cerebrospinal fluid sample compared to the reference indicates the presence of the CNS disorder in the subject and (e) treating the subject diagnosed with the CNS disorder.

In certain embodiments, the methods of treatment can include diagnosing the subject with a CNS disorder by use of the algorithm disclosed herein. For example, but not by way of limitation, the treatment method can include (a) diagnosing the subject with a CNS disorder by use of the disclosed algorithm and (b) treating the subject diagnosed with the CNS disorder.

In certain embodiments, the prophylactic methods can include (a) obtaining a cerebrospinal fluid sample from the subject, (b) determining the methylation status and/or level of one or more genomic loci present in the cerebrospinal fluid sample, (c) comparing the methylation status and/or level of the one or more genomic loci to a reference, (d) determining that the subject is at risk of developing a CNS disorder, where the difference in the methylation status and/or level of the one or more genomic loci in the cerebrospinal fluid sample compared to the reference indicates that the subject is at risk of developing a CNS disorder and (e) preventing the subject from developing the CNS disorder.

In certain embodiments, the prophylactic methods can include (a) determining that the subject is at risk of developing a CNS disorder by use of the disclosed algorithm of this Example and (b) preventing the subject from developing the CNS disorder.

Any suitable treatment methods and preventative methods known in the art can be used with the presently disclosed methods for treating and preventing CNS disorders. For example, but not by way of limitation, the subject can be treated by administration of a medication to treat the CNS disorder. Non-limiting examples of treatment methods for CNS disorders including depression, epilepsy, multiple sclerosis (MS), neurodegenerative diseases (e.g., Alzheimer's disease), neuropathic pain and schizophrenia are disclosed in DiNunzio and Williams, “CNS Disorders—Current Treatment Options and the Prospects for Advanced Therapies,” Drug Development and Industrial Pharmacy, 34:11:1141-1167 (2008).

In certain embodiments, the information provided by the methods described herein can be used by a physician in determining the most effective course of treatment (e.g., preventative or therapeutic) for the subject. A course of treatment refers to the measures taken for a patient after the prognosis or the assessment of increased risk for development of a CNS disorder is made. For example, when a subject is identified to have an increased risk of developing a CNS disorder, the physician can determine whether frequent monitoring of DNA methylation changes can be performed as a prophylactic measure. Also, when a subject is diagnosed with a CNS disorder (e.g., based on the presence of a DNA methylation pattern in a sample from a subject), it can be advantageous to follow such detection with a therapeutic treatment.

In certain embodiments, this Example further provides methods for assessing the efficacy of a therapeutic or prophylactic therapy for preventing, inhibiting or treating a CNS disorder in a subject, comprising determining the methylation status of one or more genomic loci present in a cerebrospinal fluid sample obtained from a subject prior to the therapy and determining methylation status of the one or more genomic loci present in a cerebrospinal fluid sample obtained from the subject at one or more time points during the therapeutic or prophylactic therapy, wherein the therapy is efficacious for preventing, inhibiting and/or treating a CNS disorder in a subject when there is a change in the presence and/or level of methylation of the one or more genomic loci in the second or subsequent samples, relative to the first sample.

In certain embodiments, the first sample is obtained after therapeutic treatment has begun.

In certain embodiments, the methods for monitoring the response in a subject to prophylactic or therapeutic treatment can include measuring the methylation status and/or level of one or more genomic loci in a cerebrospinal fluid sample of a subject at a first timepoint, administering a therapeutic agent, re-measuring the methylation status and/or level of the one or more genomic loci at a second timepoint, comparing the results of the first and second measurements and optionally modifying the treatment regimen based on the comparison. In certain embodiments, the first timepoint is prior to an administration of the therapeutic agent, and the second timepoint is after said administration of the therapeutic agent. In certain embodiments, the first timepoint is prior to the administration of the therapeutic agent to the subject for the first time. In certain embodiments, the dose (defined as the quantity of therapeutic agent administered at any one administration) is increased or decreased in response to the comparison. In certain embodiments, the dosing interval (defined as the time between successive administrations) can be increased or decreased in response to the comparison, including total discontinuation of treatment. In addition, the method of the present disclosure can be used to determine the efficacy of the therapeutic treatment, wherein a change in the methylation status of certain genomic loci present in a cerebrospinal fluid sample of a subject can indicate that the therapeutic treatment regimen can be altered, reduced and/or stopped.

Assays of this Example

This Example further provides assays and/or methods for determining the DNA methylation status and/or level of genomic loci that correlates with the presence, absence and/or severity of a CNS disorder.

In certain embodiments, the assay method of this Example can include comparing the methylation status and/or level of genomic loci present in a cerebrospinal fluid sample from a subject that has a CNS disorder to the methylation status and/or level of genomic loci in a cerebrospinal fluid sample from a healthy subject to determine the methylation pattern, as disclosed above, that correlates with the presence of the CNS disorder.

In certain embodiments, the assay methods of this Example can include comparing the methylation status and/or level of genomic loci in a cerebrospinal fluid sample from a subject that has a CNS disorder at an early stage to the methylation status and/or level of genomic loci in a cerebrospinal fluid sample from a subject that has the CNS disorder at a late stage to determine the methylation status and/or level that correlates with the different stages and/or severity of the CNS disorder.

Non-limiting examples of CNS disorders include brain and spinal cord tumors, brain and spinal cord infections (e.g., meningitis, brain abscesses and encephalitis), neurodegenerative diseases (e.g., Parkinson's disease, Alzheimer's disease, Huntington's disease, spinocerebellar ataxia, spinal muscular atrophy, motor neuron diseases and prion disease), CNS inflammations (e.g., CNS vasculitis, antibody-mediated inflammatory brain diseases, demyelinating conditions, Rasmussen's encephalitis and neurosarcoidosis, and secondary inflammation occurs second to another disease, e.g., meningitis, in the body), and neuropsychiatric disorders (e.g., depression, schizophrenia and autism spectrum disorders).

DNA Isolation Techniques of this Example

In certain embodiments, the methods of this Example include isolating nucleic acid from a cerebrospinal fluid sample obtained from a subject. There are several platforms that are known in the art and currently available to isolate nucleic acids from cerebrospinal fluid samples. For example, but not by way of limitation, isolation of DNA from a cerebrospinal fluid sample can be performed by extraction methods using organic solvents such as a mixture of phenol and chloroform, followed by precipitation with ethanol (see, for example, J. Sambrook et al., “Molecular Cloning: A Laboratory Manual,” 1989, 2nd Ed., Cold Spring Harbor Laboratory Press: New York, N.Y.). Additional non-limiting examples include salting out DNA extraction (see, for example, P. Sunnucks et al., Genetics, 1996, 144:747-756; and S. M. Aljanabi and I. Martinez, Nucl. Acids Res. 1997, 25:4692-4693), the trimethylammonium bromide salts DNA extraction method (see, for example, S. Gustincich et al., BioTechniques, 1991, 11:298-302) and the guanidinium thiocyanate DNA extraction method (see, for example, J. B. W. Hammond et al., Biochemistry, 1996, 240:298-300). There are also numerous commercially available kits that can be used to extract DNA from biological fluids (e.g., cerebrospinal fluid) or cells, for example, Qiagen's Gentra PureGene Cell Kit, QlAamp Circulating Nucleic Acid Kit, QiaAmp DNA Mini Kit, DNeasy Blood and Tissue Kit or QiaAmp DNA Blood Mini Kit (Qiagen, Hilden, Germany), GenomicPrep™ Blood DNA Isolation Kit (Promega, Madison, Wis.) and GFX™ Genomic Blood DNA Purification Kit (Amersham, Piscataway, N.J.) can be used to obtain DNA from a cerebrospinal fluid sample from a subject.

Methylation Detection Techniques of this Example

Various methylation analysis procedures are known in the art, and can be used with the methods of this Example. These assays allow for determination of the methylation state of one genomic locus, e.g., one or more CpG sites or islands within a nucleic acid obtained from a cerebrospinal fluid sample. In addition, the methods can be used to quantify the methylation of a genomic locus. Such assays involve, among other techniques, DNA sequencing of bisulfite-treated DNA, PCR (for sequence-specific amplification), digital PCR and use of methylation-sensitive restriction enzymes.

In certain embodiments, methylation-specific PCR can be used to determine the methylation status of a genomic loci. Methylation-specific PCR is based on a chemical reaction of sodium bisulfite with DNA that converts unmethylated cytosines, e.g., of CpG dinucleotides, to uracil or UpG, followed by traditional PCR. Methylated cytosines will not be converted in this process and primers can be designed to overlap the methylation site, e.g., CpG site, of interest, thereby allowing one to determine the methylation status of the methylation site as methylated or unmethylated. Additionally, restriction enzyme digestion of PCR products amplified from bisulfite-converted DNA may be used, e.g., by using the method described by Sadri & Hornsby (Nucl. Acids Res. 1996; 24:5058-5059) or COBRA (Combined Bisulfite Restriction Analysis) (Xiong & Laird, Nucleic Acids Res. 1997; 25:2532-2534).

In certain embodiments, whole genome bisulfite sequencing, which is a high-throughput genome-wide analysis of DNA methylation, can be used to determine the methylation status of multiple genomic loci. It is based on sodium bisulfite conversion of genomic DNA, as described above, which is then sequenced on a next-generation sequencing platform. The sequences obtained are then re-aligned to the reference genome to determine the methylation states of cytosines, e.g., of CpG dinucleotides, present within the analyzed genomic loci based on mismatches resulting from the conversion of unmethylated cytosines into uracil.

In certain embodiments, genome-wide DNA methylation profiling can be performed using commercially-available arrays, thereby allowing the interrogation of multiple genomic loci, e.g., multiple CpG sites. Non-limiting examples of such arrays include HumanMethylation BeadChips (Illumina, San Diego, Calif.) and Infinium MethylationEPIC kit (Illumina). Additional methods for analyzing the methylation state of multiple genomic loci is provided in Yong et al., Epigenetics & Chromatin 2016; 9:26, which is incorporated by reference herein.

Kits of this Example

This Example provides kits for diagnosing, monitoring, classifying and/or treating a subject with a CNS disorder. The kits of this Example comprise a means for determining and/or detecting the methylation status of one or more genomic loci.

Types of kits of this Example include, but are not limited to, packaged probe and primer sets (e.g., TaqMan probe/primer sets), arrays/microarrays, which further contain one or more probes, primers or other detection reagents for determining the methylation state and/or level of one or more genomic loci. For example, but not by way of limitation, a kit of this Example can include one or more probes or primers for detecting the methylation state of one or more genomic loci. In certain embodiments, the one or more genomic loci comprise a CpG site. In certain embodiments, one or more of the genomic loci do not comprise a CpG site. For example, but not by way of limitation, about 5% or more, about 10% or more, about 15% or more, about 20% or more, about 25% or more, about 30% or more, about 35% or more, about 40% or more, about 45% or more, about 50% or more, about 55% or more, about 60% or more, about 65% or more or about 70% or more of the one or more genomic loci detected by the primers or probes of this Example comprise one or more CpG sites.

In certain non-limiting embodiments, a primer and/or probe of this Example can be at least about 10 nucleotides or at least about 15 nucleotides or at least about 20 nucleotides in length and/or up to about 200 nucleotides or up to about 150 nucleotides or up to about 100 nucleotides or up to about 75 nucleotides or up to about 50 nucleotides in length.

In a further non-limiting embodiment, the oligonucleotide primers and/or probes can be immobilized on a solid surface or support, for example, on a nucleic acid microarray, wherein the position of each oligonucleotide primer and/or probe bound to the solid surface or support is known and identifiable.

In certain non-limiting embodiments, the kits of this Example additionally include other components such as, but not limited to, a buffer, enzymes such as DNA polymerases or ligases, nucleotides such as deoxynucleotide triphosphates, positive control sequences, negative control sequences and the like necessary to carry out an assay or reaction to detect the methylation state of a genomic locus.

In certain embodiments, the kits of this Example include a container comprising one or more probes and/or primers for detecting the methylation state of one or more genomic loci. In certain embodiments, the kits further include instructions for use, e.g., the instructions can describe that a particular methylation status of a genomic locus is indicative of a CNS disorder in a subject. The instructions can be printed directly on the container (when present), or as a label applied to the container, or as a separate sheet, pamphlet, card or folder supplied in or with the container.

Reports, Programmed Computers, and Systems of this Example

In certain embodiments, the presently disclosed diagnosis and/or monitoring of a CNS disorder in a subject based on the methylation status of one or more genomic loci can be referred to herein as a “report.” A tangible report can optionally be generated as part of a testing process (which can be interchangeably referred to herein as “reporting,” or as “providing” a report, “producing” a report or “generating” a report).

Examples of tangible reports can include, but are not limited to, reports in paper (such as computer-generated printouts of test results) or equivalent formats and reports stored on computer readable medium (such as a CD, USB flash drive or other removable storage device, computer hard drive, or computer network server, etc.). Reports, particularly those stored on computer readable medium, can be part of a database, which can optionally be accessible via the internet (such as a database of patient records or genetic information stored on a computer network server, which can be a “secure database” that has security features that limit access to the report, such as to allow only the patient and the patient's medical practitioners to view the report while preventing other unauthorized individuals from viewing the report, for example). In addition to, or as an alternative to, generating a tangible report, reports can also be displayed on a computer screen (or the display of another electronic device or instrument).

A report can include, for example, an individual's medical history, or can just include size, presence, absence or levels of one or more markers (for example, a report on computer readable medium such as a network server can include hyperlink(s) to one or more journal publications or websites that describe the medical/biological implications). Thus, for example, the report can include information of medical/biological significance as well as optionally also including information regarding the methylation status of relevant genomic loci, or the report can just include information regarding the methylation status of relevant genomic loci without other medical/biological significance.

A report can further be “transmitted” or “communicated” (these terms can be used herein interchangeably), such as to the individual who was tested, a medical practitioner (e.g., a doctor, nurse, clinical laboratory practitioner, genetic counselor, etc.), a healthcare organization, a clinical laboratory, and/or any other party or requester intended to view or possess the report. The act of “transmitting” or “communicating” a report can be by any means known in the art, based on the format of the report. Furthermore, “transmitting” or “communicating” a report can include delivering a report (“pushing”) and/or retrieving (“pulling”) a report. For example, reports can be transmitted/communicated by various means, including being physically transferred between parties (such as for reports in paper format) such as by being physically delivered from one party to another, or by being transmitted electronically or in signal form (e.g., via e-mail or over the internet, by facsimile, and/or by any wired or wireless communication methods known in the art) such as by being retrieved from a database stored on a computer network server, etc.

In certain embodiments, the disclosed subject matter provides computers (or other apparatus/devices such as biomedical devices or laboratory instrumentation) programmed to carry out the methods described herein, e.g., to perform the algorithm of this Example (see Example Embodiment E). In certain embodiments, the system can be controlled by the individual and/or their medical practitioner in that the individual and/or their medical practitioner requests the test, receives the test results back and (optionally) acts on the test results to reduce the individual's CNS disorder risk or treat the individual, such as by implementing a disorder management system.

The following Example Embodiments are offered to more fully illustrate the disclosure of this Example, but are not to be construed as limiting the scope thereof.

Example Embodiment D of Example 2: Non-Invasive Molecular Phenotyping of the Human Brain Via Epigenomic Liquid Biopsy of Cerebrospinal Fluid

Example Embodiment D discovered methods of epigenomic liquid biopsy for the comprehensive analysis of cell-free DNA (cfDNA) methylation signatures in the human central nervous system (CNS). Example Embodiment D used solution phase hybridization and high throughput bisulfite sequencing to compare DNA methylation signatures of cfDNA obtained from cerebrospinal fluid (CSF) and plasma. Recovery of cfDNA from CSF was relatively low (68 to 840/mL CSF) compared to plasma. Distributions of CpG methylation were significantly altered between CSF and plasma, both generally and at the level of specific functional elements such as exons, introns, CpG islands and shores. Sliding window analysis was used to identify differentially methylated CpG sites. Example Embodiment D found numerous gene/locus-specific differences in CpG methylation between cfDNA from CSF and plasma. These loci were more likely to be hypomethylated in CSF compared to plasma. Differentially methylated CpGs in CSF were identified in genes related to branching of neurites and neuronal development. Example Embodiment D found clear association between tissue-specific gene expression in the CNS and cfDNA methylation patterns in CSF. Ingenuity pathways analysis (IPA) of differentially methylated regions identified an enrichment of functional pathways related to neurobiology. The GTEX RNA expression database was used to analyze the presently disclosed data in the context of central nervous system (CNS)-specific gene expression. In conclusion, Example Embodiment D was the first comprehensive quantitative genome-wide analysis of DNA methylation in human CSF. The presently disclosed methods of this Example can be used for epigenomic liquid biopsy of the human CNS for molecular phenotyping of brain-derived DNA methylation signatures.

Example Embodiment D obtained CSF from healthy volunteers, and recovered brain-derived cfDNA from CSF. Example Embodiment D also quantified and catalogued the brain-derived cfDNA in a genome-wide fashion at the level of DNA methylation via bisulfite sequencing. In order to generate preliminary functional insights into brain-specific molecular phenotypes, solution-phase hybridization coupled with high-throughput sequencing was used to compare DNA methylation of cfDNA from CSF with that from plasma.

Methods

Human Samples

Plasma and CSF samples were processed for cell-free nucleic acid analysis as previously described (Chu T et al., PLoS One. 2017; 12(3):e0171882). Healthy human Lumbar puncture for CSF collection was performed under fluoroscopic guidance by the Neuroradiology Department at UPMC. All participants provided informed consent. CSF donors had no personal or first degree-relative history of psychiatric disorder or suicidal behaviour.

Preparation and Analysis of Bisulfite DNA Sequencing Libraries

DNA sequencing libraries were prepared using the SeqCap-Epi kit (Roche). Libraries were sequenced on an Illumina HiSeq 2500 instrument. Reads were quality trimmed and adaptor sequences were removed using Trim-Galore. Reads were aligned using Bismark in paired-end, bowtie2 modes. Unmapped reads were aligned using single-end, bowtie2 modes (Bismark). Duplicates were removed (Bismark). Methylation was called on pair-end and single-end files and then merged. CpG methylation information for autosomes was read into MethylSig for differential methylation analysis.

Results

Example Embodiment D examined the physical properties of cfDNA in CSF. Following extraction, DNA amount was quantified by real time PCR. The resulting yields for a series of CSF samples are shown in Table 5A.

TABLE 5A
Amounts of cfDNA recovered from a series of CSF samples.
Amount of DNA is shown in picograms per mL of starting CSF.
       Amount of cfDNA Recovered from
Sample 1 mL CSF in pg
LP667  145
LP676  164
LP418  180
LP603  228
LP429  278
LP435  143
LP630  840
LP638  309
LP673  188
LP678  402
LP366  68

Quantities of cfDNA are low, ranging from 68 pg to 840 pg per mL of input CSF (mean 268 ng/mL). This is significantly lower than cfDNA recovery from plasma (Table 5B).

TABLE 5B
Amounts of cfDNA recovered from a series of plasma samples
obtained from non-pregnant females. Amount of
DNA is shown in picograms per mL of starting plasma.
Sample DNA pg/mL of Plasma
EN7    3320
EN8    3770
EN9    4260
EN11   2830
EN12   8390
EN13   6390
EN14   2720
EN15   6920
EN16   8060
EN17   8260
EN18   3590
EN20   5350

Example Embodiment D then explored the size distribution of cfDNA from CSF and compared this to plasma. It was found that cfDNA fragments recovered from CSF had a mean size of approximately 155 bp, which was approximately 20 bp less than the mean fragment size of cfDNA recovered from plasma (Table 6).

TABLE 6
Mean fragment length of cfDNA in samples.
	Sample	Mean	Median
	Name (Type)	Fragment Length	Fragment Length
	LP374 (CSF)	154	155
	LP374 (CSF)	158	159
	EN1 (Plasma)	178	175
	EN2 (Plasma)	182	176
	EN3 (Plasma)	178	176
	EN4 (Plasma)	170	170
	EN5 (Plasma)	177	175
	EN6 (Plasma)	181	177
	EN7 (Plasma)	176	174
	EN8 (Plasma)	175	174
	EN9 (Plasma)	177	174
	EN11 (Plasma)	177	175
	EN12 (Plasma)	172	171

Example Embodiment D performed solution phase hybridization capture of bisulfate-converted cfDNA from two CSF and two plasma samples. To further explore size differences between CSF and plasma, fragment size was plotted against read count, which confirmed that cfDNA fragments from CSF were significantly shorter in length than those from plasma (FIG. 26). Furthermore, it was found that cfDNA from CSF exhibits a periodicity of fragment density peaks differing in size by approximately 10 bp (FIG. 27 and Table 7).

TABLE 7
Fragment size distribution for cfDNA from CSF (LP372
and LP374) and plasma (EN1)
	Fragment
	Length	LP372	LP374	EN1
	20	18.1655359	4.12935877	0
	21	12.1103572	8.25871754	0
	22	36.3310717	12.3880763	0
	23	0	8.25871754	0
	24	30.2758931	16.5174351	0
	25	18.1655359	16.5174351	1
	26	12.1103572	12.3880763	1
	27	30.2758931	20.6467938	0
	28	30.2758931	28.9055114	0
	29	30.2758931	66.0697403	3
	30	121.103572	119.751404	4
	31	84.7725007	90.8458929	4
	32	145.324287	115.622046	3
	33	211.931252	152.786274	3
	34	242.207145	251.890885	4
	35	260.372681	264.278961	5
	36	351.20036	280.796396	6
	37	423.862503	545.075358	16
	38	611.573041	491.393694	12
	39	708.455898	730.896502	21
	40	666.069648	743.284579	18
	41	811.393935	958.011235	18
	42	865.890543	962.140593	15
	43	847.725007	1048.85713	28
	44	1144.42876	1350.30032	38
	45	1271.58751	1643.48479	47
	46	1441.13251	1742.5894	48
	47	1592.51198	2229.85374	51
	48	1955.82269	2518.90885	78
	49	2682.44413	3274.5815	75
	50	2960.98234	3266.32279	119
	51	2809.60288	3563.63662	116
	52	2767.21663	3534.73111	105
	53	2858.04431	4021.99544	115
	54	2906.48574	4083.93582	133
	55	3039.69967	4492.74234	146
	56	3663.38306	4860.25527	174
	57	4196.23878	5520.95268	223
	58	4813.867	6838.21812	291
	59	6339.77201	7808.61743	292
	60	6618.31023	8779.01674	341
	61	6981.62095	9220.85813	316
	62	6503.26184	8704.68829	348
	63	5958.29576	9319.96274	344
	64	6479.04112	9241.50493	378
	65	6715.19309	9823.74451	424
	66	7139.05559	9943.49592	487
	67	7902.0081	10587.6759	489
	68	8513.58114	12045.3395	577
	69	10856.9353	13296.5352	689
	70	11856.0397	14308.2281	810
	71	10826.6594	14081.1134	736
	72	10669.2247	13887.0335	730
	73	10608.6729	13903.551	740
	74	10536.0108	14502.308	769
	75	11074.9217	14750.0695	853
	76	11886.3156	15596.5881	936
	77	12346.5092	16121.0166	1016
	78	15265.1053	18293.0594	1236
	79	19425.013	21522.2179	1580
	80	23173.1686	24573.814	1851
	81	22997.5684	24668.7893	1851
	82	19848.8755	22067.2933	1602
	83	16627.5205	19647.489	1309
	84	16312.6512	18590.3732	1297
	85	16845.5069	20597.2415	1406
	86	17160.3762	21447.8895	1456
	87	19588.5028	22248.9851	1603
	88	22016.6295	25589.6363	1899
	89	28132.3599	30920.6385	2252
	90	38734.9776	38745.7733	3086
	91	43439.8514	41652.8419	3526
	92	44414.7352	41809.7575	3315
	93	37475.5005	36932.9848	2766
	94	31826.0188	32126.4112	2468
	95	28762.0984	31028.0018	2284
	96	29670.3752	32118.1525	2426
	97	32782.737	33625.3685	2715
	98	37669.2662	38675.5742	3178
	99	50094.4927	47983.1489	3978
	100	50887.7211	48317.627	4978
	101	66897.6134	61651.3264	4602
	102	68877.6568	63587.9957	5316
	103	66504.0268	62200.5312	4945
	104	59861.4958	56167.538	4933
	105	54327.0626	53206.7878	4683
	106	54212.0142	52054.6967	4904
	107	57233.5483	55159.9744	5361
	108	65135.5564	61172.3208	6047
	109	77742.4383	73031.8392	7117
	110	96967.6304	89735.0954	8777
	111	114115.896	101532.673	10212
	112	115865.843	102593.919	10374
	113	99977.0542	90267.7827	9348
	114	85783.7155	80501.8492	8594
	115	81666.194	76033.883	8582
	116	85214.5287	76727.6153	9247
	117	88641.7598	79642.9426	10058
	118	98214.9972	88207.2327	11458
	119	116774.12	102602.177	13897
	120	148376.097	128472.61	17219
	121	191846.224	159182.651	22351
	122	221873.855	183426.117	25397
	123	215588.58	177603.721	24829
	124	168909.208	143379.595	20151
	125	134443.131	115650.951	17103
	126	119547.391	104823.772	16914
	127	123253.161	107887.757	18354
	128	133613.571	115605.528	20682
	129	152814.543	129438.88	23258
	130	178736.762	155924.587	28592
	131	224622.906	188067.516	36576
	132	280808.908	234440.215	47251
	133	333131.707	276683.555	57525
	134	339907.452	280536.247	59591
	135	291865.665	243396.794	54337
	136	246645.591	204498.234	50144
	137	236491.056	193827.971	53729
	138	255867.628	207343.363	63398
	139	295868.138	244363.064	78271
	140	354131.066	295988.307	96596
	141	401791.377	343785.635	110936
	142	418909.367	362132.376	118556
	143	418564.222	364655.414	121296
	144	418213.022	365960.292	125102
	145	415790.95	363618.945	130557
	146	385987.361	337583.338	134678
	147	360489.004	313653.704	139586
	148	353943.356	304057.074	150072
	149	362687.034	310907.681	163436
	150	383631.897	331847.659	178397
	151	413683.748	367376.662	195570
	152	446091.064	403347.506	210262
	153	448652.405	410627.565	215911
	154	425564.009	390885.101	216568
	155	409753.937	375044.881	223618
	156	406998.831	364622.379	242943
	157	408203.811	363412.477	271460
	158	416880.882	368929.301	308096
	159	434743.659	386099.174	345628
	160	446666.306	404082.532	378777
	161	453084.795	418754.144	404958
	162	449808.944	423007.383	436486
	163	462942.626	430469.134	486903
	164	476584.944	442246.066	558909
	165	488083.728	445896.419	623138
	166	484038.868	437022.427	657408
	167	469554.881	430138.786	670547
	168	451849.539	421310.217	672348
	169	432067.27	414979.91	662314
	170	410347.345	403169.944	648068
	171	387537.487	391421.918	626438
	172	368942.033	377828.069	613248
	173	350461.628	370304.377	610963
	174	339992.224	359402.87	616896
	175	328087.743	354881.222	619799
	176	314306.157	344124.242	614507
	177	298005.616	330745.12	591992
	178	278035.637	314904.9	559319
	179	260530.115	303350.954	523767
	180	238113.844	287106.057	486651
	181	218168.086	271587.926	453452
	182	202509.394	256123.478	426739
	183	186862.812	242525.499	402693
	184	174068.22	230884.837	386125
	185	161128.303	220681.191	374770
	186	150180.54	208334.409	363365
	187	140298.489	198139.022	351750
	188	130507.265	187956.023	339394
	189	122235.891	178252.03	320487
	190	113716.254	169233.51	299582
	191	103325.568	159356.084	278594
	192	94775.6557	149561.245	258141
	193	87939.3591	140695.512	239078
	194	80539.9308	130954.355	225250
	195	74817.787	123884.892	213210
	196	70645.769	116522.246	204351
	197	63882.1344	109605.57	196298
	198	59056.1571	103110.088	189910
	199	55374.6085	98320.0323	180675
	200	51838.3842	92993.1595	172164
	201	48598.8636	87819.073	161500
	202	44087.7555	82690.4094	149421
	203	41901.836	77503.9348	139531
	204	37723.7628	70661.5873	129951
	205	35453.0708	67329.1947	120744
	206	32104.557	63389.7865	114771
	207	30390.9415	59153.0644	108677
	208	28374.567	56192.3141	103424
	209	26140.2061	53499.9722	99371
	210	24420.5354	49081.5583	94738
	211	23136.8375	46884.7395	88154
	212	21532.2152	42895.7789	82930
	213	19079.8678	40236.4719	76873
	214	17348.0867	37255.0748	71462
	215	16258.1546	35491.8386	66474
	216	15761.6299	32716.9095	61730
	217	14496.0976	30065.8612	58942
	218	13624.1519	28802.2774	55677
	219	12952.0271	27200.0862	52944
	220	11516.9497	25081.7252	50446
	221	11323.184	23206.9963	47643
	222	10584.4522	22459.5823	44112
	223	9760.94793	20750.0278	41517
	224	9113.04382	19296.4935	38807
	225	8574.13292	18008.1336	36644
	226	8089.71864	16249.0268	33759
	227	7883.84256	15398.3789	32127
	228	6793.91041	14407.3327	30947
	229	6545.64809	13924.1978	29433
	230	6242.88916	13094.1967	27861
	231	5728.19897	12144.4441	26902
	232	5582.87469	11112.1044	25394
	233	5255.89504	10612.452	24190
	234	4941.02575	10397.7254	22850
	235	4396.05968	9319.96274	21545
	236	3923.75575	8287.62305	20590
	237	4208.34914	8010.95601	19692
	238	3651.27271	7284.18887	19022
	239	3742.10039	7139.66131	18628
	240	3505.94842	6755.63095	17904
	241	3312.1827	6239.4611	17751
	242	3221.35503	6053.63996	16986
	243	2991.25824	5657.22151	16639
	244	2991.25824	5677.86831	16073
	245	2543.17502	5128.66359	15498
	246	2367.57484	4765.28002	15162
	247	2494.73359	4426.6726	14774
	248	2252.52645	4141.74685	14574
	249	2131.42287	4042.64224	14231
	250	2076.92627	4013.73672	14306
	251	1980.04341	3832.04494	14044
	252	1919.49162	3526.47239	13829
	253	2082.98145	3472.79073	13723
	254	1877.10537	3134.18331	13464
	255	1640.95341	3204.38241	13434
	256	1531.96019	2820.35204	13150
	257	1677.28448	2762.54102	12823
	258	1580.40162	2787.31717	12774
	259	1598.56716	2494.1327	12713
	260	1392.69108	2361.99322	12779
	261	1186.81501	2295.92348	12662
	262	1519.84983	2316.57027	12851
	263	1398.74626	2452.83911	12850
	264	1471.4084	2196.81887	12819
	265	1289.75305	2072.9381	12584
	266	1223.14608	1973.83349	12280
	267	1180.75983	1973.83349	12021
	268	1162.59429	1788.01235	12208
	269	1017.27001	1709.55453	12278
	270	1168.64947	1816.91786	12571
	271	1059.65626	1763.23619	12615
	272	1077.82179	1746.71876	12973
	273	993.049294	1713.68389	13075
	274	974.883758	1610.44992	13129
	275	1089.93215	1523.73339	13062
	276	1071.76662	1552.6389	12712
	277	932.497507	1560.89762	12750
	278	1011.21483	1362.68839	12917
	279	914.331972	1358.55904	13137
	280	999.104472	1271.8425	13614
	281	914.331972	1325.52417	14599
	282	853.780185	1478.31044	14721
	283	932.497507	1391.59391	15192
	284	950.663043	1441.14621	15042
	285	944.607865	1226.41955	15132
	286	968.828579	1189.25533	15183
	287	769.007685	1280.10122	15643
	288	805.338756	1267.71314	15887
	289	896.166436	1226.41955	16424
	290	829.559471	1094.28007	17111
	291	884.056078	1280.10122	17625
	292	968.828579	1185.12597	18017
	293	762.952506	1342.0416	18332
	294	805.338756	1164.47917	19032
	295	968.828579	1296.61865	19504
	296	702.40072	1222.2902	20330
	297	993.049294	1193.38468	20884
	298	841.669828	1292.48929	21544
	299	853.780185	1176.86725	22156
	300	932.497507	1218.16084	22486
	301	823.504292	1234.67827	23103
	302	878.0009	1197.51404	23665
	303	884.056078	1267.71314	23679
	304	829.559471	1209.90212	24359
	305	1023.32519	1218.16084	25728
	306	0	1337.91224	0
	307	0	1086.02136	0
	308	0	1370.94711	0
	309	0	1238.80763	0
	310	0	1176.86725	0
	311	0	1271.8425	0
	312	0	1218.16084	0
	313	0	1309.00673	0
	314	0	1247.06635	0
	315	0	1267.71314	0
	316	0	1309.00673	0
	317	0	1300.74801	0
	318	0	1090.15072	0
	319	0	1214.03148	0
	320	0	1242.93699	0
	321	0	1230.54891	0
	322	0	1086.02136	0
	323	0	1094.28007	0
	324	0	1234.67827	0
	325	0	1205.77276	0
	326	0	1081.892	0
	327	0	1065.37456	0
	328	0	1024.08097	0
	329	0	1086.02136	0

To explore the distributions of CpG methylation in cfDNA from CSF, Example Embodiment D examined the percentages of CpGs within cfDNA fragments that are methylated at low methylation (LM) levels (<20%), high methylation (HM) levels (>80%) or at intermediate methylation (IM) levels (20-80%) in both CSF and plasma. It was found that CpG sites in CSF, as in plasma, are largely distributed in a biphasic manner with large numbers of LM sites and very few IM sites. This is consistent with previous reports from DNA methylation analysis in differentiated intact adult cells (Chu T et al., PLoS One. 2011; 6(2):e14723). cfDNA in CSF is represented by fewer IM sites than plasma and greater numbers of LM and HM sites (FIG. 28A).

Example Embodiment D further explored distributions of CpG methylation in the context of distinct genomic elements, specifically introns, exons, promoters, CpG islands and CpG island shores. The relationship between Group (CSF/Plasma) and Response (<20%, 20-80% and >80%) was tested, after controlling for Category (intergenic region, intron, exon or promoter) using Cochran-Mantel-Haenszel test. The p-value (<0.0001) indicates that the numbers of sites at <20%, 20-80% and >80% are significantly different between CSF and Plasma after adjusting for category. Specifically, distribution of DNA methylation varies across CSF and plasma samples adjusted for categories (intergenic, intron, exon or promoter) such that the CSF DNA methylome contains a greater percentage of >80% methylation in exons and introns compared to plasma (FIG. 28B). DNA methylation distributions in CGIs and CGI shores appeared to be largely similar between one of the CSF samples (LP374) and the plasma samples. In general, the rate of methylation was lower in CGIs followed by shores and other in these samples (FIG. 28C).

Example Embodiment D further explored how the methylation signatures in cfDNA fragments from CSF differ from those of plasma in a gene/locus-specific fashion. A sliding windows analysis was performed to identify genomic regions (250 bp) whose CpG methylation characteristics differ significantly between cfDNA from CSF and plasma. As shown in FIG. 29 numerous differences were identified in or around known genes that differ between CSF and plasma. Notably, around two-fold more loci were identified that were significantly (p=<0.01) more methylated in plasma than CSF (n=394) than vice versa (n=216). Thus it appears that cfDNA from CSF is relatively hypomethylated compared to cfDNA from plasma. FIG. 30 contains the 100-most differentially methylated loci identified in which cfDNA from CSF is methylated at a lower rate than that of plasma. The list clearly shows that these loci are enriched for loci that are linked to genes that encode brain- or neuron-specific proteins or are related to central nervous system molecular physiology. Examples include, ADAP1 (Stricker R et al., Biol Chem. 2014; 395(11):1321-40), ANO1 (Cho S J et al., Cell Tissue Res. 2014; 357(3):563-9), ARHGEF7 (Chia R et al., Nat Commun. 2014; 5:5827), ARRB2 (Liu X et al., Proc Natl Acad Sci USA. 2015; 112(14):4483-8), FBXO33 (Sanchez-Mora C et al., Neuropsychopharmacology. 2015; 40(4):915-26), FOXN3, KLF15 (Ohtsuka T et al., Stem Cells. 2011; 29(11):1817-28), PCBP3 (Boschi N M et al., Neuroscience. 2015; 286:1-12), RAPGEF2 (Jiang S Z et al., eNeuro. 2017; 4(5)), RFX3 (Benadiba C et al., PLoS Genet. 2012; 8(3):e1002606), RGS3 (Qiu R et al., Stem Cells. 2010; 28(9):1602-10; Qiu R et al., J Cell Biol. 2008; 181(6):973-83), RYK (Lanoue V et al., Sci Rep. 2017; 7(1):5965). These account for ˜38% of the total number of genes in this group (n=32) and the remainder are represented largely by, as yet, functionally uncharacterized loci including a number of micro RNA encoding sequences. Interestingly, FOXN3 has been shown to be upregulated and demethylated upon neuronal activation (Grassi D et al., Cereb Cortex. 2017; 27(8):4166-81). FIG. 31 contains the 100-most differentially methylated loci identified in which cfDNA of CSF is methylated at a higher rate than that of plasma. Little or no enrichment is apparent for loci that are linked to genes encoding brain- or neuron-specific proteins. Specifically the present disclosure found only four such genes, CNTFR (Askvig J M et al., Exp Neurol. 2012; 233(1):243-52), BAG3 (Santoro A et al., Cell Tissue Res. 2017; 368(2):249-58), KLF15 (Ohtsuka T et al., Stem Cells. 2011; 29(11):1817-28) and SOX1-OT (Ahmad A et al., Cell Mol Life Sci. 2017; 74(22):4245-58), contained these such loci and these account for only ˜14% of the total number of genes in this group (n=29).

To search for biological themes within the data, Ingenuity Pathways Analysis (IPA) was performed to search for ontological patterns that characterize the sample-specific differentially methylated regions. The most significant canonical pathway identified was “axonal guidance”, containing a number of differentially methylation loci including those that map to ARHGEF7, HKR1, ITGB1, PLCH2, PRKCZ, RGS3, WNT8A. All 45 “top diseases and bio-functions” identified by IPA were related to neurobiology with the top 3 consisting of the following: 1) branching of neuritis, 2) development of neurons, and 3) developmental process of synapse (Table 8).

TABLE 8
Top “diseases and bio-functions” identified by Ingenuity Pathways Analysis software
for loci whose methylation levels differ significantly between cfDNA from plasma and CSF
	Diseases or
	Functions			#
Categories	Annotation	p-Value	Molecules	Molecules
Cell Morphology, Cellular Assembly and	Branching of	9.24E−04	ADAP1,	5
Organization, Cellular Development, Cellular	neurites		ARHGEF7,
Function and Maintenance, Cellular Growth			ITGB1,
and Proliferation, Embryonic Development,			RAPGEF2,
Nervous System Development and Function,			RYK
Organismal Development, Tissue
Development
Cellular Development, Cellular Growth and	Development of	1.26E−03	ADAP1,	8
Proliferation, Nervous System Development	neurons		ARHGEF7,
and Function, Tissue Development			ITGB1,
			PDZRN3,
			PRKCZ,
			RAPGEF2,
			RYK, SUN1
Cell-To-Cell Signaling and Interaction, Cellular	Developmental	1.74E−03	ARHGEF7,	4
Assembly and Organization, Cellular	process of synapse		ITGB1,
Development, Cellular Function and			PDZRN3,
Maintenance, Cellular Growth and			SUN1
Proliferation, Nervous System Development
and Function, Tissue Development
Cellular Movement, Nervous System	Delay in migration	1.82E−03	ITGB1	1
Development and Function	of Schwann cells
Cell Morphology, Cellular Assembly and	Polarization of	1.82E−03	PRKCZ	1
Organization, Nervous System Development	axons
and Function
Cell Morphology, Nervous System	Polarization of	1.82E−03	PRKCZ	1
Development and Function	astrocytes
Cellular Assembly and Organization, Cellular	Segregation of	1.82E−03	ITGB1	1
Movement, Nervous System Development	sensory axons
and Function
Cell Morphology, Cellular Assembly and	Arborization of	1.82E−03	ITGB1	1
Organization, Cellular Development, Cellular	spinal nerve
Function and Maintenance, Cellular Growth
and Proliferation, Nervous System
Development and Function, Organismal
Development, Tissue Development
Cellular Movement, Nervous System	Migration of	2.40E−03	ITGB1,	2
Development and Function	granule cells		RGS3
Cellular Movement, Nervous System	Migration of	2.47E−03	ITGB1,	4
Development and Function	neurons		RAPGEF2,
			RGS3,
			SOX1
Cell Morphology, Cellular Assembly and	Dendritic	2.92E−03	ADAP1,	4
Organization, Cellular Development, Cellular	growth/branching		ARHGEF7,
Function and Maintenance, Cellular Growth			ITGB1,
and Proliferation, Embryonic Development,			RAPGEF2
Nervous System Development and Function,
Organismal Development, Tissue
Development
Cellular Development, Cellular Growth and	Expansion of	3.63E−03	ITGB1	1
Proliferation, Nervous System Development	satellite cells
and Function
Nervous System Development and Function,	Adhesion of brain	3.63E−03	ITGB1	1
Tissue Development	tissue
Cell Morphology, Cellular Assembly and	Neuritogenesis	5.13E−03	ADAP1,	6
Organization, Cellular Development, Cellular			ARHGEF7,
Function and Maintenance, Cellular Growth			ITGB1,
and Proliferation, Nervous System			PRKCZ,
Development and Function, Organismal			RAPGEF2,
Development, Tissue Development			RYK
Cell-To-Cell Signaling and Interaction, Cellular	Chemoattraction of	5.44E−03	RGS3	1
Movement, Nervous System Development	granule cells
and Function
Cellular Movement, Nervous System	Chemotaxis of	5.44E−03	RGS3	1
Development and Function	cerebellar granule
	cell
Cell Morphology, Cellular Assembly and	Formation of	5.44E−03	PRKCZ	1
Organization, Cellular Development, Cellular	supernumerary
Function and Maintenance, Cellular Growth	axons
and Proliferation, Nervous System
Development and Function, Organismal
Development, Tissue Development
Cell Morphology, Cellular Assembly and	Extension of	6.77E−03	ITGB1,	3
Organization, Cellular Function and	neurites		PRKCZ,
Maintenance, Nervous System Development			RAPGEF2
and Function
Cellular Assembly and Organization, Cellular	Retraction of	6.79E−03	ARHGEF7,	2
Compromise, Nervous System Development	neurites		RYK
and Function
Nervous System Development and Function,	Abnormal	9.05E−03	SUN1	1
Neurological Disease, Organ Morphology,	morphology of
Organismal Development, Organismal Injury	cerebellum fissure
and Abnormalities
Cell Cycle, Nervous System Development and	Mitogenesis of	9.05E−03	PRKCZ	1
Function	nervous tissue cell
	lines
Cellular Development, Cellular Growth and	Proliferation of	1.26E−02	ITGB1	1
Proliferation, Nervous System Development	satellite cells
and Function
Cellular Movement, Endocrine System	Migration of	1.26E−02	ITGB1	1
Development and Function, Nervous System	lutenizing hormone-
Development and Function	releasing hormone
	neurons
Cell-To-Cell Signaling and Interaction,	Binding of microglia	1.26E−02	ITGB1	1
Hematological System Development and
Function, Immune Cell Trafficking,
Inflammatory Response, Nervous System
Development and Function
Cellular Assembly and Organization, Cellular	Retraction of axons	1.44E−02	RYK	1
Compromise, Nervous System Development
and Function
Cell-To-Cell Signaling and Interaction,	Binding of Schwann	1.44E−02	ITGB1	1
Nervous System Development and Function	cells
Nervous System Development and Function	Analgesic tolerance	1.62E−02	ARRB2	1
Cellular Assembly and Organization, Nervous	Sorting of axons	1.80E−02	ITGB1	1
System Development and Function
Cell Morphology, Cellular Assembly and	Extension of	1.80E−02	RAPGEF2	1
Organization, Cellular Development, Cellular	dendrites
Function and Maintenance, Cellular Growth
and Proliferation, Embryonic Development,
Nervous System Development and Function,
Organismal Development, Tissue
Development
Nervous System Development and Function	Myelination of	1.80E−02	ITGB1	1
	peripheral nervous
	system
Cell Morphology, Cellular Assembly and	Size of dendritic	1.98E−02	ITGB1	1
Organization, Nervous System Development	trees
and Function, Tissue Morphology
Cell Morphology, Cellular Assembly and	Morphogenesis of	2.16E−02	ARHGEF7	1
Organization, Cellular Development, Cellular	dendritic spines
Function and Maintenance, Cellular Growth
and Proliferation, Embryonic Development,
Nervous System Development and Function,
Organismal Development, Tissue
Development
Nervous System Development and Function,	Quantity of radial	2.69E−02	ITGB1	1
Tissue Morphology	glial cells
Cell-To-Cell Signaling and Interaction,	Synaptic	2.69E−02	PRKCZ	1
Nervous System Development and Function	transmission of CA1
	neuron
Cellular Movement, Nervous System	Migration of	2.87E−02	RGS3	1
Development and Function	cerebellar granule
	cell
Cellular Movement, Nervous System	Migration of	3.04E−02	ITGB1	1
Development and Function	oligodendrocytes
Cellular Assembly and Organization, Nervous	Complexity of	3.04E−02	ITGB1	1
System Development and Function	dendritic trees
Cell Morphology, Cell-To-Cell Signaling and	Loss of synapse	3.40E−02	ITGB1	1
Interaction, Cellular Assembly and
Organization, Nervous System Development
and Function, Neurological Disease, Tissue
Morphology
Cell Morphology, Cellular Assembly and	Neuritogenesis of	3.75E−02	ITGB1	1
Organization, Cellular Development, Cellular	pheochromocytoma
Function and Maintenance, Cellular Growth	cell lines
and Proliferation, Nervous System
Development and Function, Organismal
Development, Tissue Development
Nervous System Development and Function	Chemically-elicited	3.75E−02	ARRB2	1
	antinociception
Cell Death and Survival, Nervous System	Survival of	3.92E−02	ITGB1	1
Development and Function	oligodendrocytes
Embryonic Development, Nervous System	Formation of brain	4.30E−02	RAPGEF2,	4
Development and Function, Organ			RYK,
Development, Organismal Development,			SOX1,
Tissue Development			SUN1
Auditory and Vestibular System Development	Abnormal	4.44E−02	SUN1	1
and Function, Auditory Disease, Cell	morphology of
Morphology, Nervous System Development	outer hair cells
and Function, Neurological Disease, Organ
Morphology, Organismal Development,
Organismal Injury and Abnormalities, Tissue
Morphology
Cell Morphology, Cellular Assembly and	Formation of	4.79E−02	ITGB1,	2
Organization, Cellular Development, Cellular	dendrites		RAPGEF2
Function and Maintenance, Cellular Growth
and Proliferation, Nervous System
Development and Function, Organismal
Development, Tissue Development

To further explore the functional significance of the data, Example Embodiment D used the Genotype-Tissue Expression (GTEx) project database. GTEx is an on-going effort to build a comprehensive public resource to study tissue-specific gene expression and regulation. Example Embodiment D identified genes whose expressions are low in whole blood and highly expressed in neuronal tissues. It was rationalized that leukocytes are the major (though probably not exclusive) contributors to cfDNA in plasma whereas neuronal tissues are the primary contributors to cfDNA in CSF. Thus, by identifying tissue-specific differences in gene expression in this context, it was hoped to illuminate the DNA methylation differences identified between cfDNA from plasma and CSF. As expected, a significant number of the differentially methylated regions overlap with genes whose expressions are high in the brain and low in whole blood. Examples of these genes are shown in Table 9 and FIG. 32.

TABLE 9
Differentially methylated regions that overlap with genes whose expressions are
high in the brain and low in whole blood.
				Difference %
		%	%	Methylation
Chromo	Start	Methylation	Methylation	Plasma		Gene
some	Coordinate	Plasma	CSF	Minus CSF	p Value	Name
chr7	158915201	65.14%	83.05%	−17.91%	5.72E−07	WDR60
chr3	126345501	84.19%	98.46%	−14.27%	2.91E−07	KLF15
chr21	45634201	88.06%	99.15%	−11.09%	1.16E−21	PCBP3
chr19	37293451	84.02%	69.00%	15.02%	7.36E−06	HKR1
chr8	139704801	16.59%	1.16%	15.42%	1.01E−12	KCNK9
chr19	37293351	84.57%	69.00%	15.57%	3.14E−06	HKR1
chr19	37293401	85.09%	69.00%	16.09%	1.46E−06	HKR1
chr19	37293301	86.43%	69.89%	16.54%	1.56E−06	HKR1
chr10	133113251	90.10%	69.80%	20.30%	1.20E−07	ADGRA1
chr7	158915451	67.25%	46.36%	20.88%	1.65E−05	WDR60
chr19	37293501	81.03%	54.55%	26.49%	1.10E−06	HKR1
chr10	133113051	84.77%	58.10%	26.68%	5.04E−08	ADGRA1
chr3	40076901	77.97%	50.30%	27.67%	6.05E−19	MYRIP
chr3	40077051	78.19%	49.84%	28.34%	6.86E−20	MYRIP
chr3	40076951	79.55%	50.30%	29.25%	2.94E−22	MYRIP
chr3	40077001	80.28%	50.30%	29.98%	1.27E−23	MYRIP
chr3	40077101	73.45%	42.98%	30.47%	3.83E−17	MYRIP
chr14	89161951	85.57%	48.05%	37.52%	1.35E−10	FOXN3
chr14	89162001	85.57%	48.05%	37.52%	1.35E−10	FOXN3
chr14	89161901	86.40%	48.05%	38.35%	4.91E−11	FOXN3
chr14	89161851	87.29%	48.05%	39.24%	1.85E−11	FOXN3
chr4	159421451	68.88%	26.56%	42.32%	2.83E−11	RAPGEF2
chr14	89161801	90.92%	48.05%	42.87%	1.79E−13	FOXN3
chr4	159421401	72.70%	26.56%	46.13%	1.09E−13	RAPGEF2
chr4	159421501	72.82%	26.56%	46.25%	5.82E−14	RAPGEF2
chr4	159421551	72.82%	26.56%	46.25%	5.82E−14	RAPGEF2
chr4	159421601	73.76%	26.56%	47.20%	3.73E−14	RAPGEF2
chr21	45634451	77.96%	29.60%	48.36%	3.26E−30	PCBP3
chr3	126345251	79.17%	28.57%	50.60%	2.04E−13	KLF15

Of twenty-nine windows that map to genes whose expressions are at least 5-fold higher in brain than whole blood, 26 were found to significantly hypomethylated in cfDNA from CSF compared to plasma. Examples of genes displaying elevated gene expression and altered DNA methylation in brain versus whole blood include KLF15, which is thought to play a critical role in the maintenance of neural stem cells at late embryonic stages and functions as a transcriptional activator to promote dopamine D2 receptor expression in neurons (Ohtsuka T et al., Stem Cells. 2011; 29(11):1817-28; Zhou J et al., Biochem Biophys Res Commun. 2017; 492(2):269-74). KCNK9, which encodes the protein TASK-3 (a potassium channel protein containing a two pore-forming P domain) and is highly expressed in the cerebellum. The synaptic protein encoding gene ADGRA1 (Pandya N J et al., Sci Rep. 2017; 7(1):12107), which is highly expressed in the frontal cortex. MYRIP, which encodes a scaffolding protein involved in exocytosis, and is expressed most highly in the amygdala, anterior cingulate cortex and cerebellum. The transcriptional repressor FOXN3, which is very highly expressed in the cerebellum. RAPGEF2, which is involved in cerebral cortex development and D1 Dopamine receptor-dependent ERK phosphorylation in brain (Jiang S Z et al., eNeuro. 2017; 4(5); Ye T et al., Nat Commun. 2014; 5:4826).

Discussion of this Example

The presently disclosed subject matter of this Example relates to a comprehensive quantitative genome-wide analysis of DNA methylation in human cerebrospinal fluid. The presently disclosed methods and the resulting data of this Example demonstrate that epigenomic liquid biopsy of the human central nervous system can be used for molecular phenotyping of brain DNA methylation signatures.

The presently disclosed subject matter of this Example reveals differences in the physical properties of cfDNA from CSF compared to that form plasma, with fragments of the former existing in a notably shorter state. This is reminiscent of cfDNA from the fetus during pregnancy which is known to be represented by fragments of lower molecular weight than maternal cfDNA (Yu S C et al., Proc Natl Acad Sci USA. 2014; 111(23):8583-8). The presently disclosed subject matter of this Example further identified clear periodicity in fragment size of cfDNA in CSF which suggests the presence of a nucleosome footprint that provides further information about the molecular phenotype of the CNS as has been suggested in the context of plasma cfDNA (Teo Y V et al., Aging Cell. 2019; 18(1):e12890; Ulz P et al., Nat Genet. 2016; 48(10):1273-8).

Further evidence that information relating to the molecular phenotype of the CNS may be obtained via epigenomic analysis of CSF cfDNA comes from the observation that DNA methylation signatures showed clear differences between cfDNA from CSF and plasma. Direct comparison of specific loci to identify differentially methylated CpG sites revealed that CSF-derived fragments were generally more likely to me less methylated than their plasma-derived counterparts. Furthermore, differentially methylated sites appeared to be correlated with tissue-specific gene expression patterns of the CNS.

Example Embodiment D reveals notable differences in the global distribution of DNA methylation between cfDNA from CSF and plasma, with fewer numbers of CpG sites existing in an intermediately methylated state (20-80%) in CSF versus plasma. This may reflect the fact that cfDNA in CSF is derived from relatively fewer distinct cell lineages than cfDNA in plasma which presumably consists of multiple contributions from a large and diverse range of different organ systems and cell-lineages. This is logical if one considers that, in its simplest form, a haploid CpG site in a single cell exists in either a state of complete hyper or hypo methylation. Thus, the % methylation or methylation rate at a given CpG site in a population of cfDNA fragments represents some complex combination of this binary state that is likely the result of the contributions of many millions or even billions of cells.

Although DNA methylation is, at some level, an indirect corollary to gene expression in the context of non-invasive molecular phenotyping, it has the distinct advantage of being relatively stable compared to cell-free RNA and thus represents a potentially attractive substrate for analysis. Sample stability means that any future clinical analysis of DNA methylation in cfDNA from CSF could be centralized in a specialized testing laboratory following sample transit.

One advantage of the presently disclosed method of this Example is its targeted nature. The solution-phase hybridization of ˜80 Mb of the human genome enables a systematic analysis of known structural genes and regulatory elements at a read depth that is higher relative to cost than could be achieved by whole genome shotgun bisulfite sequencing. Even though the yield of cfDNA per mL of CSF was lower than that of plasma, the recent emergence of new approaches to DNA methylation analysis at single base resolution will likely increase the efficiency of these assays and enable the generation of richer data sets from many more individual samples that reflect a range of pathobiological and normal phenotypes.

The presently disclosed subject matter of this Example relating to epigenomic liquid biopsy can be used for the molecular profiling of CNS phenotypes in a variety of research and clinical settings.

Example Embodiment E of Example 2: Estimation of Abnormal Spinal Fluid Methylome Variation in Targeted Regions for Diagnosis of CNS Abnormality

Provided below is an algorithm that can be used to diagnose a subject with a CNS disorder. The presently disclosed subject matter of this Example Embodiment E provides that the methylome(s) of the central nervous, or structures therein, could be affected by certain abnormalities, and that the changes of these methylomes can lead to changes in the methylation patterns of the DNA fragments found in maternal cerebrospinal fluid (CSF), which are released by CNS tissues. An algorithm was developed to identify the changes of methylation patterns in the methylome of CSF caused by CNS phenotypes. The main insight behind this algorithm of this Example was that the methylome of the DNA fragments in CSF is a mixture of a variety of component methylomes of CNS origin, and that the proportion of these different component methylomes in the mixture varies from subject to subject, even among the population with normal CNS phenotype. By constructing a model of CSF methylome as a linear combination of various component methylomes of CNS origins, the algorithm of this Example can accurately predict the methylation patterns of a new CSF sample under the hypothesis that it is from a normal individual. Consequently, the algorithm exhibited high sensitivity in detecting abnormal methylation patterns in a CSF sample caused by changes of the methylomes of some CNS tissues when the sample is from an affected individual.

Let i be any CpG site in human genome, z_i,j be the methylation level of CpG site i in a CSF sample j, p_i,r,j be the proportion of the r^th component methylome m_r,j of CNS origin in maternal plasma sample j at site i, m_i,r,j be the methylation level of CpG i in methylome m_r,j. The hypothesis is:

Z_i,j=Σ_r=1^Rp_i,r,jm_i,r,j  (1)

where p_i,r,j, m_i,r,j>=0, m_i,r,j<=1, p_i,1,j+ . . . +p_i,R,c=1.

It is further assumed that there is a set of CpG sites S such that, for any CpG site i in S, and any CSF j from a normal individual, it has m_I,r,j=m_I,r and p_I,r,j=p_r,j.

That is, it is assumed that in any CSF sample from a normal individual, the proportions of different component methylomes in the mixture are the same for all CpG sites in S. It is also assumed that, by restricting to the set of CpG sites S, CSF samples from all normal individuals have the same set of component methylomes. They are called restricted reference component methylomes (RRCM), and are labeled as m₁^S, . . . , m_R^S, or simply m₁, . . . , m_R when there is no confusion. For any CSF sample j from a normal individual, its methylome restricted to set of CpG sites in S can be expressed as a weighted average of the restricted reference component methylomes. More precisely, let z_j^S be the methylome of CSF sample C restricted to S, then for some mixture vector p_j=[p_j,1 . . . , p_j,R]^T, it has:

z_j^s=[m₁^S, . . . ,m_R^S]p_j  (2)

Finally, it is assumed that the set S is the union of two disjoint subsets C and T, where T is a union of K non-empty sets T_k such that T=U_k=1^KT_k where the index k represents the k^th type of abnormal CNS phenotype. T_k's do not need to be disjoint. Moreover, T_k itself is the union of two disjoint sets D_k and V_k. Either D_k or V_k could be empty, but not both. It is assumed that for any CSF sample, including one from an abnormal individual, when restricted to CpG sites in C, its methylome can always be expressed as a weighted average of the restricted reference component methylomes. That is, it has: z_j^C=[m₁^C, . . . , m_R^C]p_j regardless whether j is from an abnormal individual. C is called the set of reference CpG sites. On the other hand, for a CSF sample l from an abnormal individual, when restricted to CpG sites in S=C∪T, its methylome can no longer be expressed as a weighted average of the restricted reference component methylomes. That is, it has: w₁^S≠[m₁^S, . . . , m_R^S]p_l for any mixture vector p_l. More specifically, for a CSF sample l from the k^th type of abnormal individual, it has: 1), w_j^C=[m₁^C, . . . , m_R^C]p_l, 2), if D_K is non-empty, then w_lDK=[m_1,k^Dk, . . . , m_R,k^Dk]p_l such that [m₁^Dk, . . . , m_R^Dk]≠[m_1,k^Dk, . . . , m_R,k^Dk], and 3), if V_k is non-empty, then w_l^Vk=[m₁^Vk, . . . , m_R^Vk]q_l such that p_l≠q_l. In other words, in a CSF sample from the k^th type of abnormal individual, if the set D_k is not empty, the component methylomes of the sample l restricted to D_k are no longer the same as the reference component methylome restricted to D_k. If the set V_k is not empty, in this CSF sample, the proportion of the reference component methylomes restricted to V_k is no longer the same as the proportion of the reference component methylome restricted to R.

T is called the target set of CpG sites, D_k is called the differential methylation target set, V_k is called the copy number variation target set, and T_k is called the target set for the k^th type of abnormal individual.

The main steps of the presently disclosed algorithm are:

• • 1) Identify the sets of reference CpG sites C, and T₁, . . . , T_K for the list of K types of abnormal individuals.
  • 2) Estimate the restricted reference component methylomes m₁, . . . , m_R, or R predictor methylomes n₁, . . . , n_R that are independent linear combinations of the reference component methylomes such that n_r=[m₁, . . . , m_R]q_r for R linearly independent mixture vectors q₁, . . . , q_R.
  • 3) (Optional) If the reference component methylomes are available, estimate the proportions of these components at the reference CpG sites C for the test CSF samples.
  • 4) Predict the methylation level of the test CSF samples at the target set T_k of CpG sites, under the hypothesis that the sample is from a normal individual.
  • 5) Compare the predicted methylation levels at D_k and V_k against the observed methylation levels, and reject the null hypothesis that a test sample is from a normal individual if the observed methylation levels are significantly different form the predicted levels.

The presently disclosed algorithm of this Example can be implemented in a variety of ways. For example, given the methyl-seq data for a set of CSF samples from normal individuals, the presently disclosed EM algorithm or the data augmentation method can be applied to estimate the component methylomes, then use the maximum likelihood method to estimate the proportion of these component methylomes in the test sample. Below are exemplary simple implementations of the presently disclosed algorithm that use linear regression.

In the first simple implementation of the presently disclosed algorithm of this Example, it is assumed the restricted methylome of a CSF sample from a normal individual can be approximated by a mixture of two restricted reference methylomes, one representing the DNA fragments from a first specific CNS region, another representing the DNA fragments from a second specific CNS region. It is further assumed that the estimations of these two reference component methylomes are available. For example, in the implementation below, the methylome of oligodendrocytes is used as an approximation to the oligodendrocyte methylome in the CSF sample, and the methylome of neuronal cell samples from healthy individuals (HI) is used as an approximation to the neuronal methylome in the CSF sample. The implementation of the algorithm includes the following steps:

• • 1. Identify the reference set C, and the target sets T₁, . . . , T_K.
    • 1.1 Collect the methylation data for a set of oligodendrocyte samples, a set of neuronal cell samples, and a set of CSF samples, all from normal individuals. For each type of abnormal individuals, collect a set of oligodendrocyte samples, a set of neuronal cell samples, and a set of CSF samples from that type of abnormal individuals. All these samples should have matched age, race, and other relevant parameters. These are the training data.
    • 1.2 Let x_i,j be the observed methylation level of CpG site i in a normal oligodendrocyte sample j, and y_i,l the observed methylation level of CpG site i in a normal neuronal cell sample l, s_x,i² the sample variance of x_i,j over all normal oligodendrocyte samples, s_y,i² the sample variance of y_i,j over all normal neuronal cell samples. Identify the CpG sites S₀ such that for any i∈S₀, it has both s_x,i²<c₀ and s_y,i²<c₀ for some constant c₀. These are CpG sites with stable methylation levels in each type of normal cells.
    • 1.3 Let x_i,j be the observed methylation level of CpG site i in an oligodendrocyte sample j, including normal and abnormal, and y_i,l the observed methylation level of CpG site i in a neuronal cell sample l, including normal and abnormal, s_x,i² the sample variance of x_i,j over all oligodendrocyte samples, including normal and abnormal, s_y,i² the sample variance of y_i,j over all neuronal cell samples, including normal and abnormal. Identify the CpG sites S₁ such that for any i∈S₁, it has both s_x,i²<c₀ and s_y,i²<c₀ for some constant c₀, and that the statistical test for the difference between {x_i,j0: j0 is a normal oligodendrocyte sample} and {x_i,j2: jk is an abnormal oligodendrocyte sample of type k} is not significant for all abnormal types of oligodendrocyte, and that the statistical test for the difference between {y_i,j0: j0 is a normal neuronal cell sample} and {y_i,j2: jk is an abnormal neuronal cell sample of type k} is not significant for all abnormal types of neuronal cell. These are CpG sites with stable methylation levels in each type of cells, and with no difference in methylation level between normal and any abnormal samples. Let x_i be the sample mean of x_i,j over all oligodendrocyte samples, including normal and abnormal, y_i the sample mean of y_i,j over all neuronal cell samples, including normal and abnormal. Identify the subset C₀ of S₁ such that for any i∈C₀, it has |x_i−y_i|>c₁ for some constant c₁. These are CpG sites that are stably methylated in each cell type, with no difference between the normal and abnormal samples of the same cell type, and differentially methylated between different types of cells.
    • 1.4 Let x^R0 be the vector of x_i for all i∈C₀, and y^C0 be the vector of y_i for all i∈C₀, where x_i is the mean methylation at site i in all oligodendrocyte samples y_i the mean methylation at site i in all neuronal cell samples. Note that by the way the set C₀ is selected, there is no difference in the methylation level of any CpG sites in C₀ between normal and abnormal oligodendrocyte samples, or between normal and abnormal neuronal cell samples. Let z_j^C0 be the observed methylation levels of CpG sites in C₀ for a CSF sample j of the k^th abnormal type. (For convenience, the normal CSF sample is called as sample of the 0^th abnormal type). For each sample j belonging to the k^th abnormal type, regress z_j^C0 against x^C0 and y^C0, with the constraints that the intercept must be 0, and the coefficients must be non-negative and add to 1, and get the residual e_j^C0. Identify the subset C₀^k of C₀ such that for any CpG i in C₀^k, it has

$\frac{e_{i,k}^2}{s_{i,k}}<c_2,$

• • • and e_i,k²<c₃ for some constants c₂ and c₃, where e_i,k² is the mean of the squared difference between estimated and observed methylation levels of CpG site i in all CSF samples of the k^th abnormal type, and s_i,k² the sample variances of methylation levels of CpG site i in the same set of CSF samples. Repeat the above procedure for each type of abnormal CSF samples, the intersection of the subsets C=∩_k=0^K C₀^k is the reference set of J CpG sites. These are CpG sites where their methylation levels in both normal and any type of abnormal CSF samples can be accurately predicted by the reference component methylomes from normal individuals.
    • 1.5 Let T₀=S₀\S₁. Let x^C and x^T0 be the vectors of x_i and x_h for all i∈C and h∈T₀ respectively, and y^C and y^T0 be the vectors of y_i and y_h for all i∈C and h∈T₀ respectively, where x_i, x_h, y_i, and y_h are mean methylation level of sites for a normal oligodendrocyte or neuronal cell at sites i and h respectively. Let z_j^C and z_j^T0 and be the observed methylation levels of CpG sites in C and T₀ respectively for a normal CSF sample j, w_lk^C and w_lg^T0 the observed methylation level of CpG sites in C and T₀ respectively for a CSF sample l_k from an individual with the k^th type of abnormality, w_lg^C and w_lg^T0 the observed methylation level of CpG sites in C and T₀ respectively for a CSF sample l_g from an individual with the g^th type of abnormality, where g≠k. For each j, l_k, and l_g, regress z_j^C, w_lk^C, and w_lg^C respectively against x^C and y^C, with the constraints that the intercept must be 0, and the coefficients must be non-negative and add to 1. Apply the fitted models respectively to x^T0 and y^T0 to predict z_j^T0, w_lk^T0, and w_lg^T0 respectively, and get the differences e_j^T0, e_lk^T0 and e_lg^T0 between the predicted values and observed values. Let e_i, e_i,k, and e_i,g be the means of the sets of differences {e_jⁱ∈e_j^T0: j is a normal CSF sample}, {e_lk^T0: l_k is a CSF sample of th k^th abnormal type} and {e_lg^T0: l_g is a CSF sample of the g^th abnormal type} for CpG site i respectively. Identify the subset T_k of T₀ such that for any i∈T_k, it has |e_i|<c_2,0, |e_ik|>c_2,k, and |e_ik−e_ig|>c_3,k, for some constants c_2,0, c_2,k, and c_3,k, for all g≠k. T_k is the target set for the k^th type of the abnormal individual. These are the sites where the methylation of a normal CSF sample can be accurately predicted, the observed methylation in a CSF sample of the k^th abnormal type will deviate from the prediction, and deviation will be different from that of a CSF sample of any other abnormal type.
  • 2. Estimate fraction of the new CSF samples to be tested. Recall that x^c and y^c are mean vectors of the methylation levels of the training oligodendrocyte and training neuronal cell data for the CpG sites in the reference set C. For any new CSF sample t to be tested, let z_t^C be the observed methylation levels of CpG sites in C. Regress z_t^C against x^C and y^C, with the constraints that the intercept must be 0, and the coefficients must be non-negative and add to 1. The estimated coefficient for x^C is the estimated oligodendrocyte fraction for the CSF sample t.
  • 3. Test if the new CSF samples are from the k^th type of abnormal individual. For the new CSF sample t, let x^Tk and y^Tk be mean vectors of the methylation levels of the training oligodendrocyte and training neuronal cells data for the CpG sites in the target set T_k identified in step 1 of this algorithm, apply the fitted regression models obtained from the step 2 of this algorithm to X^Tk and y^Tk to predict the methylation levels of CpG sites in T_k for sample t under the hypothesis that sample t is from a normal pregnancy. Let n_k be the number of CpG sites in T_k. Define functions

$f_k\left(x_1,\dots ,x_{\mathrm{nk}}\right)={{\Sigma }_i\left(-1\right)}^{I\_\left(e_{i_k}-e_i\right)}{x}_i\mathrm{and}{f}_{k,g}\left(x_1,\dots ,x_{\mathrm{nk}}\right)={{\Sigma }_i\left(-1\right)}^{I\_\left(e_{i_k}-e_{i_g}\right)}{x}_i,$

• • where I_(⋅)=I_(−∞,0)(⋅), that is, the indicator function for the interval (−∞, 0), e_i, e_ik and e_ig are estimations obtained from step 1.5 of the algorithm. It will be said the sample is from the k^th type of abnormal individual if f_k(e_1t−e₁, . . . , e_nk,t−e_nk)>c_4,k, and f_k,g (e_1t−e_1g, . . . , e_nk,t−e_nk,g)>c_5,g for all g≠k, where e_it is the difference between the observed methylation level of the CpG site i∈T_k for sample t and the predicted value by the fitted model obtained from step 2, and g is any type of abnormal individual that is different form the k^th type of abnormal individual.

Other ways of implementing the presently disclosed algorithm of this Example can be developed by modifying the simple implementation presented above. Specifically, it does not need to assume that there are only two component reference methylomes that make up the CSF methylomes, nor does it need to approximate them by the oligodendrocyte and HI methylomes. Instead, a set of predictor methylomes can be collected that are mixtures of component reference genomes, as long as the number of the predictor methylomes is the same as the number of the reference component methylomes, and the mixture vectors of the predictor methylomes are linearly independent. For example, they can be methylomes of CSF samples with known different proportion of oligodendrocyte and neuronal cell DNAs.

In the presently disclosed algorithm of this Example, the difference between observed methylation levels in certain target regions and the predicted methylation levels as the test statistic to determine if in a CSF sample the methylome has been affect by some type of CNS abnormality. To illustrate the advantage of this approach, it is assumed that the mixture vector p_i for the methylome of a normal CSF sample j followed a Dirichlet's distribution with parameters α_i= . . . =α_R. Furthermore, for CpG site i, its methylation levels in the R reference vector p_j for component methylomes are m_i,r=(r−1)/(R−1). It can be shown that the methylation level of i in sample j then has a mean of 0.5, and a variance of

$\frac{R+1}{12\left(R-1\right)\left(R{\alpha }_1+1\right)}.$

If there is a methyl-seq library of in sample j with a coverage of N for CpG site i, the variance of the measured methylation level z_i,j is

${\sigma }_1^2=\frac{1}{4N}+\frac{N-1}{N}\frac{R+1}{12\left(R-1\right)\left(R{\alpha }_1+1\right)}.$

In other words, if z_i,j is used as a test statistic to detect abnormal CNS using CSF sample, under the null hypothesis, the test statistic has a variance of σ₁². However, in the presently disclosed algorithm of this Example, it is first estimated the mixture vector p_i, then predict z_i,j by Σ_rm_i,rp_r,j. Note that in a methyl-seq data, it can get millions of CpG sites covered in each library, and that the variance of the coefficients in a linear regression model is inversely proportional to sample size. Thus it is possible to get highly accurate estimation of the mixture vector p_i, even if it is taken into account that adjacent CpG sites tend to have correlated methylation levels. Assuming an accurate estimate of Σ_rm_i,rp_r,j can be obtained, the variance of the difference z_i,j−Σ_rm_i,rp_r,j between the observed methylation level and the prediction will be

$\frac{1}{4N}-\frac{1}{N}\frac{R+1}{12\left(R-1\right)\left(R{\alpha }_1+1\right)}.$

In other words, under the null hypothesis, the test static z_i,j−Σ_r M_i,r p_r,j used in the presently disclosed algorithm has a much smaller variance than the other candidate test statistic z_i,j. This in turns means that the presently disclosed test will achieve a higher power at the same level of type I error.

Although the presently disclosed subject matter of this Example and its advantages have been described in detail, it should be understood that various changes, substitutions and alterations can be made herein without departing from the spirit and scope of the invention. Moreover, the scope of this Example is not intended to be limited to the particular embodiments of the process, machine, manufacture, and composition of matter, means, methods and steps described in the specification. As one of ordinary skill in the art will readily appreciate from the invention of the presently disclosed subject matter, processes, machines, manufacture, compositions of matter, means, methods, or steps, presently existing or later to be developed that perform substantially the same function or achieve substantially the same result as the corresponding embodiments described herein may be utilized according to the presently disclosed subject matter. Accordingly, the appended claims are intended to include within their scope such processes, machines, manufacture, compositions of matter, means, methods, or steps.

Example of Embodiments for Example 2

B1. A method for diagnosing, prognosing, classifying and/or monitoring a CNS disorder in a subject comprising:

(a) obtaining a cerebrospinal fluid sample from the subject;

(b) determining the methylation status and/or level of one or more genomic loci in the cerebrospinal fluid sample;

(c) comparing the methylation status and/or level of the one or more genomic loci to a reference; and

(d) diagnosing the CNS disorder in the subject,

wherein the difference in the methylation status and/or level of the one or more genomic loci in the cerebrospinal fluid sample compared to the reference indicates the presence of the CNS disorder in the subject.

B2. The method of embodiment B1, wherein an increase in the level of methylation of the one or more genomic loci in the cerebrospinal fluid sample indicates the presence of the CNS disorder in the subject or a decrease in the level of methylation of the one or more genomic loci in the cerebrospinal fluid sample indicates the presence of the CNS disorder in the subject.
B3. The method of embodiment B1, wherein a decrease in the level of methylation of at least one of the one or more genomic loci in the cerebrospinal fluid sample and an increase in the level of methylation of at least one of the one or more genomic loci in the cerebrospinal fluid sample indicates the presence of the CNS disorder in the subject.
B4. A method of treating a CNS disorder in a subject comprising:

(a) obtaining a cerebrospinal fluid sample from the subject;

(b) determining the methylation status and/or level of one or more genomic loci present in the cerebrospinal fluid sample;

(c) comparing the methylation status and/or level of the one or more genomic loci to a reference;

(d) diagnosing a CNS disorder in the subject, wherein the difference in the methylation status and/or level of the one or more genomic loci in the cerebrospinal fluid sample compared to the reference indicates the presence of the CNS disorder in the subject; and

(e) treating the subject diagnosed with the CNS disorder.
B5. The method of any one of embodiments B1-B4, wherein the reference is the methylation status and/or level of the one or more genomic loci in a cerebrospinal fluid sample obtained from a subject that does not have the CNS disorder.
B6. A method of treating a CNS disorder in a subject comprising:

(a) measuring the methylation status and/or level of one or more genomic loci present in a cerebrospinal fluid sample from the subject prior to a treatment of the CNS disorder;

(b) measuring the methylation status and/or level of one or more genomic loci present in a cerebrospinal fluid sample from the subject during the treatment of the CNS disorder;

(c) continuing the treatment if the difference in the methylation status and/or level of the one or more genomic loci between the cerebrospinal fluid samples from prior to and during the treatment of the CNS disorder indicates the subject is responsive to the treatment.

B7. The method of embodiment B6, further comprising (d) administering a different treatment to the subject if the difference in the methylation status and/or level of the one or more genomic loci between the cerebrospinal fluid samples from prior to and during the treatment of the CNS disorder indicates the subject is not responsive to the treatment.
B8. The method of embodiment B6, wherein an increase in the level of methylation of the one or more genomic loci in the cerebrospinal fluid sample indicates the subject is responsive to the treatment, or a decrease in the level of methylation of the one or more genomic loci in the cerebrospinal fluid sample indicates the subject is responsive to the treatment.
B9. The method of embodiment B6, wherein a decrease in the level of methylation of at least one of the one or more genomic loci in the cerebrospinal fluid sample and an increase in the level of methylation of at least one of the one or more genomic loci in the cerebrospinal fluid sample indicates the subject is responsive to the treatment.
B10. The method of embodiment B7, wherein an increase in the level of methylation of the one or more genomic loci in the cerebrospinal fluid sample indicates the subject is responsive to the treatment, or a decrease in the level of methylation of the one or more genomic loci in the cerebrospinal fluid sample indicates the subject is not responsive to the treatment.
B11. The method of embodiment B7, wherein a decrease in the level of methylation of at least one of the one or more genomic loci in the cerebrospinal fluid sample and an increase in the level of methylation of at least one of the one or more genomic loci in the cerebrospinal fluid sample indicates the subject is not responsive to the treatment.
B12. The method of any one of embodiments B1-B11, wherein the one or more genomic loci comprise one or more CpG sites.
B13. The method of any one of embodiments B1-B12, wherein the CNS disorder is selected from the group consisting of brain and spinal cord tumors, brain and spinal cord infections, brain and spinal cord inflammations, neuropsychiatric disorders, and neurodegenerative diseases.
B14. The method of any one of embodiments B1-B13, wherein the one or more genomic loci are present within nucleic acids isolated from the cerebrospinal fluid sample.
B15. The method of any one of embodiments B1-B14, wherein the one or more genomic loci are present within cell-free nucleic acids isolated from the cerebrospinal fluid sample.
B16. A method of treating a CNS disorder in a subject comprising;

(a) diagnosing a CNS disorder in the subject by utilization of the algorithm disclosed in Example Embodiment E; and

(b) treating the subject diagnosed with the CNS disorder.

B17. The method of any one of embodiments B1-B16, wherein the subject is human.
B18. A kit for diagnosing, prognosing and/or monitoring a CNS disorder in a subject comprising a means for determining and/or detecting the methylation status of one or more genomic loci.
B19. The kit of embodiment B18, wherein the means comprises one or more primers and/or probes for determining and/or detecting the methylation status of the one or more genomic loci.

Abstract of this Example (Example 2)

Example 2 provides methods for diagnosing, prognosing, monitoring, classifying and/or treating central nervous system disorders, e.g., brain and spinal cord tumors, brain and spinal cord infections, brain and spinal cord inflammations, neuropsychiatric disorders, and neurodegenerative diseases. This Example further provides algorithms and kits for diagnosing, prognosing, monitoring, classifying and/or treating central nervous system disorders.

Example 3—Method for Non-Invasive Detection of Fetal Aneuploidy and/or Sub-Chromosomal Fetal Copy Number Variations by Bisulfite Sequencing of Maternal Plasma

Field of this Example

The methods disclosed in this Example are related to the field of prenatal diagnosis, specifically to non-invasive methods to detect fetal aneuploidy in a biological sample including maternal plasma DNA.

Background of this Example

Definitive prenatal diagnosis is currently performed via amniocentesis (AF) or chorionic villus sampling (CVS) to obtain fetal or placental cells, respectively. Chromosome analysis is then achieved using conventional karyotyping or, more recently, array comparative genomic hybridization (aCGH). AF and CVS are invasive procedures that have significant risk of miscarriage, fetal morbidity and considerable parental stress and there have been intense efforts to develop non-invasive alternatives.

Summary of this Example

Disclosed in this Example are diagnostic methods that can be used to detect fetal aneuploidy and/or sub-chromosomal fetal copy number variations (e.g., microdeletions and/or microduplications) in maternal plasma while reducing cost and complexity. The method uses DNA methylation signatures identified using biochemical methods with a computational approach to detect fetal aneuploidy and/or sub-chromosomal fetal copy number variations.

Description of this Example

Proof of concept has been demonstrated for the detection of pregnancy related disease via bisulfite sequencing of maternal plasma. One intermediate step of the method includes the estimation of a percentage of DNA fragments in the maternal plasma that originate from the fetus. The method has been modified to detect fetal aneuploidy and/or sub-chromosomal fetal copy number variations.

The approach of this Example involves the targeted methylation of specific regions of genomic DNA such as commonly aneuploid chromosomes and/or copy number variation regions (such as microdeletion regions). Such an approach would involve a parallel strategy in which regions of interest are targeted in a multiplex fashion.

Terms of this Example

Unless otherwise noted, technical terms are used according to conventional usage. Definitions of common terms in molecular biology may be found in Benjamin Lewin, Genes V, published by Oxford University Press, 1994 (ISBN 0-19-854287-9); Kendrew et al. (eds.), The Encyclopedia of Molecular Biology, published by Blackwell Science Ltd., 1994 (ISBN 0-632-02182-9); and Robert A. Meyers (ed.), Molecular Biology and Biotechnology: a Comprehensive Desk Reference, published by VCH Publishers, Inc., 1995 (ISBN 1-56081-569-8).

In order to facilitate review of the various embodiments of this Example, the following explanations of specific terms are provided:

Amplification: To increase the number of copies of a nucleic acid molecule. The resulting amplification products are called “amplicons.” Amplification of a nucleic acid molecule (such as a DNA or RNA molecule) refers to use of a technique that increases the number of copies of a nucleic acid molecule in a sample. An example of amplification is the polymerase chain reaction (PCR), in which a sample is contacted with a pair of oligonucleotide primers under conditions that allow for the hybridization of the primers to a nucleic acid template in the sample. The primers are extended under suitable conditions, dissociated from the template, re-annealed, extended, and dissociated to amplify the number of copies of the nucleic acid. This cycle can be repeated. The product of amplification can be characterized by such techniques as electrophoresis, restriction endonuclease cleavage patterns, oligonucleotide hybridization or ligation, and/or nucleic acid sequencing.

Other examples of in vitro amplification techniques include quantitative real-time PCR; reverse transcriptase PCR (RT-PCR); real-time PCR (rt PCR); real-time reverse transcriptase PCR (rt RT-PCR); nested PCR; strand displacement amplification (see U.S. Pat. No. 5,744,311); transcription-free isothermal amplification (see U.S. Pat. No. 6,033,881); repair chain reaction amplification (see PCT Publication No. WO 90/01069); ligase chain reaction amplification (see European patent publication No. EP-A-320 308); gap filling ligase chain reaction amplification (see U.S. Pat. No. 5,427,930); coupled ligase detection and PCR (see U.S. Pat. No. 6,027,889); and NASBA™ RNA transcription-free amplification (see U.S. Pat. No. 6,025,134), amongst others.

Allele (Haplotype): A 5′ to 3′ sequence of nucleotides found at a set of one or more polymorphic sites in a locus on a single chromosome from a single individual. “Allelic pair” is the two alleles found for a locus in a single individual. With regard to a population, alleles are the ordered, linear combination of polymorphisms (e.g., single nucleotide polymorphisms (SNPs) in the sequence of each form of a gene (on individual chromosomes) that exist in the population. “Haplotyping” is a process for determining one or more alleles in an individual and includes use of family pedigrees, molecular techniques and/or statistical inference. “Haplotype data” or “allele data” is the information concerning one or more of the following for a specific gene: a listing of the allelic pairs in an individual or in each individual in a population; a listing of the different alleles in a population; frequency of each allele in that or other populations, and any known associations between one or more alleles and a trait.

Bisulfite: All types of bisulfites, such as sodium bisulfite, that are capable of chemically converting a cytosine (C) to a uracil (U) without chemically modifying a methylated cytosine and therefore can be used to differentially modify a DNA sequence based on the methylation status of the DNA.

Bisulfite treatment: The treatment of DNA with bisulfite or a salt thereof, such as sodium bisulfite (NaHSO₃). Bisulfite reacts readily with the 5,6-double bond of cytosine, but poorly with methylated cytosine. Cytosine reacts with the bisulfite ion to form a sulfonated cytosine reaction intermediate which is susceptible to deamination, giving rise to a sulfonated uracil. The sulfonate group can be removed under alkaline conditions, resulting in the formation of uracil. Uracil is recognized as a thymine by polymerases and amplification will result in an adenine-thymine base pair instead of a cytosine-guanine base pair.

Cell-free DNA: DNA which is no longer fully contained within an intact cell, for example DNA found in plasma or serum.

Chromosomal abnormality: A chromosome, or a segment of a chromosome, with DNA deletions or duplications, such as chromosomal aneuploidy. The term also encompasses translocation of extra chromosomal sequences to other chromosomes.

Chromosomal aneuploidy or aneuploidy: The abnormal presence (hyperploidy) or absence (hypoploidy) of a chromosome, such as chromosome 13, 18 or 21. In some cases, the abnormality can involve more than one chromosome, or more than one portion of one or more chromosomes. The most common chromosome aneuploidy is trisomy, such as trisomy 21, where the genome of an afflicted patient has three chromosomes 21, as compared to two chromosomes 21. In rarer cases, the patient may have an extra piece of chromosome 21 (less than full length) in addition to the normal pair. In yet other cases, a portion of chromosome 21 may be translocated to another chromosome, such as chromosome 14. In this example, chromosome 21 is referred as the “chromosome relevant to the chromosomal aneuploidy” and a second, chromosome that is present in the normal pair in the patient's genome, for example chromosome 1, is a “reference chromosome.” There are also cases where the number of a relevant chromosome is less than the normal number of 2. Turner syndrome is one example of a chromosomal aneuploidy where the number of X chromosome in a female subject has been reduced from two to one.

Chromosomal Copy number variation: A microdeletion is a small deletion in a chromosome, such as a deletion of about 10 kilobases to about 5 million base pairs, for example a deletion of about 100 to about 5 million base pairs, such as about 1,000 kilobases, such as about 100 kilobases, 200 kilobases, 400 kilobases, 500 kilobases, 750 kilobases, or about 1 million base pairs. Similarly, a “microduplication” is a small duplication in a chromosome, such as a duplication of about 10 kilobases to about 5 million base pairs, for example a deletion of about 100 to about 5 million base pairs, such as about 1,000 kilobases, such as about 100 kilobases, 200 kilobases, 400 kilobases, 500 kilobases, 750 kilobases, or about 1 million base pairs. A chromosomal fetal copy number variation is either a microdeletion or a microduplication in genomic DNA of a fetus of a pregnant woman.

CpG-containing genomic sequence: A segment of DNA sequence at a defined location in the genome of an individual such as a human fetus or a pregnant woman. Typically, a “CpG-containing genomic sequence” is at least 15 nucleotides in length and contains at least one cytosine. In some embodiments, a CpG containing sequence can be at least 30, 50, 80, 100, 150, 200, 250, or 300 nucleotides in length and contains at least 2, 5, 10, 15, 20, 25, or 30 cytosines. For any specific “CpG-containing genomic sequence” at a given location, for example, within a region centering around a given genetic locus on chromosome 21 nucleotide sequence variations can exist from individual to individual and from allele to allele even for the same individual. Typically, such a region centering around a defined genetic locus (e.g., a CpG island) contains the locus as well as upstream and/or downstream sequences. Each of the upstream or downstream sequence (counting from the 5′ or 3′ boundary of the genetic locus, respectively) can be as long as 1 kb, in other cases may be as long as 5 kb, 2 kb, 750 bp, 500 bp, 200 bp, or 100 bp. A “CpG-containing genomic sequence” can encompass a coding or a non-coding, nucleic acid sequence, and thus can include a nucleotide sequence transcribed (or not transcribed) for protein production. Thus, a CpG containing genomic sequence can be a nucleotide sequence can be a protein-coding sequence, a non protein-coding sequence or a combination thereof.

Control DNA: Genomic DNA obtained from an individual that is used for comparative purposes, such as DNA from a healthy individual who does not have a chromosomal abnormality. In some embodiments, a control DNA sample can be obtained from plasma of a female carrying a healthy fetus who does not have a chromosomal abnormality, which can serve as a negative control. When certain chromosome anomalies are known, the control can also be established standards that are indicative of a specific disease or condition.

To screen for three different chromosomal aneuploidies in a maternal plasma of a pregnant female, a panel of control DNAs that have been isolated from plasma of mothers who are known to carry a fetus with, for example, chromosome 13, 18, or 21 trisomy, and a mother who is pregnant with a fetus who does not have a chromosomal abnormality can be used as a control.

Copy number: The number of copies of a section of DNA in a genome. Copy number analysis usually refers to the process of analyzing data produced by a test for DNA copy number variation in patient's sample. Such analysis helps detect chromosomal copy number variation that may cause or may increase risks of various critical disorders. Copy number variation can be detected with various types of tests, including, but not limited to, such as methylation status, fluorescent in situ hybridization, comparative genomic hybridization high-resolution array-based tests based on array comparative genomic and SNP array technologies. The methods disclosed herein can be used to determine the copy number of a specific locus of interest.

DNA (deoxyribonucleic acid): DNA is a long chain polymer which comprises the genetic material of most living organisms (some viruses have genes comprising ribonucleic acid (RNA)). The repeating units in DNA polymers are four different nucleotides, each of which comprises one of the four bases, adenine, guanine, cytosine and thymine bound to a deoxyribose sugar to which a phosphate group is attached. Triplets of nucleotides (referred to as codons) code for each amino acid in a polypeptide, or for a stop signal (termination codon). The term codon is also used for the corresponding (and complementary) sequences of three nucleotides in the mRNA into which the DNA sequence is transcribed.

Unless otherwise specified, any reference to a DNA molecule is intended to include the reverse complement of that DNA molecule. Except where single-strandedness is required by the text herein, DNA molecules, though written to depict only a single strand, encompass both strands of a double-stranded DNA molecule. Thus, a reference to the nucleic acid molecule that encodes a protein, or a fragment thereof, encompasses both the sense strand and its reverse complement. Thus, for instance, it is appropriate to generate probes or primers from the reverse complement sequence of the disclosed nucleic acid molecules.

DNA sequencing: The process of determining the nucleotide order of a given DNA molecule. The general characteristics of “parallel or massively parallel sequencing” are that the sequence of the target genetic material is then performed in parallel and the sequence information is captured by a computer. For example, the sequencing can be performed using sequencing-by-synthesis with reversible terminations (ILLUMINA® Genome Analyzer or Hi Seq, Next-Seq, Nova-Seq etc., semiconductor sequencing (Thermo FishernIon Torrent) or nanopore sequencing (Oxford Nanopore).

Differentially Modifies (methylated or non-methylated DNA): A reagent that modifies methylated or non-methylated DNA, respectively, in a process through which distinguishable products result from methylated and non-methylated DNA, thereby allowing the identification of the DNA methylation status. Such processes may include, but are not limited to, chemical reactions (such as conversion by bisulfite) and enzymatic treatment (such as cleavage by a methylation-dependent endonuclease), or an antibody that specifically binds a methylated (or non-methylated) DNA sequence. Thus, an enzyme that preferentially cleaves or digests methylated DNA is one capable of cleaving or digesting a DNA molecule at a significantly higher efficiency when the DNA is methylated, whereas an enzyme that preferentially cleaves or digests unmethylated DNA exhibits a significantly higher efficiency when the DNA is not methylated.

Gene: A segment of DNA that contains the coding sequence for a protein, wherein the segment may include promoters, exons, introns, and other untranslated regions that control expression.

Genotype: An unphased 5′ to 3′ sequence of nucleotide pair(s) found at a set of one or more polymorphic sites in a locus on a pair of homologous chromosomes in an individual. “Genotyping” is a process for determining a genotype of an individual.

Genomic segment: A contiguous sequence of genomic DNA no more than 2000 bases in length.

Hybridization: Oligonucleotides and their analogs hybridize by hydrogen bonding, which includes Watson-Crick, Hoogsteen or reversed Hoogsteen hydrogen bonding, between complementary bases. Generally, nucleic acids consist of nitrogenous bases that are either pyrimidines (cytosine (C), uracil (U), and thymine (T)) or purines (adenine (A) and guanine (G)). These nitrogenous bases form hydrogen bonds between a pyrimidine and a purine, and the bonding of the pyrimidine to the purine is referred to as “base pairing.” More specifically, A will hydrogen bond to T or U, and G will bond to C. “Complementary” refers to the base pairing that occurs between two distinct nucleic acid sequences or two distinct regions of the same nucleic acid sequence. For example, an oligonucleotide can be complementary to a specific genetic locus, so it specifically hybridizes with a mutant allele (and not the reference allele) or so that it specifically hybridizes with a reference allele (and not the mutant allele).

“Specifically hybridizable” and “specifically complementary” are terms that indicate a sufficient degree of complementarity such that stable and specific binding occurs between the oligonucleotide (or its analog) and the DNA or RNA target, such that the target can be distinguished. The oligonucleotide or oligonucleotide analog need not be 100% complementary to its target sequence to be specifically hybridizable. An oligonucleotide or analog is specifically hybridizable when binding of the oligonucleotide or analog to the target DNA or RNA molecule interferes with the normal function of the target DNA or RNA, and there is a sufficient degree of complementarity to avoid non-specific binding of the oligonucleotide or analog to non-target sequences under conditions where specific binding is desired, for example under physiological conditions in the case of in vivo assays or systems. Such binding is referred to as specific hybridization. In one example, an oligonucleotide is specifically hybridizable to DNA or RNA nucleic acid sequences including an allele of a gene, wherein it will not hybridize to nucleic acid sequences containing a polymorphism.

Hybridization conditions resulting in particular degrees of stringency will vary depending upon the nature of the hybridization method of choice and the composition and length of the hybridizing nucleic acid sequences. Generally, the temperature of hybridization and the ionic strength (especially the Na⁺ concentration) of the hybridization buffer will determine the stringency of hybridization, though wash times Also influence stringency. Calculations regarding hybridization conditions required for attaining particular degrees of stringency are discussed by Sambrook et al. (ed.), Molecular Cloning: A Laboratory Manual, 2nd ed., vol. 1-3, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y., 1989, chapters 9 and 11.

Increase or a Decrease: A significantly significant positive or negative change, respectively, in quantity from a control value. An increase is a positive change, such as a 50%, 100%, 200%, 300%, 400% or 500% increase as compared to the control value. A decrease is a negative change, such as a 50%, 100%, 200%, 300%, 400% or 500% decrease as compared to a control value.

Isolated: An “isolated” biological component (such as a nucleic acid molecule, protein or organelle) has been substantially separated or purified away from other biological components in the cell of the organism in which the component naturally occurs, i.e., other chromosomal and extra-chromosomal DNA and RNA, proteins and organelles. Nucleic acids and proteins that have been “isolated” include nucleic acids and proteins purified by standard purification methods. The term also embraces nucleic acids and proteins prepared by recombinant expression in a host cell as well as chemically synthesized nucleic acids.

Locus: A location on a chromosome or DNA molecule corresponding to a gene or a physical or phenotypic feature, where physical features include polymorphic sites.

Methylation: The addition of a methyl group (—CH₃) to cytosine nucleotides of CpG sites in DNA. DNA methylation, the addition of a methyl group onto a nucleotide, is a post-replicative covalent modification of DNA that is catalyzed by a DNA methyltransferase enzyme. In biological systems, DNA methylation can serve as a mechanism for changing the structure of DNA without altering its coding function or its sequence.

Methylation sequencing assay: A sequencing assay that detects the methylation status of one or more CpG sites in DNA. A non-limiting example of a methylation sequencing assay is a sequencing assay performed on bisulfite-treated and amplified genomic DNA.

Methylation status: The state of methylation of a genomic sequence. This refers to the characteristics of a DNA segment at a particular genomic locus relevant to methylation. Such characteristics include, but are not limited to, whether any of the cytosine (C) residues within this DNA sequence are methylated, location of methylated C residue(s), percentage of methylated C at any particular stretch of residues, and allelic differences in methylation. The methylation profile affects the relative or absolute concentration of methylated C or unmethylated C at any particular stretch of residues in a biological sample.

Methyl-sensitive enzymes: DNA restriction endonucleases that are dependent on the methylation state of their DNA recognition site for activity. For example, there are methyl-sensitive enzymes that cleave at their DNA recognition sequence only if it is not methylated. Thus, an unmethylated DNA sample will be cut into smaller fragments than a methylated DNA sample. Similarly, a hypermethylated DNA sample will not be cleaved. In contrast, there are methyl-sensitive enzymes that cleave at their DNA recognition sequence only if it is methylated. As used herein, the terms “cleave”, “cut” and “digest” are used interchangeably.

Methyl-sensitive enzymes that digest unmethylated DNA suitable for use in methods of the invention include, but are not limited to, HpaII, HhaI, MaeII, BstUI and AciI. One enzyme is HpaII that cuts only the unmethylated sequence CCGG. Enzymes that digest only methylated DNA include, but are not limited to, DpnI, which cuts at a recognition sequence GATC, and McrBC, which belongs to the family of AAA⁺ proteins (New England BioLabs, Inc., Beverly, Mass.).

Cleavage methods and procedures for selected restriction enzymes for cutting DNA at specific sites are well known to the skilled artisan. For example, many suppliers of restriction enzymes provide information on conditions and types of DNA sequences cut by specific restriction enzymes, including New England BioLabs, Promega Corporation, Boehringer-Mannheim, and the like. Sambrook et al. (See Sambrook et al., Molecular Biology: A Laboratory Approach, Cold Spring Harbor, N.Y. 1989) provide a general description of methods for using restriction enzymes and other enzymes.

Oligonucleotide: An oligonucleotide is a plurality of joined nucleotides joined by native phosphodiester bonds, between about 6 and about 300 nucleotides in length. An oligonucleotide analog refers to moieties that function similarly to oligonucleotides but have non-naturally occurring portions. For example, oligonucleotide analogs can contain non-naturally occurring portions, such as altered sugar moieties or inter-sugar linkages, such as a phosphorothioate oligodeoxynucleotide. Functional analogs of naturally occurring polynucleotides can bind to RNA or DNA, and include peptide nucleic acid (PNA) molecules.

In several examples, oligonucleotides and oligonucleotide analogs can include linear sequences up to about 200 nucleotides in length, for example a sequence (such as DNA or RNA) that is at least 6 bases, for example at least 8, 10, 15, 20, 25, 30, 35, 40, 45, 50, 100 or even 200 bases long, or from about 6 to about 70 bases, for example about 10-25 bases, such as 12, 15 or 20 bases.

Polymorphic marker: A segment of genomic DNA that exhibits heritable variation in a DNA sequence between individuals. Such markers include, but are not limited to, single nucleotide polymorphisms (SNPs), restriction fragment length polymorphisms (RFLPs), short tandem repeats, such as di-, tri- or tetra-nucleotide repeats (STRs), and the like. Polymorphic markers can be used to specifically differentiate between a maternal and paternal allele in the enriched fetal nucleic acid sample.

Polymorphism: A variation in a gene sequence. The polymorphisms can be those variations (DNA sequence differences) which are generally found between individuals or different ethnic groups and geographic locations which, while having a different sequence, produce functionally equivalent gene products. Typically, the term can also refer to variants in the sequence which can lead to gene products that are not functionally equivalent. Polymorphisms also encompass variations which can be classified as alleles and/or mutations which can produce gene products which may have an altered function. Polymorphisms also encompass variations which can be classified as alleles and/or mutations which either produce no gene product or an inactive gene product or an active gene product produced at an abnormal rate or in an inappropriate tissue or in response to an inappropriate stimulus. Alleles are the alternate forms that occur at the polymorphism.

Polymorphisms can be referred to, for instance, by the nucleotide position at which the variation exists, by the change in amino acid sequence caused by the nucleotide variation, or by a change in some other characteristic of the nucleic acid molecule or protein that is linked to the variation.

Primers: Primers are nucleic acid molecules, usually DNA oligonucleotides of about 10-50 nucleotides in length (longer lengths are also possible). Typically, primers are at least about 15 nucleotides in length, such as at least about 20, 25, 30, or 40 nucleotides in length. For example, a primer can be about 10-50 nucleotides in length, such as, 10-30, 15-20, 15-25, 15-30, or 20-30 nucleotides in length. Primers can also be of a maximum length, for example no more than 25, 30, 40, or 50 nucleotides in length. Forward and reverse primers may be annealed to a complementary target DNA strand by nucleic acid hybridization to form hybrids between the primers and the target DNA strand, and then extended along the target DNA strand by a DNA polymerase enzyme to form an amplicon. One of skill in the art will appreciate that the hybridization specificity of a particular probe or primer typically increases with its length. Thus, for example, a probe or primer including 20 consecutive nucleotides typically will anneal to a target with a higher specificity than a corresponding probe or primer of only 15 nucleotides. In some embodiments, forward and reverse primers are used in combination in a bisulfite amplicon sequencing assay.

Sample: A sample, such as a biological sample, is a sample obtained from a subject. As used herein, biological samples include all clinical samples useful for detection of fetal aneuploidy, including, but not limited to, cells, tissues, and bodily fluids, such as: blood; derivatives and fractions of blood, such as serum; urine; sputum; or CVS samples. In a particular example, a sample includes blood obtained from a human subject, such as whole blood or serum. Microdeletions and microduplications can be measured in samples isolated from a subject.

Sensitivity and specificity: Statistical measurements of the performance of a binary classification test. Sensitivity measures the proportion of actual positives which are correctly. Specificity measures the proportion of negatives which are correctly identified.

Sequence Read: A sequence (e.g., of about 300 bp) of contiguous base pairs of a nucleic acid molecule. The sequence read may be represented symbolically by the base pair sequence (in ATCG) of the sample portion. It may be stored in a memory device and processed as appropriate to determine whether it matches a reference sequence or meets other criteria. A sequence read may be obtained directly from a sequencing apparatus or indirectly from stored sequence information concerning a sample.

Standard control: A value reflective of the ratio, or the amount or concentration of a fetal genomic sequence located on a chromosome relevant to a particular chromosomal aneuploidy (such as trisomy 13, 18, or 21) over the amount or concentration of a fetal genetic marker located on a reference chromosome, as the amounts or concentrations are found in a biological sample (for example, blood, plasma, or serum) from an average, healthy pregnant woman carrying a chromosomally normal fetus. A “standard control” can be determined differently and represent different value depending on the context in which it is used. For instance, when used in an epigenetic-genetic dosage method where an epigenetic marker is measured against a genetic marker, the “standard control” is a value reflective of the ratio, or the amount or concentration of a fetal genomic sequence located on a chromosome relevant to a particular chromosomal aneuploidy (for example, trisomy 13, 18, or 21) over the amount or concentration of a fetal genetic marker located on a reference chromosome, as the amounts or concentrations are found in a biological sample (such as blood, plasma, or serum) from an average, healthy pregnant woman carrying a chromosomally normal fetus. In some embodiments, a standard control is determined based on an average healthy pregnant woman at a certain gestational age.

Subject: Living multi-cellular vertebrate organisms, a category that includes human and non-human mammals (such as laboratory or veterinary subjects).

Target sequence or region: A sequence of nucleotides located in a particular region in the human genome. The target can be for instance a coding sequence; it can also be the non-coding strand that corresponds to a coding sequence. The target can also be a non-coding sequence, such as an intronic sequence.

Unless otherwise explained, all technical and scientific terms used in this Example have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs. The singular terms “a,” “an,” and “the” include plural referents unless context clearly indicates otherwise. Similarly, the word “or” is intended to include “and” unless the context clearly indicates otherwise. It is further to be understood that all base sizes or amino acid sizes, and all molecular weight or molecular mass values, given for nucleic acids or polypeptides are approximate, and are provided for description. Although methods and materials similar or equivalent to those described herein can be used in the practice or testing of this disclosure, suitable methods and materials are described below. The term “comprises” means “includes.”

This Example provides an improved method for the detection of fetal aneuploidy and/or the detection of sub-chromosomal fetal copy number variations via bisulfate sequencing of maternal plasma. One intermediate step of the method is to estimate the percentage of DNA fragments in the maternal plasma that originate from the fetus. In some embodiments, the method for the detection of fetal aneuploidy includes the following steps.

First, the method comprises estimating the percentage of fetal DNA fragments in the sample using exclusively the reads aligned to the target chromosome for which we would like to test fetal aneuploidy, e.g., chromosome 21. Herein, this estimation is called F_A.

Next, the method comprises estimating the percentage of fetal DNA fragment in the sample using the reads aligned to the reference chromosomes that we believe are not affected by fetal aneuploidy, e.g., chromosome 1, 2, 3, etc. Herein, this estimation is called F_D.

The next step comprises calculating the difference between F_A−F_D.

If |F_A−F_D|>c₁, for some threshold c₁>0, we claim that the fetus has aneuploidy in the target chromosome. Specifically, if F_A−F_D>c₁, it is indicated that the fetus has one more extra copy in the target chromosome. If F_A−F_D<−c₁, it is indicated that the fetus has only one copy in the target chromosome.

If |F_A−F_D|>c₂, for some threshold c₂ such that c₁>c₂>0, and F_D<c₃ for threshold c₃>0, it is undecided whether the fetus has aneuploidy in the target chromosome.

If |F_A−F_D|<c₂, it is undecided whether the fetus is not affected by aneuploidy in the target chromosome.

In some embodiments, the method for the detection of sub-chromosomal fetal copy number variations includes the following steps:

First, the method comprises estimating the percentage of fetal DNA fragments in the sample using exclusively the reads aligned to a target genomic copy number variation region. The target copy number variation region may be on a target chromosome, or any chromosome. Herein, this estimation is called F_AM.

Next, the method comprises estimating the percentage of fetal DNA fragment in the sample using the reads aligned to a reference or control copy number variation region that is known to not be significantly affected by fetal copy number variations. The reference copy number variation region may be on a reference chromosome or any chromosome. Herein, this estimation is called F_DM.

The next step comprises calculating the difference between F_AM−F_DM.

If |F_AM−F_DM|>c₁, for some threshold c₁>0, we claim that the sample has the presence of sub-chromosomal fetal microdeletions. Specifically, if F_AM−F_DM>c₁, it is indicated that the fetal DNA in the sample has sub-chromosomal copy number variations. If F_AM−F_DM<−c₁, it is indicated that the fetal DNA in the sample does not have sub-chromosomal copy number variations.

If |F_AM−F_DM|>c₂, for some threshold c₂ such that c₁>c₂>0, and F_D<c₃ for constant c₃>0, it is undecided whether the fetus has sub-chromosomal copy number variations.

If |F_A−F_D|<c₂, it is undecided whether the fetus is not affected by sub-chromosomal copy number variations.

Example Embodiment F of Example 3: Detection of Trisomy 21 by Bisulfite Sequencing of Maternal Plasma

Sequencing Datasets Used were as Follows:

CM dataset: CVS and MBC genome-wide targeted bisulfite sequencing data: We have 5 bisulfite sequencing libraries from 5 normal CVS samples and 6 bisulfite sequencing libraries from 6 normal MBC samples. Genome-wide targeting was achieved via solution-phase hybridization using Agilent kit. There are about 2.8 million CpG sites (ranging from 1.2 to 3.5 million) covered by 5 or more reads in each library. In total we have 2.75 million CpG sites that are covered by 5 or more reads in at least 3 libraries in each group.

PnP dataset: Normal Pregnant and Non pregnant plasma targeted bisulfite sequencing data: We have 6 bisulfite sequencing libraries from 6 normal maternal plasma samples and 9 bisulfite sequencing libraries from 9 normal nonpregnant woman plasma samples. Genome-wide targeting was achieved via solution-phase hybridization using the Roche kit. There are about 7.3 million CpG sites (ranging from 5.3 to 8.8 million) covered by 5 or more reads in each library. We have in total 6.84 million CpG sites that are covered by 5 or more reads in at least 4 libraries in each group.

T21N dataset: Maternal plasma (Normal and Trisomy 21) targeted bisulfite sequencing data included 14 libraries from 12 normal samples, and 12 libraries from 12 T21 samples. Genome-wide targeting was achieved via solution-phase hybridization using the Roche kit. In each library, there are about 7.7 million CpG sites covered by 5 or more reads (ranging from 4.86 million to 9.95 million). We have in total about 6.86 million CpG sites that are covered by 5 or more reads in at least 8 libraries in each group.

Model selection: The method first comprises determining the most appropriate predictive model for analyzing the methylation levels of libraries in the T21N data set.

Maternal plasma DNA can be considered as a mixture of the DNA from maternal tissues and fetal tissues. If we assume that among the fetal tissues, placenta contributes the most to the maternal plasma DNA, and among the maternal tissues, maternal white blood cells contribute the most to the maternal plasma DNA, we can approximate the fetal tissue signature using placental reference data, and the maternal tissue signature using maternal white blood cell reference data. This suggests that, for the first predictive model, we can predict the maternal plasma DNA methylation signature as a mixture of the average placental (CVS) DNA methylation signature and the average maternal white blood cells (MBC) DNA methylation signature. Herein this is referred to as the CM model.

Secondly, we can use the DNA from non-pregnant women's plasma to approximate the maternal plasma DNA coming from the maternal tissues. Therefore, as the second predictive model, we can predict the maternal plasma DNA methylation as a mixture of the average placenta (CVS) DNA methylation and the average non-pregnant women plasma (nP) DNA methylation. Herein this is referred to as the CnP model.

Finally, as discussed in Example 1, we can use mixtures of the maternal and fetal tissues to predict the maternal plasma DNA methylation, as long as the signals in these mixtures have different mix proportions. Therefore, as the third predictive model, we can predict the maternal plasma DNA methylation as a mixture of a certain average maternal plasma (P) DNA methylation and the average non-pregnant women plasma (nP) DNA methylation. (Note that because different maternal plasma samples have different proportions of fetal DNA, average maternal plasma methylation will be different for different sets of samples). Herein this is referred to as the PnP model.

Selection of CpG sites: CpG sites were selected in the target chromosome (chr21) and reference chromosomes (chr1, . . . , 12, 14, 15). We experimented with several different ways of identifying informative CpG sites for the detection of aneuploidy. In addition to the use of all CpG sites, we also considered the set of CpG sites that are differentially methylated between the maternal tissues (those that have the most significant influence on the maternal plasma DNA methylation pattern) and the fetal tissues (those that have the most significant influence on the maternal plasma methylation pattern).

For the CM model, the CpG sites must be shared by the CM data and the T21N data. In total we found 2031580 shared CpG sites. Among them, 20859 are located in the target chromosome chr21, 1431135 are located in the reference chromosomes (chr1, . . . , 12, 14, 15). Furthermore, if we require the CpG sites to be differentially methylated between the CVS and MBC samples, with p value <=0.05 and difference in methylation level >=20, the number of CpG sites reduces to 648617, with 7409 in the target chromosome and 455660 in the reference chromosomes.

For the CnP model, the CpG sites must be shared by all three data sets: CM data, PnP data, and T21N data. In total we found 1970505 shared CpG sites. Among them, 20015 are located in the target chromosome chr21, 1391135 are located in the reference chromosomes (chr1, . . . , 12, 14, 15). Furthermore, if we require the CpG sites to be differentially methylated between the CVS and MBC samples, with p value <=0.05 and difference in methylation level >=20, the number of CpG sites reduces to 637599, with 7243 in the target chromosome and 448842 in the reference chromosomes.

For the PnP model, the CpG sites must be shared by two data sets: PnP data, and T21N data. In total we found 6386312 shared CpG sites. Among them, 69344 are located in the target chromosome chr21, 4498308 are located in the reference chromosomes (chr1, . . . , 12, 14, 15). Furthermore, if we require the CpG sites to be differentially methylated between the pregnant and non-pregnant plasma samples, with p value <=0.05 and difference in methylation level between 2 and 20, the number of CpG sites reduces to 546671, with 6316 in the target chromosome and 392185 in the reference chromosomes.

Table 10 below summarizes the CpG sites used in our study:

TABLE 10
CpG sites used in Example Embodiment F.
Prediction	CpG sites		Total	Target
	present	Selection	CpG	Chr	Reference
Model	in	Criteria	sites	sites	Chr sites
CM	CM, T21N	None	2031580	20859	1431135
CM	CM, T21N	P <= .05,	648617	7409	455660
		DM >= 20
CnP	CM, PnP,	None	1970505	20015	1391135
	T21N
CnP	CM, PnP,	P <= .05,	637599	7243	448842
	T21N	DM >= 20 (CM)
PnP	PnP, T21N	None	6386312	69344	4498308
PnP	PnP, T21N	P <= .05,	546671	6316	392185
		2 <= DM <= 20

Application of the modified algorithm: The modified algorithm of this Example was applied to test to test which libraries in T21N are from trisomy 21 pregnancies.

We predicted the fetal frequency using the CpG sites shown in Table 10. For the CM model, we used the average methylation level of the selected CpG sites respectively in the 5 CVS libraries and 6 MBC libraries from the CM data set. For the CnP model, we used the average methylation level of the selected CpG sites respectively in the 5 CVS libraries from the CM data set and 6 maternal plasma libraries from the PnP data set. For the PnP model, we used the average methylation level of the selected CpG sites respectively in the 6 maternal plasma libraries and 9 nonpregnant plasma libraries from the PnP data set. The results of these analyses, shown at FIGS. 33-38, correspond to the 6 rows of Table 10, respectively.

Results: In each figure for the CM models (FIGS. 33 and 34) or CnP models (FIGS. 33 and 34), the Y axis in both panels shows the difference between the fetal frequency predicted by the CpG sites in the target chromosome (chr21) and the fetal frequency predicted by the CpG sites in the reference chromosomes. The red dots represent the normal maternal plasma samples, the blue dots the trisomy 21 (T21) maternal plasma samples. As expected, the difference is close to 0 for the normal samples, but much higher in the T21 samples. The X axis of the right panel is the estimated fetal frequency using the CpG sites in the reference chromosome. It is clear that, for the CM and CnP models, the difference between the chr21 predicted frequency and the reference chromosomes predicted fetal frequency for the T21 samples is positively correlated with the estimated fetal frequency, with a slope of about 0.5. Note that the estimation of fetal frequency probably is higher than expected in the CM model and will need some calibration. This is most likely because CVS and MBC represent the majority, though not all, types of tissues that contribute to the maternal plasma DNA.

For the PnP models (FIGS. 37 and 38), the X axis in the right panel of the figures cannot be interpreted as fetal frequency. It can be, however, interpreted as the ratio of the fetal frequency in each library to the average fetal frequency in the 6 maternal plasma samples from the PnP data. The Y axis in the two panels, therefore, can be interpreted as the ratio of the difference between the chr21 predicted frequency and the reference chromosomes predicted fetal frequency to the average fetal frequency of the 6 maternal plasma samples from the PnP data.

Example Embodiments for Example 3

C1. A method, comprising:

obtaining sequence reads of a methylation sequencing assay covering genomic segments of a maternal plasma sample from a pregnant subject;

estimating a first percentage of fetal DNA molecules (F_A) in the maternal plasma sample based on sequence reads that are aligned to a target chromosome;

estimating a second percentage of fetal DNA molecules (F_D) in the maternal plasma sample based on sequence reads that are aligned to one or more reference chromosomes; and

indicating the presence of a fetal aneuploidy in the pregnant subject responsive to an absolute value difference between the first percentage and the second percentage being larger than a threshold c₁, wherein c₁>0,

wherein the maternal plasma sample comprises maternal DNA and fetal DNA.

C2. The method of embodiment C1, wherein the target chromosome is chromosome 21, chromosome 13, chromosome 18, or chromosome X.
C3. The method of embodiments C1 or C2, wherein the one or more reference chromosomes include one or more of chromosomes 1-12, 14-17, 19, 20 and 22.
C4. The method of any of embodiments C1-C3, wherein indicating fetal aneuploidy comprises indicating that a fetus of the pregnant subject has an abnormal number of copies of the target chromosome.
C5. The method of any of embodiments C1-C4, wherein the pregnant subject is a human.
C6. A method for detecting fetal aneuploidy in a maternal plasma DNA sample obtained from a pregnant woman as described herein.
C7. A method for detecting sub-chromosomal fetal copy number variation in a maternal plasma DNA sample obtained from a pregnant woman as described herein.
C8. A method, comprising:

obtaining sequence reads of a methylation sequencing assay covering genomic segments of a maternal plasma sample from a pregnant subject;

estimating a first percentage of fetal DNA molecules (F_AM) in the maternal plasma sample based on sequence reads that are aligned to a target copy number variation region of a chromosome of a genome;

estimating a second percentage of fetal DNA molecules (F_DM) in the maternal plasma sample based on sequence reads that are aligned to one or more reference regions of the genome; and

indicating the presence of a sub-chromosomal fetal copy number variation in the pregnant subject responsive to an absolute value difference between the first percentage and the second percentage being larger than a threshold c₁, wherein c₁>0, wherein the maternal plasma sample comprises maternal DNA and fetal DNA.

Abstract of this Example (Example 3)

Non-invasive methods are disclosed in this Example for prenatal diagnosis. These methods can be used to detect fetal aneuploidy and/or sub-chromosomal fetal copy number variations. The methods are performed on a biological sample including maternal plasma DNA, such as a plasma or blood sample from a pregnant woman.

Example 4—Methods and Materials for Assessing and Treating Endometriosis

Field of this Example

This Example relates to methods and materials involved in assessing a mammal (e.g., a human) for and/or treating a mammal (e.g., human) having or developing endometriosis. For example, this Example provides methods and materials for using DNA methylation profiles of nucleic acid within a liquid biopsy (e.g., a plasma sample, a urine sample, a tampon blood sample, or a cervical or vaginal swab sample) to determine whether or not a mammal has, or is developing, endometriosis. This Example also provides methods, algorithms, and kits for diagnosing, prognosing, monitoring, classifying, and/or treating endometriosis.

Background of this Example

Endometriosis is a debilitating disease involving the growth of uterine tissue outside the uterus. The primary symptoms are pelvic pain and infertility. Nearly half of affected women have chronic pelvic pain, and in 70 percent of those, the pain occurs during menstruation. Pain with sex also is common. Infertility occurs in up to half of women affected. Less common symptoms include urinary or bowel symptoms. About 25 percent of women have no symptoms. Endometriosis can have both social and psychological effects.

Currently, definitive diagnosis is achieved by surgical biopsy, which is achieved laparoscopically. Because of the invasive nature of this method, diagnosis and therefore treatment, are frequently considerably delayed, and the consequence of this is prolonged and progressive pain and a risk of infertility.

Summary of this Example

This Example provides methods and materials involved assessing a mammal (e.g., a human) for and/or treating a mammal (e.g., human) having or developing endometriosis. For example, this Example provides methods and materials for using DNA methylation profiles of nucleic acid within a liquid biopsy (e.g., a plasma sample, a urine sample, a tampon blood sample, or a cervical or vaginal swab sample) to determine whether or not a mammal has, or is developing, endometriosis. Determining if a mammal (e.g., a human) has, or is likely to develop, endometriosis by assessing DNA methylation profiles of nucleic acid within a liquid biopsy (e.g., a plasma sample, a urine sample, a tampon blood sample, or a cervical swab sample) can aid in the identification of mammals (e.g., humans) that should be treated in a particular manner (e.g., by administering a hormone therapy, by administering a pain medication, and/or by performing a surgical treatment), for example, early in the disease process.

This Example also provides methods for diagnosing, prognosing, monitoring, classifying, and/or treating endometriosis. For example, the methods described in this Example can include determining the methylation status of one or more genomic loci in a sample (e.g., a plasma sample, a urine sample, a tampon blood sample, or a cervical swab sample) of a mammal (e.g., a female human). This Example further provides algorithms and kits for diagnosing, prognosing, monitoring, classifying, and/or treating endometriosis.

In one aspect, this Example provides a method for diagnosing, prognosing, classifying, and/or monitoring endometriosis in a mammal (e.g., a human female) comprising: (a) obtaining a sample (e.g., a plasma sample, a urine sample, a tampon blood sample, or a cervical or vaginal swab sample) from the mammal; (b) determining the methylation status and/or level of one or more genomic loci in the sample; (c) comparing the methylation status and/or level of the one or more genomic loci to a reference; and (d) identifying the mammal as having endometriosis, wherein the difference in the methylation status and/or level of the one or more genomic loci in the sample compared to the reference indicates the presence of endometriosis in the mammal.

In another aspect, this Example provides a method of treating endometriosis in a mammal (e.g., a human female) comprising: (a) obtaining a sample (e.g., a plasma sample, a urine sample, a tampon blood sample, or a cervical or vaginal swab sample) from the mammal; (b) determining the methylation status and/or level of one or more genomic loci present in the sample; (c) comparing the methylation status and/or level of the one or more genomic loci to a reference or a set of predicted; (d) identifying the mammal as having endometriosis, wherein the difference in the methylation status and/or level of the one or more genomic loci in the sample compared to the reference or a set of predicted values indicates the presence of endometriosis in the mammal; and (e) treating the mammal by administering a hormone therapy, by administering a pain medication, and/or by performing a surgical treatment.

In some cases, an increase in the level of methylation of the one or more genomic loci in a sample (e.g., a plasma sample, a urine sample, a tampon blood sample, or a cervical or vaginal swab sample) can indicate the presence of endometriosis in the mammal or a decrease in the level of methylation of the one or more genomic loci in a sample (e.g., a plasma sample, a urine sample, a tampon blood sample, or a cervical swab sample) indicates the presence of the endometriosis in the mammal. In some cases, a decrease in the level of methylation of at least one of the one or more genomic loci in a sample and an increase in the level of methylation of at least one of the one or more genomic loci in the sample can indicate the presence of endometriosis in the mammal.

In some cases, the reference can be the methylation status and/or level of the one or more genomic loci in a sample (e.g., a plasma sample, a urine sample, a tampon blood sample, or a cervical swab sample) obtained from a mammal (e.g., a human female) that does not have endometriosis.

In another aspect, this Example provides a method of treating endometriosis in a mammal (e.g., a human female) comprising: (a) measuring the methylation status and/or level of one or more genomic loci present in a sample (e.g., a plasma sample, a urine sample, a tampon blood sample, or a cervical or vaginal swab sample) from the mammal prior to a treatment of endometriosis; (b) measuring the methylation status and/or level of one or more genomic loci present in a sample (e.g., a plasma sample, a urine sample, a tampon blood sample, or a cervical or vaginal swab sample) from the mammal during the treatment of endometriosis; and (c) continuing the treatment if the difference in the methylation status and/or level of the one or more genomic loci between the samples from prior to and during the treatment of endometriosis indicates the subject is responsive to the treatment. In some cases, the method further comprises (d) administering a different treatment to the mammal if the difference in the methylation status and/or level of the one or more genomic loci between the samples from prior to and during the treatment of endometriosis indicates the subject is not responsive to the treatment.

In some cases, an increase in the level of methylation of the one or more genomic loci in the sample indicates that the mammal is responsive to the treatment, or a decrease in the level of methylation of the one or more genomic loci in the sample indicates the mammal is responsive to the treatment. In certain embodiments, a decrease in the level of methylation of at least one of the one or more genomic loci in the sample and an increase in the level of methylation of at least one of the one or more genomic loci in the sample indicates the mammal is responsive to the treatment. In certain embodiments, an increase in the level of methylation of the one or more genomic loci in the sample indicates the mammal is responsive to the treatment, or a decrease in the level of methylation of the one or more genomic loci in the sample indicates the mammal is not responsive to the treatment. In certain embodiments, a decrease in the level of methylation of at least one of the one or more genomic loci in the sample and an increase in the level of methylation of at least one of the one or more genomic loci in the sample indicates the subject is not responsive to the treatment.

This Example further provides algorithms for diagnosing and/or monitoring a mammal having endometriosis. In certain embodiments, the algorithm of this Example can be used to classify endometriosis of a mammal (e.g., a human female).

In another aspect, this Example provides a kit for diagnosing, prognosing, and/or monitoring endometriosis in a mammal comprising a means for determining and/or detecting the methylation status of one or more genomic loci. In certain embodiments, the means comprises one or more primers and/or probes for determining and/or detecting the methylation status of the one or more genomic loci.

In certain embodiments, the one or more genomic loci are present within nucleic acids isolated from the sample. In certain embodiments, the one or more genomic loci are present within cell-free nucleic acids isolated from the sample.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Although methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, suitable methods and materials are described below. All publications, patent applications, patents, and other references mentioned herein are incorporated by reference in their entirety. In case of conflict, the present specification, including definitions, will control. In addition, the materials, methods, and examples are illustrative only and not intended to be limiting.

Other features and advantages of the invention will be apparent from the following detailed description, and from the claims.

Description of this Example

This Example provides methods for diagnosing, prognosing, monitoring, classifying, and/or treating endometriosis. For example, the methods described herein can include determining the methylation status of one or more genomic loci in a sample of a mammal (e.g., a human female). In some cases, the methods described herein can include the use of an algorithm to diagnose, prognose, monitor, classify, and/or assist in the treatment of endometriosis.

Unless defined otherwise, all technical and scientific terms used in this Example generally have their ordinary meanings in the art, within the context of this Example and in the specific context where each term is used. The following references provide one of skill with a general definition of many of the terms used in this Example: Singleton et al., Dictionary of Microbiology and Molecular Biology (2nd ed. 1994); The Cambridge Dictionary of Science and Technology (Walker ed., 1988); The Glossary of Genetics, 5th Ed., R. Rieger et al. (eds.), Springer Verlag (1991); and Hale & Marham, The Harper Collins Dictionary of Biology (1991). Certain terms are discussed below, or elsewhere in the specification, to provide additional guidance to the practitioner in describing the compositions and methods of this Example and how to make and use them.

The terms “comprise(s),” “include(s),” “having,” “has,” “can,” “contain(s),” and variants thereof, as used herein, are intended to be open-ended transitional phrases, terms or words that do not preclude the possibility of additional acts or structures. The singular forms “a,” “an,” and “the” include plural references unless the context clearly dictates otherwise. The present Example also contemplates other embodiments “comprising,” “consisting of” and “consisting essentially of,” the embodiments or elements presented herein, whether explicitly set forth or not.

As used herein, the use of the word “a” or “an” when used in conjunction with the term “comprising” in the claims and/or the specification may mean “one,” but it is also consistent with the meaning of “one or more,” “at least one,” and “one or more than one.” Still further, the terms “having,” “including,” “containing,” and “comprising” are interchangeable, and one of skill in the art is cognizant that these terms are open ended terms.

The term “about” or “approximately” means within an acceptable error range for the particular value as determined by one of ordinary skill in the art, which will depend in part on how the value is measured or determined, i.e., the limitations of the measurement system. For example, “about” can mean within 3 or more than 3 standard deviations, per the practice in the art. Alternatively, “about” can mean a range of up to 20%, preferably up to 10%, more preferably up to 5%, and more preferably still up to 1% of a given value. Alternatively, particularly with respect to biological systems or processes, the term can mean within an order of magnitude, preferably within 5-fold, and more preferably within 2-fold, of a value.

In certain embodiments, the term “biomarker” refers to a marker (e.g., DNA methylation status) that allows detection of a disease and/or disorder in an individual, including detection of the disease or the disorder in its early stages. In certain embodiments, the term “biomarker” refers to a marker (e.g., DNA methylation status) that allows the characterization of a phenotype of a disease and/or a disorder in an individual. Early stage of a disease, as used herein, refers to the time period between the onset of the disease and the time point that signs or symptoms of the disease emerge. In certain non-limiting embodiments, the presence, absence, and/or level of a biomarker in a sample of a mammal (e.g., a human) is determined by comparing to a reference control.

The terms “reference sample,” “reference control,” “control,” or “reference,” as used interchangeably herein, refers to a control for a methylation status of a genomic locus that is to be detected in a sample of a mammal. In certain embodiments, a reference sample can be a sample from a healthy individual, e.g., an individual that does not have endometriosis. In certain embodiments, a reference sample can be a sample from a control individual that does not have the disease or phenotype to be detected by a biomarker disclosed herein. In certain embodiments, a control or reference can be the presence, absence, and/or a particular level of a methylation state of a genomic locus in a healthy individual. In certain embodiments, a reference can be a predetermined presence, absence, and/or particular level of a methylation state of a genomic locus that indicates a subject does not have endometriosis. In certain embodiments, a reference can be the methylation status of a locus in an individual having a disease or a phenotype, e.g., an individual that has endometriosis, where the methylation status of the locus is known to be not associated with the disease or the phenotype.

The term “a set of predicted values” refers to the methylation status of certain genomic loci for a sample. The status of those loci is not directly measured from that sample. Rather, it is inferred from measurements of other loci for that sample and/or measurements of other samples. The inference of the predicted values is based on some mathematical/statistical models. The models usually assume that the sample for which the methylation status of those loci is to be predicted has a normal phenotype. This assumption may be either correct or wrong, but its correctness is not required for the inference of the predicted values.

The term “slightly invasive or non-invasive method” refers to a method that does not involve the removal of tissues by biopsy from the uterus or endometrium. In certain embodiments, slightly invasive or non-invasive methods, as described herein, include obtaining plasma, urine, or a cervical or vaginal swab from a subject.

The term “patient” or “subject,” as used interchangeably herein, refers to any warm-blooded animal, e.g., human or non-human. Non-limiting examples of non-human subjects include mammals, non-human primates, dogs, cats, mice, rats, guinea pigs, rabbits, fowl, pigs, horses, cows, goats, sheep, etc. In certain embodiments, the subject is human.

The term “nucleic acid,” “nucleic acid molecule,” or “polynucleotide” includes any compound and/or substance that comprises a polymer of nucleotides. Each nucleotide is composed of a base, specifically a purine- or pyrimidine base (i.e., cytosine (C), guanine (G), adenine (A), thymine (T), or uracil (U)), a sugar (i.e., deoxyribose or ribose), and a phosphate group. In certain embodiments, the nucleic acid molecule is described by the sequence of bases, whereby said bases represent the primary structure (linear structure) of a nucleic acid molecule. The sequence of bases is typically represented from 5′ to 3′. These terms encompass deoxyribonucleic acid (DNA) including, e.g., complementary DNA (cDNA) and genomic DNA, ribonucleic acid (RNA), in particular messenger RNA (mRNA), synthetic forms of DNA or RNA, and mixed polymers comprising two or more of these molecules. The herein described nucleic acid molecule can contain naturally occurring or non-naturally occurring nucleotides. Examples of non-naturally occurring nucleotides include modified nucleotide bases with derivatized sugars or phosphate backbone linkages or chemically modified residues.

The term “isolated” (e.g., isolated genomic DNA) refers to a biological component that has been substantially separated or purified away from other biological components in the cell of the organism in which the component naturally occurs, e.g., other chromosomal and extra-chromosomal DNA and RNA, proteins, and organelles. Nucleic acids, e.g., DNA, that have been “isolated” include nucleic acids purified by standard purification methods.

The term “genomic locus” or “genomic DNA locus,” as used herein, refers to any fixed position in a genome. For example, a genomic locus can refer to a genomic element, a chromosomal region, a gene, a region of a gene, e.g., an exon or intron, a regulatory region of a gene, e.g., a promoter or enhancer, a CpG site, a CpG island, or a CpG island shore. For example, a genomic locus can include one or more CpG sites, e.g., between about 1 to about 100 CpG sites. In certain embodiments, a genomic locus can be of any particular length, e.g., between about 1 to about 10,000 nucleotides in length.

As used interchangeably herein, “methylation state,” “methylation profile,” “methylation status,” and “methylation level” refer to the presence, absence, percentage, and/or quantity of methylation at a particular nucleotide, or nucleotides, within a DNA region, e.g., a genomic locus. The methylation status of a particular DNA sequence (e.g., a genomic locus) can indicate the methylation state of every nucleotide in the sequence, indicate the methylation state of any of the nucleotides (e.g., cytosines) in the sequence, can indicate the methylation state of a subset of the nucleotides (e.g., of cytosines), can indicate the percentage or fraction of methylated cytosines at any particular stretch of nucleotides within the sequence or can indicate the average rate of methylation of all the cytosines (or a subset of the cytosines) present in a nucleic acid.

As used herein, a “methylated nucleic acid molecule” refers to a nucleic acid molecule that contains one or more methylated nucleotides that is/are methylated.

As used herein, a “CpG site” or “methylation site” is a nucleotide within a nucleic acid that is susceptible to methylation either by natural occurring events in vivo or by an event instituted to chemically methylate the nucleotide in vitro.

A “CpG island,” as used herein, describes a segment of a nucleic acid, e.g., DNA sequence, that have a high frequency of CpG dinucleotide repeats. See, e.g., Illingworth and Bird, FEBS Letters, 2009; 583:1713-1720. For example, Yamada et al. (Genome Research, 2004; 14:247-266) have described a set of standards for determining a CpG island: it must be at least 400 nucleotides in length, has a GC content greater than 50%, and an OCF/ECF ratio greater than 0.6. Others (Takai et al., Proc. Natl. Acad. Sci. U.S.A., 2002; 99:3740-3745) have defined a CpG island less stringently as a sequence of at least 200 nucleotides in length, having a greater than 50% GC content and an OCF/ECF ratio greater than 0.6.

A “CpG island shore,” as used herein, refers to methylation hotspots that are present a short distance, e.g., less than 2 kb, from CpG islands.

The term “methylome,” as used herein, refers to the amount or pattern of methylation at different sites or regions within a population of cells. The methylome can correspond to all of the genome, a subset of the genome (e.g., repeat elements in the genome), or a portion of the subset (e.g., those areas found to be associated with endometriosis). A methylome from plasma can be referred to a “plasma fluid methylome,” or a “plasma fluid DNA methylome.” The plasma fluid methylome is an example of a cell-free methylome that includes cell-free DNA (cfDNA).

As used herein, the term “increase” refers to alter positively by at least about 2%, including, but not limited to, alter positively by about 5%, by about 10%, by about 15%, by about 20%, by about 25%, by about 30%, by about 35%, by about 40%, by about 45%, by about 50%, by about 55%, by about 60%, by about 65%, by about 70%, by about 75%, by about 80%, by about 85%, by about 90%, by about 95% or by about 100%.

As used herein, the terms “reduce,” “reduction,” or “decrease” refers to alter negatively by at least about 2%, including, but not limited to, alter negatively by about 5%, by about 10%, by about 15%, by about 20%, by about 25%, by about 30%, by about 35%, by about 40%, by about 45%, by about 50%, by about 55%, by about 60%, by about 65%, by about 70%, by about 75%, by about 80%, by about 85%, by about 90%, by about 95% or by about 100%.

As described herein, this Example provides methods for diagnosing, monitoring, classifying, and/or treating endometriosis by analyzing the methylation status of one or more genomic loci in a sample (e.g., a plasma sample, a urine sample, a tampon blood sample, or a cervical or vaginal swab sample) of a mammal (e.g., a human female). In certain embodiments, the methods can include using an algorithm described herein. In certain embodiments, the methods described herein can allow for the early diagnosis or screening of a subject with endometriosis, e.g., the subject does not have any symptoms, or only have early symptoms of endometriosis.

In certain embodiments, samples obtained for use in the methods described herein can include cfDNA, which carries DNA methylation information from the cell of origin. cfDNA can arise from cellular apoptosis and necrosis, and can be generated from active secretory processes, with the formation of extracellular vesicles. DNA methylation signatures are highly tissue-specific, and include in vivo information relating to the tissue source of cfDNA. In certain embodiments, the methods described herein can include analyzing cfDNA in a sample (e.g., a plasma sample, a urine sample, a tampon blood sample, or a cervical or vaginal swab sample), to identify genetic phenotypes that are drivers and/or consequences of endometriosis.

The sample from the subject can be collected using any appropriate technique. For example, a blood sample, a plasma sample, a urine sample, a tampon blood sample, or a cervical or vaginal swab sample can be collected using standard methods. In some cases, the sample can be collected from the subject before the subject has any symptom of endometriosis, i.e., a non-symptomatic subject. In certain embodiments, the sample can be collected from the non-symptomatic subject who is at high risk of endometriosis. In certain embodiments, the sample can be collected from the subject who has previously received or is currently receiving a treatment for endometriosis. In certain embodiments, two or more samples (e.g., two or more, three or more, four or more, five or more, six or more or seven or more samples) can be obtained before and during the subject is receiving a treatment for endometriosis (e.g., serially obtained samples).

Diagnostic, Prognostic, Classification, and Monitoring Methods of this Example

This Example provides diagnostic and prognostic methods for diseases and/or disorders that are characterized by differential methylation of genomic loci. For example, this Example provides methods for diagnosing, prognosing, classifying, and/or monitoring endometriosis in a subject that includes analyzing the methylation status of certain genomic loci.

In certain embodiments, the analyzed genomic loci can include one or more genomic loci that exhibit differential methylation in a sample from a subject that has endometriosis compared to a reference sample. For example, the methods described herein can include assessing the methylation status of one or more genomic loci, e.g., about 5 or more, about 10 or more, about 50 or more, about 100 or more, about 500 or more, about 1,000 or more, about 5,000 or more, about 10,000 or more, about 25,000 or more, about 50,000 or more or about 100,000 or more genomic loci in a sample of a subject. In certain embodiments, the one or more genomic loci can be one or more promoter regions of one or more genes, one or more exons of one or more genes, one or more introns of one or more genes, one or more CpG sites, one or more CpG islands, one or more CpG island shores, one or more enhancers of one or more genes, or a combination thereof. In certain embodiments, the genomic loci are present in intergenic regions. In certain embodiments, the genomic loci are present on a particular chromosome.

In certain embodiments, this Example provides methods for diagnosing, prognosing, and/or monitoring endometriosis in a subject by detecting the DNA methylation profiles associated with endometriosis. In certain embodiments, the methods described herein can include (a) obtaining a sample from the subject, (b) determining the methylation status of one or more genomic loci present in the sample, e.g., present within cfDNA in a plasma sample, a urine sample, a tampon blood sample, or a cervical or vaginal swab sample, (c) comparing the methylation status of the one or more genomic loci to a reference or a set of predicted values, and (d) diagnosing endometriosis in the subject. In certain embodiments, the difference in the methylation status of the one or more genomic loci in the sample compared to the reference or a set of predicted values indicates the presence of endometriosis in the subject. In certain embodiments, the difference in the methylation status also can indicate the severity of endometriosis.

In certain embodiments, the methods described herein for diagnosing, prognosing, and/or monitoring endometriosis in a subject can include (a) obtaining a sample from the subject, (b) determining the level of methylation of one or more genomic loci present in the sample, (c) comparing the level of methylation of the one or more genomic loci to a reference or a set of predicted values, and (d) diagnosing endometriosis in the subject. In certain embodiments, the difference in the level of methylation of the one or more genomic loci in the sample compared to the reference or a set of predicted values indicates the presence of endometriosis in the subject. In certain embodiments, the difference in the methylation level also can indicate the severity of endometriosis.

In certain embodiments, diagnosing endometriosis in the subject can include characterizing a phenotype of the endometriosis, wherein the difference in the methylation status of the one or more genomic loci in the sample compared to the reference or a set of predicted values indicates the phenotype of the endometriosis. In certain embodiment, the phenotype of the endometriosis can include the severity of the endometriosis, prognosis of the endometriosis, molecular expression profile of the endometriosis, responsiveness of the endometriosis to certain treatments, or any combinations thereof.

In certain embodiments, the methods described herein for determining if a subject is at risk of developing endometriosis in the subject can include (a) obtaining a sample from the subject, (b) determining the level of methylation of one or more genomic loci present in the sample, (c) comparing the level of methylation of the one or more genomic loci to a reference or a set of predicted values, and (d) determining that the subject is at risk of developing endometriosis, wherein the difference in the level of methylation of the one or more genomic loci in the sample compared to the reference or a set of predicted values indicates that the subject is at risk.

In certain embodiments, diagnosing, prognosing, and/or monitoring of a subject with endometriosis can be based on a higher or lower methylation level of the genomic locus in the sample of the subject relative to the methylation level in a reference sample, e.g., a sample from a subject that does not have endometriosis. In certain embodiments, a difference of greater than about 5%, greater than about 10%, greater than about 15%, greater than about 20%, greater than about 25%, greater than about 30%, greater than about 35%, greater than about 40%, greater than about 45%, greater than about 50%, greater than about 55%, greater than about 60%, greater than about 65%, greater than about 70%, greater than about 75%, greater than about 80%, greater than about 85%, greater than about 90% or greater than about 95% in the methylation (e.g., level, percentage and/or fraction) of the one or more genomic loci in a sample obtained from a subject compared to a control can be indicative that the subject has endometriosis or is at risk of developing endometriosis. In certain embodiments, the difference can be a decrease in methylation (e.g., level, percentage, and/or fraction) of the genomic loci in the sample of the subject. Alternatively, the difference can be an increase in methylation (e.g., level, percentage, and/or fraction) of the genomic loci in the sample of the subject. In certain embodiments, the difference can be a decrease in the methylation of a genomic locus and an increase in the methylation of a different genomic locus in the sample obtained from the subject. In certain embodiments, a decrease in the level of methylation of one or more genomic loci in the sample and the increase in the level of methylation of one or more different genomic loci in the sample can indicate the presence of endometriosis.

In certain embodiments, diagnosis of a subject with endometriosis can be based on the methylated or unmethylated state of a genomic locus, e.g., a CpG site. In certain embodiments, a genomic locus, e.g., a CpG site, in a sample from a subject diagnosed with endometriosis can be methylated and the genomic locus, e.g., the CpG site, in a reference sample can be unmethylated. In certain embodiments, a genomic locus in a sample from a subject diagnosed with endometriosis can be unmethylated and the genomic locus in a reference sample can be methylated.

Diagnostic, Prognostic, Classification, and Monitoring Methods Using an Algorithm of this Example

This Example also provides diagnostic and prognostic methods for diseases and/or disorders that are characterized by differential methylation of genomic loci by using an algorithm as described, for example, in Example Embodiment H. For example, this Example provides methods for diagnosing, prognosing, classifying, and/or monitoring endometriosis in a subject that includes analyzing the methylation status of certain genomic loci and/or genomic fractions.

Methods for Treating Endometriosis

This Example also provides methods for treating a subject having endometriosis. For example, a mammal (e.g., a human female) that was identified as having endometriosis as described herein (or identified as being at risk of developing endometriosis as described herein) can be administered one or more hormone therapies, one or more pain medications, or a combination thereof to treat endometriosis. Examples of hormone therapies that can be used as described herein include, without limitation, gonadotropin-releasing hormone therapies (e.g., elagolix), estrogen therapies, progestin therapies, estrogen and progestin combination therapies, progesterone therapies, progesterone and progestin combination therapies, danazol therapies, and gestrinone therapies. Examples of pain medications that can be used as described herein include, without limitation, nonsteroidal anti-inflammatory drugs. In some cases, a mammal (e.g., a human female) that was identified as having endometriosis as described herein (or identified as being at risk of developing endometriosis as described herein) can be treated using a surgical procedure to treat endometriosis. Examples of surgical procedures that can be used as described herein include, without limitation, laparoscopic surgeries to remove one or more endometriosis patches, laparotomy surgeries to remove one or more endometriosis patches, and surgeries to sever pelvic nerves (e.g., presacral neurectomy or laparoscopic uterine nerve ablation surgeries).

In some cases, the information provided by the methods described herein can be used by a clinician or physician in determining the most effective course of treatment (e.g., preventative or therapeutic) for the subject. A course of treatment refers to the measures taken for a patient after the prognosis or the assessment of increased risk for development of endometriosis is made. For example, when a subject is identified to have an increased risk of developing endometriosis, the physician can determine whether frequent monitoring of DNA methylation changes can be performed as a prophylactic measure. Also, when a subject is diagnosed with endometriosis (e.g., based on the presence of a DNA methylation pattern in a sample from a subject), it can be advantageous to follow such detection with a therapeutic treatment.

In some cases, this Example provides methods for assessing the efficacy of a therapeutic or prophylactic therapy for treating endometriosis in a subject, comprising determining the methylation status of one or more genomic loci present in a sample obtained from a subject prior to the therapy and determining methylation status of the one or more genomic loci present in a sample obtained from the subject at one or more time points during the therapeutic or prophylactic therapy, wherein the therapy is efficacious for treating endometriosis in a subject when there is a change in the presence and/or level of methylation of the one or more genomic loci in the second or subsequent samples, relative to the first sample. In certain embodiments, the first sample is obtained after therapeutic treatment has begun.

In certain embodiments, the methods for monitoring the response in a subject to prophylactic or therapeutic treatment of endometriosis can include measuring the methylation status and/or level of one or more genomic loci in a sample of a subject at a first time-point, administering a therapeutic agent, re-measuring the methylation status and/or level of the one or more genomic loci at a second time-point, comparing the results of the first and second measurements and optionally modifying the treatment regimen based on the comparison. In certain embodiments, the first time-point can be prior to an administration of the therapeutic agent, and the second time-point can be after said administration of the therapeutic agent. In certain embodiments, the first time-point can be prior to the administration of the therapeutic agent to the subject for the first time. In certain embodiments, the dose (defined as the quantity of therapeutic agent administered at any one administration) can be increased or decreased in response to the comparison. In certain embodiments, the dosing interval (defined as the time between successive administrations) can be increased or decreased in response to the comparison, including total discontinuation of treatment. In addition, the methods described herein can be used to determine the efficacy of the therapeutic treatment, wherein a change in the methylation status of certain genomic loci present in a sample of a subject can indicate that the therapeutic treatment regimen can be altered, reduced, and/or stopped.

Assays of this Example

This Example also provides assays and/or methods for determining the DNA methylation status and/or level of genomic loci that correlates with the presence, absence, and/or severity of endometriosis. In some cases, the assay method can include comparing the methylation status and/or level of genomic loci present in a sample from a subject that has endometriosis to the methylation status and/or level of genomic loci in a sample from a healthy subject to determine the methylation pattern, as described above, that correlates with the presence of endometriosis. In some cases, the assay methods can include comparing the methylation status and/or level of genomic loci in a sample from a subject that has endometriosis at an early stage to the methylation status and/or level of genomic loci in a sample from a subject that has endometriosis at a late stage to determine the methylation status and/or level that correlates with the different stages and/or severity of endometriosis.

DNA Isolation Techniques of this Example

In certain embodiments, the methods described herein can include isolating nucleic acid from a sample (e.g., a plasma sample, a urine sample, a tampon blood sample, or a cervical or vaginal swab sample) obtained from a subject. Any appropriate technique can be used to isolate nucleic acids from a sample. For example, isolation of DNA from a plasma sample can be performed by extraction methods using organic solvents such as a mixture of phenol and chloroform, followed by precipitation with ethanol (see, for example, J. Sambrook et al., “Molecular Cloning: A Laboratory Manual,” 1989, 2nd Ed., Cold Spring Harbor Laboratory Press: New York, N.Y.). Additional non-limiting examples include salting out DNA extraction (see, for example, P. Sunnucks et al., Genetics, 1996, 144:747-756; and S. M. Aljanabi and I. Martinez, Nucl. Acids Res. 1997, 25:4692-4693), the trimethylammonium bromide salts DNA extraction method (see, for example, S. Gustincich et al., BioTechniques, 1991, 11:298-302), and the guanidinium thiocyanate DNA extraction method (see, for example, J. B. W. Hammond et al., Biochemistry, 1996, 240:298-300). There are also numerous commercially available kits that can be used to extract DNA from biological fluids (e.g., plasma samples) or cells. For example, Qiagen's Gentra PureGene Cell Kit, QlAamp Circulating Nucleic Acid Kit, QiaAmp DNA Mini Kit, DNeasy Blood and Tissue Kit or QiaAmp DNA Blood Mini Kit (Qiagen, Hilden, Germany), GenomicPrep™ Blood DNA Isolation Kit (Promega, Madison, Wis.) and GFX™ Genomic Blood DNA Purification Kit (Amersham, Piscataway, N.J.) can be used to obtain DNA from a sample from a subject.

Methylation Detection Techniques of this Example

Various methylation analysis procedures are known in the art, and can be used with the methods described herein. These assays allow for determination of the methylation state of one genomic locus, e.g., one or more CpG sites or islands within a nucleic acid obtained from a sample. In addition, the methods can be used to quantify the methylation of a genomic locus. Such assays involve, among other techniques, DNA sequencing of bisulfite-treated DNA, PCR (for sequence-specific amplification), digital PCR and use of methylation-sensitive restriction enzymes.

In certain embodiments, methylation-specific PCR can be used to determine the methylation status of a genomic loci. Methylation-specific PCR is based on a chemical reaction of sodium bisulfite with DNA that converts unmethylated cytosines, e.g., of CpG dinucleotides, to uracil or UpG, followed by traditional PCR. Methylated cytosines will not be converted in this process, and primers can be designed to overlap the methylation site, e.g., CpG site, of interest, thereby allowing one to determine the methylation status of the methylation site as methylated or unmethylated. Additionally, restriction enzyme digestion of PCR products amplified from bisulfite-converted DNA may be used, e.g., by using the method described by Sadri & Hornsby (Nucl. Acids Res. 1996; 24:5058-5059) or COBRA (Combined Bisulfite Restriction Analysis) (Xiong & Laird, Nucleic Acids Res. 1997; 25:2532-2534).

In certain embodiments, whole genome bisulfite sequencing, which is a high-throughput genome-wide analysis of DNA methylation, can be used to determine the methylation status of multiple genomic loci. It is based on sodium bisulfite conversion of genomic DNA, as described above, which is then sequenced on a next-generation sequencing platform. The sequences obtained are then re-aligned to the reference genome to determine the methylation states of cytosines, e.g., of CpG dinucleotides, present within the analyzed genomic loci based on mismatches resulting from the conversion of unmethylated cytosines into uracil.

In certain embodiments, genome-wide DNA methylation profiling can be performed using commercially-available arrays, thereby allowing the interrogation of multiple genomic loci, e.g., multiple CpG sites. Non-limiting examples of such arrays include HumanMethylation BeadChips (Illumina, San Diego, Calif.) and Infinium MethylationEPIC kit (Illumina). Additional methods for analyzing the methylation state of multiple genomic loci are provided in Yong et al., Epigenetics & Chromatin 2016; 9:26, which is incorporated by reference herein.

Kits of this Example

This Example provides kits for diagnosing, monitoring, classifying, and/or treating a subject with endometriosis. The kits described herein can comprise a means for determining and/or detecting the methylation status of one or more genomic loci.

Kits of this Example can include, without limitation, packaged probe and primer sets (e.g., TaqMan probe/primer sets), arrays/microarrays, which further contain one or more probes, primers, or other detection reagents for determining the methylation state and/or level of one or more genomic loci. For example, a kit described herein can include one or more probes or primers for detecting the methylation state of one or more genomic loci. In certain embodiments, the one or more genomic loci comprise a CpG site. In certain embodiments, one or more of the genomic loci do not comprise a CpG site. For example, about 5% or more, about 10% or more, about 15% or more, about 20% or more, about 25% or more, about 30% or more, about 35% or more, about 40% or more, about 45% or more, about 50% or more, about 55% or more, about 60% or more, about 65% or more or about 70% or more of the one or more genomic loci detected by the primers or probes can comprise one or more CpG sites.

In certain non-limiting embodiments, a primer and/or probe described herein can be at least about 10 nucleotides or at least about 15 nucleotides or at least about 20 nucleotides in length, and/or up to about 200 nucleotides or up to about 150 nucleotides or up to about 100 nucleotides or up to about 75 nucleotides or up to about 50 nucleotides in length.

In a further non-limiting embodiment, the oligonucleotide primers and/or probes can be immobilized on a solid surface or support, for example, on a nucleic acid microarray, wherein the position of each oligonucleotide primer and/or probe bound to the solid surface or support is known and identifiable.

In certain non-limiting embodiments, the kits described herein can additionally include other components such as a buffer, enzymes such as DNA polymerases or ligases, nucleotides such as deoxynucleotide triphosphates, positive control sequences, and/or negative control sequences necessary to carry out an assay or reaction to detect the methylation state of a genomic locus.

In certain embodiments, the kits described herein can include a container comprising one or more probes and/or primers for detecting the methylation state of one or more genomic loci. In certain embodiments, the kits further include instructions for use, e.g., the instructions can describe that a particular methylation status of a genomic locus is indicative of endometriosis in a subject. The instructions can be printed directly on the container (when present), or as a label applied to the container, or as a separate sheet, pamphlet, card or folder supplied in or with the container.

Reports, Programmed Computers, and Systems of this Example

In certain embodiments, a diagnosis and/or monitoring of endometriosis in a subject based on the methylation status of one or more genomic loci as described herein can be referred to herein as a “report.” A tangible report can optionally be generated as part of a testing process (which can be interchangeably referred to herein as “reporting,” or as “providing” a report, “producing” a report or “generating” a report).

Examples of tangible reports can include, without limitation, reports in paper (such as computer-generated printouts of test results) or equivalent formats and reports stored on computer readable medium (such as a CD, USB flash drive or other removable storage device, computer hard drive, or computer network server, etc.). Reports, particularly those stored on computer readable medium, can be part of a database, which can optionally be accessible via the internet (such as a database of patient records or genetic information stored on a computer network server, which can be a “secure database” that has security features that limit access to the report, such as to allow only the patient and the patient's medical practitioners to view the report while preventing other unauthorized individuals from viewing the report, for example). In addition to, or as an alternative to, generating a tangible report, reports can also be displayed on a computer screen (or the display of another electronic device or instrument).

A report can include, for example, an individual's medical history, or can just include size, presence, absence, or levels of one or more markers (for example, a report on computer readable medium such as a network server can include hyperlink(s) to one or more journal publications or websites that describe the medical/biological implications). Thus, for example, the report can include information of medical/biological significance as well as optionally also including information regarding the methylation status of relevant genomic loci, or the report can just include information regarding the methylation status of relevant genomic loci without other medical/biological significance.

A report can further be “transmitted” or “communicated” (these terms can be used herein interchangeably), such as to the individual who was tested, a medical practitioner (e.g., a doctor, nurse, clinical laboratory practitioner, genetic counselor, etc.), a healthcare organization, a clinical laboratory, and/or any other party or requester intended to view or possess the report. The act of “transmitting” or “communicating” a report can be by any means known in the art, based on the format of the report. Furthermore, “transmitting” or “communicating” a report can include delivering a report (“pushing”) and/or retrieving (“pulling”) a report. For example, reports can be transmitted/communicated by various means, including being physically transferred between parties (such as for reports in paper format) such as by being physically delivered from one party to another, or by being transmitted electronically or in signal form (e.g., via e-mail or over the internet, by facsimile, and/or by any wired or wireless communication methods known in the art) such as by being retrieved from a database stored on a computer network server, etc.

In certain embodiments, the disclosed subject matter provides computers (or other apparatus/devices such as biomedical devices or laboratory instrumentation) programmed to carry out the methods described herein, e.g., to perform the algorithm disclosed herein (see Example Embodiment H). In certain embodiments, the system can be controlled by the individual and/or their medical practitioner in that the individual and/or their medical practitioner requests the test, receives the test results back and (optionally) acts on the test results to reduce the individual's endometriosis risk or treat the individual, such as by implementing an endometriosis management system.

Example Embodiment G of Example 4: Discovery of Putative

Uterine/Endometrial-Derived Nucleic Acids in Human Plasma Via Epigenomic Liquid Biopsy

Example Embodiment G provides an analysis of epigenomic liquid biopsy of human plasma for the discovery of putative uterine/endometrial-derived cell-free DNA (cfDNA) methylation signatures. Example Embodiment G used solution phase hybridization and high throughput bisulfate sequencing to compare DNA methylation signatures of cfDNA obtained from the plasma samples of women who had previously had a hysterectomy with those from control women who had not previously had a hysterectomy.

A total of n=9 women (the case group) who had previously had a hysterectomy and n=11 women (the control group) who had not previously had a hysterectomy were included. No women in the control group were pregnant at the time of analysis.

Methods

DNA was extracted from plasma volumes ranging from ˜6.3 to 8.1 mL using the NucleoSnap DNA Plasma Kit (Macherey-Nagel) and quantified by Agilent Bioanalyzer. The average yield of cfDNA was 5.625 ng/mL plasma. DNA sequencing libraries were prepared using the Kapa Hyper Prep Kit (Roche). Zymo DNA Methylation Direct Kit was used for bisulfite conversion and targeted libraries were prepared using the SeqCap-Epi (Roche). Libraries were sequenced on an Illumina HiSeq 2500 instrument using 150 bp, paired-end reads.

Reads were trimmed for quality and adaptor sequences using Trim-Galore. The reads were aligned to the human reference sequence (GRCh38/hg20) using Bismark in paired-end and single-end (unaligned paired-end reads), bowtie2 modes. Read duplicates were removed using Bismark. Methylation was called on paired-end and single-end files and then merged. Average on-target coverage was 44.49×.

CpG methylation calls with read depth of at least 10× were read into MethylSig for differential methylation analysis. MethylKit was used for sliding window analysis with a window size of 250 bp and step size of 50 bp. Differentially methylated regions were identified according to (insert paper reference).

Results

Analysis of plasma epigenomic liquid biopsy data using MethylSig revealed 42,403 significant CpG sites that were differentially methylated between cases and controls. Similarly, analysis of the same plasma epigenomic liquid biopsy data using MethylKit revealed 16,562 significant 250 bp sliding windows (identified by Bonferroni corrected p-value and a confidence interval methylation difference of at least 5%) that were differentially methylated between cases and controls.

To further explore the data, Example Embodiment G used the Genotype-Tissue Expression (GTEx) project database. GTEx is an on-going effort to build a comprehensive public resource to study tissue-specific gene expression and regulation. Example Embodiment G identified genes whose expressions are low in whole blood and comparatively highly expressed in uterine tissues. It was rationalized that leukocytes are a major (though not exclusive) contributor to cfDNA in plasma. Furthermore, the uterus is likely to be the sole contributor of uterine/endometrial-derived cfDNA in plasma. Thus, the identification of tissue-specific differences in gene expression in this context, was carried out to illuminate the DNA methylation data and assist in the discovery of putative uterine/endometrial-derived cfDNA in plasma. Specifically, genes were identified whose expressions were elevated by a minimum of 5-fold in uterus compared to whole blood and whose expressions were below two transcripts per million (TPM) (from GTEx). These were then merged with the list of n=16,562 significant CpGs identified via sliding windows analysis (see above). This identified 3,538 significantly differentially methylated windows (not shown) located within and adjacent to the tissue-specific differentially expressed genes. The top n=30 most significantly differentially methylated windows from this output are listed in Table 11. This list of CpG sites represents examples of specific DNA methylation differences that may be detected in the relevant biological samples listed herein, demonstrating that DNA methylation differences such as those listed and many others exist and can be used to differentiate samples as described herein, for example, as described using procedures similar to those set forth in Example Embodiment H.

TABLE 11
						Fold
Chromsome	Coordinate	methControl	methCase	meth.diff	pvalue	Change
chr19	5229601	15.04%	30.22%	15.18%	1.12E−131	41.3317959
chr19	5229551	12.48%	27.81%	15.33%	1.31E−121	41.3317959
chr19	5229501	11.21%	27.39%	16.17%	9.03E−121	41.3317959
chr19	5229451	11.38%	30.18%	18.80%	3.98E−116	41.3317959
chr19	719301	22.35%	41.33%	18.97%	2.94E−114	54.1302115
chr1	240492801	35.76%	49.22%	13.47%	4.27E−114	7.42897026
chr19	719251	22.30%	39.77%	17.46%	7.39E−114	54.1302115
chr10	133529601	6.01%	12.98%	6.97%	1.57E−113	5.95805128
chr10	133529551	6.74%	13.90%	7.16%	6.14E−113	5.95805128
chr1	240492851	35.08%	47.05%	11.97%	3.27E−109	7.42897026
chr19	5229651	16.64%	31.59%	14.96%	8.96E−108	41.3317959
chr10	133529651	6.16%	13.22%	7.06%	6.55E−107	5.95805128
chr1	240492751	36.14%	49.74%	13.60%	8.86E−107	7.42897026
chr1	240492901	35.19%	46.74%	11.55%	1.43E−102	7.42897026
chr10	789451	15.66%	30.20%	14.54%	1.75E−102	26.0717831
chr6	37650051	32.66%	43.72%	11.06%	1.85E−102	5.70832736
chr19	719201	22.10%	38.04%	15.94%	6.21E−97	54.1302115
chr10	133529501	7.35%	14.96%	7.62%	5.78E−96	5.95805128
chr1	201650401	42.76%	29.94%	−12.82%	9.46E−96	9.64816418
chr6	37648901	6.77%	14.26%	7.49%	3.13E−94	5.70832736
chr1	201650451	44.10%	32.00%	−12.10%	6.68E−92	9.64816418
chr11	67071551	17.15%	27.45%	10.30%	4.55E−90	21.1637777
chr19	5229401	12.57%	33.69%	21.13%	9.26E−90	41.3317959
chr3	75396351	52.34%	64.56%	12.22%	1.40E−88	6.87247351
chr19	719351	28.35%	48.18%	19.83%	2.46E−88	54.1302115
chr6	37650001	33.33%	44.40%	11.07%	1.49E−87	5.70832736
chr6	37648951	6.55%	13.54%	6.99%	2.08E−87	5.70832736
chr10	133528001	8.42%	17.33%	8.91%	2.04E−86	5.95805128
chr1	201650351	40.04%	26.56%	−13.48%	6.13E−86	9.64816418
chr3	75396251	48.26%	58.90%	10.64%	2.97E−85	6.87247351

Discussion of this Example

This example demonstrates a comprehensive quantitative genome-wide analysis of DNA methylation in human plasma from women who have previously had a hysterectomy and women who have not previously had a hysterectomy. These results demonstrate that epigenomic liquid biopsy of human plasma can be used for quantitative analysis of cfDNA methylation and, in this example, the discovery of putative uterine/endometrial-derived DNA methylation signatures.

Example Embodiment H of Example 4: Estimation of Abnormal Plasma Methylome Variation in Targeted Regions for Diagnosis of Endometriosis

Provided below is an algorithm that can be used to diagnose a subject with endometriosis. The presently disclosed subject matter provides that the methylome(s) of uterus of a mammal, or structures therein (e.g., endometrium), could be affected by certain abnormalities (e.g., endometriosis), and that the changes of these methylomes can lead to changes in the methylation patterns of the DNA fragments found in plasma, which are released by uterus/endometrium-derived tissues. An algorithm was developed to identify the changes of methylation patterns in the methylome of plasma caused by uterine/endometrial phenotypes.

The main insight behind this algorithm was that the methylome of the DNA fragments in plasma is a mixture of a variety of component methylomes of uterine/endometrial origin, and that the proportion of these different component methylomes in the mixture varies from subject to subject, even among the population with normal uterine/endometrial phenotypes. By constructing a model of plasma methylome as a linear combination of various component methylomes of uterine/endometrial and other origins, the algorithm can accurately predict the methylation patterns of a new plasma sample under the hypothesis that it is from a normal individual.

Consequently, the algorithm of this Example has high sensitivity for detecting abnormal methylation patterns in a plasma sample caused by changes of the methylomes of some uterine/endometrial or other relevant tissues when the sample is from an affected individual.

The procedure can be applied with little modification to the diagnosis and phenotyping of endometriosis using other types of biopsy samples, such as cervical swabs, vaginal swabs, urine, and tampon blood, provided that the DNA fragments from the tissues affected by endometriosis can be found in those biopsy samples.

Let i be any CpG site in human genome, z_i,j be the methylation level of CpG site i in a plasma sample j, p_i,r,j be the proportion of the r^th component methylome m_r,j of ovarian origin in plasma sample j at site i, m_i,r,j be the methylation level of CpG i in methylome m_r,j. The hypothesis is:

Z_i,j=Σ_r=1^Rp_i,r,jm_i,r,j  (1)

where p_I,r,j, m_I,r,j>=0, m_I,r,j<=1, p_I,1,j+ . . . +p_I,R,j=1.

It is further assumed that there is a set of CpG sites S such that, for any CpG site i in S, and any plasma j from a normal individual, it has m_I,r,j=m_I,r and p_I,r,j=p_r,j.

That is, it is assumed that in any plasma sample from a normal individual, the proportions of different component methylomes in the mixture are the same for all CpG sites in S. It is also assumed that, by restricting to the set of CpG sites S, plasma samples from all normal individuals have the same set of component methylomes. They are called restricted reference component methylomes (RRCM), and are labeled as m₁^S, . . . , m_R^S or simply m₁, . . . , m_R when there is no confusion. For any plasma sample j from a normal individual, its methylome restricted to set of CpG sites in S can be expressed as a weighted average of the restricted reference component methylomes. More precisely, let z_i^s be the methylome of plasma sample C restricted to S, then for some mixture vector p_j=[p_j,l . . . , p_j,R]^T, it has:

z_j^s=[m₁^S, . . . ,m_R^S]p_j  (2)

Finally, it is assumed that the set S is the union of two disjoint subsets C and T, where T is a union of K non-empty sets T_k such that T=U_k=1^KT_k where the index k represents the k^th type of abnormal uterine/endometrial phenotype. T_k's do not need to be disjoint. Moreover, T_k itself is the union of two disjoint sets D_k and V_k. Either D_k or V_k could be empty, but not both. It is assumed that for any plasma sample, including one from an abnormal individual, when restricted to CpG sites in C, its methylome can always be expressed as a weighted average of the restricted reference component methylomes. That is, it has: z_j^C=[m₁^C, . . . , m_R^C]p_j regardless whether j is from an abnormal individual. C is called the set of reference CpG sites. On the other hand, for a plasma sample l from an abnormal individual, when restricted to CpG sites in S=CUT, its methylome can no longer be expressed as a weighted average of the restricted reference component methylomes. That is, it has: w₁^S≠[m₁^S, . . . , m_R^S]p_l for any mixture vector p_l. More specifically, for a plasma sample l from an individual with the k^th type of abnormal phenotype, it has: 1), w_j^C=[m₁^C, . . . , m_R^C]p_l, 2), if D_K is non-empty, then w_lDK=[m_1,k^Dk, . . . , m_R,k^Dk]p_l such that [m₁^Dk, . . . , m_R^Dk]≠[m_1,k^Dk, . . . , m_R,k^Dk], and 3), if V_k is non-empty, then w_l^Vk=[m₁^Vk, . . . , m_R^Vk]q_l such that p_l≠q_l. In other words, in a plasma sample from the k^th type of abnormal individual, if the set D_k is not empty, the component methylomes of the sample l restricted to D_k are no longer the same as the reference component methylome restricted to D_k. If the set V_k is not empty, in this plasma sample, the proportion of the reference component methylomes restricted to V_k is no longer the same as the proportion of the reference component methylome restricted to R.

T is called the target set of CpG sites, D_k is called the differential methylation target set, V_k is called the copy number variation target set, and T_k is called the target set for the k^th type of abnormal phenotype.

The main steps of the algorithm of this Example are:

• • 1) Identify the sets of reference CpG sites C, and T₁, . . . , T_K for the list of K types of abnormal individuals.
  • 2) Estimate the restricted reference component methylomes m₁, . . . , m_R, or R predictor methylomes n₁, . . . , n_R that are independent linear combinations of the reference component methylomes such that n_r=[m₁, . . . , m_R]q_r for R linearly independent mixture vectors q₁, . . . , q_R.
  • 3) (Optional) If the reference component methylomes are available, estimate the proportions of these components at the reference CpG sites C for the test plasma samples.
  • 4) Predict the methylation level of the test plasma samples at the target set T_k of CpG sites, under the hypothesis that the sample is from a normal individual.
  • 5) Compare the predicted methylation levels at D_k and V_k against the observed methylation levels, and reject the null hypothesis that a test sample is from a normal individual if the observed methylation levels are significantly different form the predicted levels.

The algorithm of this Example can be implemented in a variety of ways. For example, given the methyl-seq data for a set of plasma samples from normal individuals, the presently disclosed EM algorithm or the data augmentation method can be applied to estimate the component methylomes, then use the maximum likelihood method to estimate the proportion of these component methylomes in the test sample. Below are exemplary simple implementations of the presently disclosed algorithm that use linear regression.

In the simplest implementation of the algorithm of this Example, it is assumed the restricted methylome of a plasma sample from a normal individual can be approximated by a mixture of two restricted reference methylomes. It is further assumed that the estimations of these two reference component methylomes are available. For example, in the implementation below, for the genomic loci of interest, the plasma methylome is approximated by the mixture of leukocyte and uterine/endometrial- or other relevant tissue/cell-derived methylomes. The implementation of the algorithm includes the following steps:

1. Identify the Reference Set C, and the Target Sets T₁, . . . , T_K.

• • 1.1 Collect the methylation data for a set of leukocyte samples, a set of uterine/endometrial- or other relevant tissue/cell-derived samples, and a set of plasma samples, all from normal individuals. For each type of abnormal individuals, collect a set of leukocyte-derived samples, a set of uterine/endometrial- or other relevant tissue/cell-derived samples, and a set of plasma samples from that type of abnormal individuals. All these samples should have matched age, race, and other relevant parameters. These are the training data.
  • 1.2 Let x_i,j be the observed methylation level of CpG site i in a normal leukocyte-derived sample j, and y_i,l the observed methylation level of CpG site i in a normal uterine/endometrial- or other relevant tissue/cell-derived sample l, s_x,i² the sample variance of x_i,j over all normal leukocyte-derived samples, s_y,i² the sample variance of y_i,j over all normal uterine/endometrial- or other relevant tissue/cell-derived samples. Identify the CpG sites S₀ such that for any i∈S₀, it has both s_x,i²<c₀ and s_y,i²<c₀ for some constant c₀. These are CpG sites with stable methylation levels in each type of normal cells.
  • 1.3 Let x_i,j be the observed methylation level of CpG site i in a leukocyte-derived sample j, including normal and abnormal, and y₀ the observed methylation level of CpG site i in a uterine/endometrial- or other relevant tissue/cell-derived sample 1, including normal and abnormal, s_x,i² the sample variance of x_i over all leukocyte-derived samples, including normal and abnormal, s_y,i² the sample variance of y_i,j over all uterine/endometrial- or other relevant tissue/cell-derived samples, including normal and abnormal. Identify the CpG sites S₁ such that for any i∈S_i, it has both s_x,i²<c₀ and s_y,i²<c₀ for some constant c₀, and that the statistical test for the difference between {x_i,j0: j0 is a normal leukocyte—derived sample} and {x_i,jk: jk is an abnormal leukocyte—derived sample of type k} is not significant for all abnormal types of leukocyte-derived, and that the statistical test for the difference between {y_i,j0: j0 is a normal uterine/endometrial—or other relevant tissue/cell—derived sample} and {y_i,jk: jk is an abnormal uterine/endometrial—or other relevant tissue/cell—derived sample of type k} is not significant for all abnormal types of uterine/endometrial- or other relevant tissue/cell-derived sample. These are CpG sites with stable methylation levels in each type of cells, and with no difference in methylation level between normal and any abnormal samples. Let x_i be the sample mean of x_i,j over all leukocyte-derived samples, including normal and abnormal, y_i the sample mean of y_i,j over all uterine/endometrial- or other relevant tissue/cell-derived samples, including normal and abnormal. Identify the subset C₀ of S₁ such that for any i∈C₀, it has |x_i−y_i|>c₁ for some constant c₁. These are CpG sites that are stably methylated in each cell type, with no difference between the normal and abnormal samples of the same cell type, and differentially methylated between different types of cells.
  • 1.4 Let x^R0 be the vector of x_i for all i∈C₀, and y^C0 be the vector of y_i for all i∈C₀, where x_i is the mean methylation at site i in all leukocyte-derived samples y_i the mean methylation at site i in all uterine/endometrial- or other relevant tissue/cell-derived samples. Note that by the way the set C₀ is selected, there is no difference in the methylation level of any CpG sites in C₀ between normal and abnormal leukocyte-derived samples, or between normal and abnormal uterine/endometrial- or other relevant tissue/cell-derived samples. Let z_j^C0 be the observed methylation levels of CpG sites in C₀ for a plasma sample j of the k^th abnormal type. (For convenience, the normal plasma sample is called as sample of the 0th abnormal type). For each sample j belonging to the k^th abnormal type, regress z_j^C0 against x^C0 and y^C0, with the constraints that the intercept must be 0, and the coefficients must be non-negative and add to 1, and get the residual e_j^C0. Identify the subset C₀^k of C₀ such that for any CpG i in C₀^k, it has

$\frac{e_{i,k}^2}{s_{i,k}}<c_2,$

• • and e_i,k²<c₃ for some constants c₂ and c₃, where e_i,k² is the mean of the squared difference between estimated and observed methylation levels of CpG site i in all plasma samples of the k^th abnormal type, and s_i,k² the sample variances of methylation levels of CpG site i in the same set of plasma samples. Repeat the above procedure for each type of abnormal plasma samples, the intersection of the subsets C=∩_k=0^K C₀^k is the reference set of CpG sites. These are CpG sites where their methylation levels in both normal and any type of abnormal plasma samples can be accurately predicted by the reference component methylomes from normal individuals.
  • 1.5 Let T₀=S₀\ S₁. Let x^C and x^T0 be the vectors of x_i and x_h for all i∈C and h∈T₀ respectively, and y^C and y^T0 be the vectors of y_i and y_h for all i∈C and h∈T₀ respectively, where x_i, x_h, y_i, and y_h are mean methylation level of sites for a normal leukocyte-derived or uterine/endometrial- or other relevant tissue/cell-derived sample at sites i and h respectively. Let z_j^C and z_j^T0 and be the observed methylation levels of CpG sites in C and T₀ respectively for a normal plasma sample j, w_lk^C and w_lk^T0 the observed methylation level of CpG sites in C and T₀ respectively for a plasma sample l_k from an individual with the k^th type of abnormality, w_lg^C and w_lg^T0 the observed methylation level of CpG sites in C and T₀ respectively for a plasma sample l_g from an individual with the g^th type of abnormality, where g≠k. For each j, l_k, and l_g, regress z_j^C, w_lk^C, and w_lg^C respectively against x^C and y^C, with the constraints that the intercept must be 0, and the coefficients must be non-negative and add to 1. Apply the fitted models respectively to x^T0 and y^T0 to predict z_j^T0, w_lk^T0 and w_lg^T0 respectively, and get the differences e_j^T0, e_lk^T0 and e_lg^T0 between the predicted values and observed values. Let e_i, e_i,k, and e_i,g be the means of the sets of differences {e_j^T0: j is a normal plasma sample}, {e_lk^T0: l_k is a plasma sample of th k^th abnormal type} and {e_lg^T0: l_g is a plasma sample of the g^th abnormal type} for CpG site i respectively. Identify the subset T_k of T₀ such that for any i∈T_k, it has |e_i|<c_2,0, |e_i,k|>c_2,k, and |e_i,k−e_i,g|>c_3,k, for some constants c_2,0, C_2,k, and C_3,k, for all g≠k. T_k is the target set for the k^th type of the abnormal individual. These are the sites where the methylation of a normal plasma sample can be accurately predicted, the observed methylation in a plasma sample of the k^th abnormal type will deviate from the prediction, and deviation will be different from that of a plasma sample of any other abnormal type.

2. Estimate Fraction of the New Stool Samples to be Tested

Recall that x^c and y^c are mean vectors of the methylation levels of the training leukocyte-derived and training uterine/endometrial- or other relevant tissue/cell-derived sample data for the CpG sites in the reference set C. For any new plasma sample t to be tested, let z_t^C be the observed methylation levels of CpG sites in C. Regress z_t^C against x^C and y^C, with the constraints that the intercept must be 0, and the coefficients must be non-negative and add to 1. The estimated coefficients are the estimated fractions of the component methylomes for the plasma sample t.

3. Test if the New Plasma Samples are from the k^th Type of Abnormal Individual.

For the new plasma sample t, let x^Tk and y^Tk be mean vectors of the methylation levels of the training leukocyte-derived and training uterine/endometrial- or other relevant tissue/cell-derived sample data for the CpG sites in the target set T_k identified in step 1 of this algorithm, apply the fitted regression models obtained from the step 2 of this algorithm to x^Tk and y^Tk to predict the methylation levels of CpG sites in T_k for sample t under the hypothesis that sample t is from a normal. Let n_k be the number of CpG sites in T_k. Define functions f_k(x₁, . . . , x_nk)=Σ_i(−1)^{I_(ei,k−ei)}x_i and f_k,g(x₁, . . . , x_nk)=Σ_i(−1)^{I_(ei,k−ei,g)}x_i, where I_(⋅)=I_(−∞,0)(⋅), that is, the indicator function for the interval (−∞, 0), e_i, e_i,k and e_i,g are estimations obtained from step 1.5 of the algorithm. It will be said the sample is from an individual with the k^th type of abnormal phenotype if f_k(e_1,t−e₁, . . . , e_nk,t−e_nk)>c_4,k, and f_k,g(e_1,t−e_1,g, . . . , e_nk,t−e_nk,g)>c_5,g for all g k, where e_i,t is the difference between the observed methylation level of the CpG site i∈T_k for sample t and the predicted value by the fitted model obtained from step 2, and g≠k is any type of abnormal phenotype that is different form the k^th type of abnormal phenotype.

Other ways of implementing the algorithm of this Example can be developed by modifying the simple implementation presented above. Specifically, it does not need to assume that there are only two component reference methylomes that make up the plasma methylomes, nor does it need to approximate them by mixtures of the component methylomes. Instead, a set of predictor methylomes can be collected that are themselves mixtures of component reference genomes, as long as the number of the predictor methylomes is the same as the number of the reference component methylomes, and the mixture vectors of the predictor methylomes are linearly independent. For example, they can be methylomes of plasma samples with known different proportion of leukocyte-derived and uterine/endometrial- or other relevant tissue/cell-derived sample DNAs.

In the algorithm of this Example, the difference between observed methylation levels in certain target regions and the predicted methylation levels as the test statistic to determine if in a plasma sample the methylome has been affected by endometriosis. To illustrate the advantage of this approach, it is assumed that the mixture vector p_i for the methylome of a normal plasma sample j followed a Dirichlet's distribution with parameters α₁= . . . =α_R. Furthermore, for CpG site i, its methylation levels in the R reference vector p_i for component methylomes are m_i,r=(r−1)/(R−1). It can be shown that the methylation level of i in sample j then has a mean of 0.5, and a variance of

$\frac{R+1}{12\left(R-1\right)\left(R{\alpha }_1+1\right)}.$

If there is a methyl-seq library of sample j with a coverage of N for CpG site i, the variance of the measured methylation level z_i,j is

${\sigma }_1^2=\frac{1}{4N}+\frac{N-1}{N}\frac{R+1}{12\left(R-1\right)\left(R{\alpha }_1+1\right)}.$

In other words, if z_i,j is used as a test statistic to detect and phenotype endometriosis using a plasma sample, under the null hypothesis, the test statistic has a variance of σ₁². However, in the presently disclosed algorithm, it is first estimated the mixture vector p_j, then predicted z_i,j by Σ_rm_i,rp_r,j. Note that in methyl-seq data, millions of CpG sites can be contained in each library, and that the variance of the coefficients in a linear regression model is inversely proportional to sample size. Thus it is possible to get highly accurate estimation of the mixture vector p_i, even if it is taken into account that adjacent CpG sites tend to have correlated methylation levels. Assuming an accurate estimate of Σ_rm_i,rp_r,j can be obtained, that is, the error of the estimation can be ignored, the variance of the difference z_i,j−Σ_rm_i,rp_r,j between the observed methylation level and the prediction will be

$\frac{1}{4N}-\frac{1}{N}\frac{R+1}{12\left(R-1\right)\left(R{\alpha }_1+1\right)}.$

In other words, under the null hypothesis, the test static z_i,j−Σ_rM_i,r p_r,j used in the presently disclosed algorithm has a much smaller variance than the other candidate test statistic z_i,j. This in turns means that the presently disclosed test will achieve a higher power at the same level of type I error.

Examples of Embodiments for Example 4

D1. A method for diagnosing, prognosing, classifying, and/or monitoring endometriosis in a mammal, comprising:

(a) obtaining a sample from the mammal;

(b) determining the methylation status and/or level of one or more genomic loci in the sample;

(c) comparing the methylation status and/or level of the one or more genomic loci to a reference or a set of predicted values; and

(d) diagnosing endometriosis in the mammal,

wherein the difference in the methylation status and/or level of the one or more genomic loci in the sample compared to the reference or a set of predicted values indicates the presence of endometriosis in the mammal.

D2. The method of embodiment D1, wherein an increase in the level of methylation of the one or more genomic loci in the sample indicates the presence of endometriosis in the mammal or a decrease in the level of methylation of the one or more genomic loci in the sample indicates the presence of endometriosis in the mammal.
D3. The method of embodiment D1, wherein a decrease in the level of methylation of at least one of the one or more genomic loci in the sample and an increase in the level of methylation of at least one of the one or more genomic loci in the sample indicates the presence of endometriosis in the mammal.
D4. A method of treating endometriosis in a mammal, comprising:

(a) obtaining a sample from the mammal;

(b) determining the methylation status and/or level of one or more genomic loci present in the sample;

(c) comparing the methylation status and/or level of the one or more genomic loci to a reference or a set of predicted values;

(d) diagnosing endometriosis in the mammal, wherein the difference in the methylation status and/or level of the one or more genomic loci in the sample compared to the reference indicates the presence of endometriosis in the mammal; and

(e) administering a hormone therapy, a pain medication, or both to said mammal.

D5. The method of any one of embodiments D1-D4, wherein the reference is the methylation status and/or level of the one or more genomic loci in a sample obtained from a mammal that does not have endometriosis.
D6. The method of any one of embodiments D1-D5, wherein said sample is a plasma sample.
D7. A method of treating endometriosis comprising:

(a) measuring the methylation status and/or level of one or more genomic loci present in a sample from a mammal prior to a treatment of endometriosis;

(b) measuring the methylation status and/or level of one or more genomic loci present in a sample from the mammal during the treatment of endometriosis; and

(c) continuing the treatment if the difference in the methylation status and/or level of the one or more genomic loci between the samples from prior to and during the treatment of endometriosis indicates the mammal is responsive to the treatment.

D8. The method of embodiment D7, further comprising (d) administering a different treatment to the mammal if the difference in the methylation status and/or level of the one or more genomic loci between the samples from prior to and during the treatment of endometriosis indicates the mammal is not responsive to the treatment.
D9. The method of embodiment D7, wherein an increase in the level of methylation of the one or more genomic loci in the sample indicates the mammal is responsive to the treatment, or a decrease in the level of methylation of the one or more genomic loci in the sample indicates the mammal is responsive to the treatment.
D10. The method of embodiment D7, wherein a decrease in the level of methylation of at least one of the one or more genomic loci in the sample and an increase in the level of methylation of at least one of the one or more genomic loci in the sample indicates the mammal is responsive to the treatment.
D11. The method of embodiment D7, wherein an increase in the level of methylation of the one or more genomic loci in the sample indicates the mammal is not responsive to the treatment, or a decrease in the level of methylation of the one or more genomic loci in the sample indicates the mammal is not responsive to the treatment.
D12. The method of embodiment D7, wherein a decrease in the level of methylation of at least one of the one or more genomic loci in the sample and an increase in the level of methylation of at least one of the one or more genomic loci in the sample indicates the mammal is not responsive to the treatment.
D13. The method of any one of embodiments D7-D12, wherein said sample is a plasma sample.
D14. The method of any one of embodiments D1-D13, wherein the one or more genomic loci comprise one or more CpG sites.
D15. The method of any one of embodiments D1-D14, wherein the one or more genomic loci are present within nucleic acids isolated from the sample.
D16. The method of any one of embodiments D1-D15, wherein the one or more genomic loci are present within cell-free nucleic acids isolated from the sample.
D17. A method of treating endometriosis, comprising;

(a) diagnosing endometriosis in a mammal by utilization of the algorithm disclosed in Example Embodiment H; and

(b) administering a hormone therapy, a pain medication, or both to said mammal to treat said endometriosis.

D18. The method of any one of embodiments D1-D17, wherein said mammal is a human.
D19. A kit for diagnosing, prognosing, and/or monitoring endometriosis in a mammal comprising a means for determining and/or detecting the methylation status of one or more genomic loci.
D20. The kit of embodiment D19, wherein the means comprises one or more primers and/or probes for determining and/or detecting the methylation status of the one or more genomic loci.

Abstract of this Example (Example 4)

This Example provides methods for diagnosing, prognosing, monitoring, classifying, and/or treating endometriosis. For example, algorithms, kits, and methods for diagnosing, prognosing, monitoring, classifying, and/or treating endometriosis are provided.

Example 5—Methods and Materials for Assessing and Treating Necrotizing Enterocolitis

Field of this Example

This Example relates document relates to methods and materials involved assessing a mammal (e.g., a human infant) for and/or treating a mammal (e.g., human infant) having or developing necrotizing enterocolitis. For example, this Example provides methods and materials for using DNA methylation profiles of nucleic acid within a biopsy (e.g., a plasma sample, a blood sample, or a stool sample) to determine whether or not a mammal has, or is developing, necrotizing enterocolitis. This Example also provides methods, algorithms, and kits for diagnosing, prognosing, monitoring, classifying, and/or treating necrotizing enterocolitis.

Background of this Example

Necrotizing enterocolitis affects low birth weight infants in the first weeks of life with a reported frequency of between 1 and 5 percent of neonatal intensive care unit admissions with mortality rates of between 15 and 30 percent. Necrotizing enterocolitis currently is suspected through the visualization of distended abdomen and confirmed via x-ray. By the time final diagnosis occurs through this process, however, the disease has already progressed to the point that typically requires surgical intervention, which carries a 50 percent mortality rate.

Summary of this Example

This Example provides methods and materials involved in assessing a mammal (e.g., a human) for and/or treating a mammal (e.g., human) having or developing necrotizing enterocolitis. For example, this Example provides methods and materials for using DNA methylation profiles of nucleic acid within a biopsy (e.g., a plasma sample, a blood sample, or a stool sample) to determine whether or not a mammal has, or is developing, necrotizing enterocolitis. Determining if a mammal (e.g., a human) has, or is likely to develop, necrotizing enterocolitis by assessing DNA methylation profiles of nucleic acid within a biopsy (e.g., a plasma sample, a blood sample, or a stool sample) can aid in the identification of mammals (e.g., humans) that should be treated in a particular manner (e.g., by administering an antibiotic therapy, by feeding intravenously as opposed to by mouth, and/or by performing a surgical treatment), for example, early in the disease process.

This Example also provides methods for diagnosing, prognosing, monitoring, classifying, and/or treating necrotizing enterocolitis. For example, the methods described herein can include determining the methylation status of one or more genomic loci in a sample (e.g., a plasma sample, blood sample, or a stool sample) of a mammal (e.g., a human infant). This Example further provides algorithms and kits for diagnosing, prognosing, monitoring, classifying, and/or treating necrotizing enterocolitis.

In one aspect, the present disclosure provides a method for diagnosing, prognosing, classifying, and/or monitoring necrotizing enterocolitis in a mammal (e.g., a human infant) comprising: (a) obtaining a sample (e.g., a plasma sample, a blood sample, or a stool sample) from the mammal; (b) determining the methylation status and/or level of one or more genomic loci in the sample; (c) comparing the methylation status and/or level of the one or more genomic loci to a reference or a set of predicted values; and (d) identifying the mammal as having necrotizing enterocolitis, wherein the difference in the methylation status and/or level of the one or more genomic loci in the sample compared to the reference or a set of predicted values indicates the presence of necrotizing enterocolitis in the mammal.

In another aspect, this Example provides a method of treating necrotizing enterocolitis in a mammal (e.g., a human infant) comprising: (a) obtaining a sample (e.g., a plasma sample, a blood sample, or a stool sample) from the mammal; (b) determining the methylation status and/or level of one or more genomic loci present in the sample; (c) comparing the methylation status and/or level of the one or more genomic loci to a reference or a set of predicted values; (d) identifying the mammal as having necrotizing enterocolitis, wherein the difference in the methylation status and/or level of the one or more genomic loci in the sample compared to the reference or a set of predicted values indicates the presence of necrotizing enterocolitis in the mammal; and (e) treating the mammal by administering an antibiotic therapy, by feeding intravenously as opposed to by mouth, and/or by performing a surgical treatment.

In some cases, an increase in the level of methylation of the one or more genomic loci in a sample (e.g., a plasma sample, a blood sample, or a stool sample) can indicate the presence of necrotizing enterocolitis in the mammal or a decrease in the level of methylation of the one or more genomic loci in a sample (e.g., a plasma sample, a blood sample, or a stool sample) indicates the presence of the necrotizing enterocolitis in the mammal. In some cases, a decrease in the level of methylation of at least one of the one or more genomic loci in a sample and an increase in the level of methylation of at least one of the one or more genomic loci in the sample can indicate the presence of necrotizing enterocolitis in the mammal.

In some cases, the reference can be the methylation status and/or level of the one or more genomic loci in a sample (e.g., a plasma sample, a blood sample, or a stool sample) obtained from a mammal (e.g., a human infant) that does not have necrotizing enterocolitis.

In another aspect, this Example provides a method of treating necrotizing enterocolitis in a mammal (e.g., a human infant) comprising: (a) measuring the methylation status and/or level of one or more genomic loci present in a sample (e.g., a plasma sample, a blood sample, or a stool sample) from the mammal prior to a treatment of necrotizing enterocolitis; (b) measuring the methylation status and/or level of one or more genomic loci present in a sample (e.g., a plasma sample, a blood sample, or a stool sample) from the mammal during the treatment of necrotizing enterocolitis; and (c) continuing the treatment if the difference in the methylation status and/or level of the one or more genomic loci between the samples from prior to and during the treatment of necrotizing enterocolitis indicates the subject is responsive to the treatment. In some cases, the method further comprises (d) administering a different treatment to the mammal if the difference in the methylation status and/or level of the one or more genomic loci between the samples from prior to and during the treatment of necrotizing enterocolitis indicates the subject is not responsive to the treatment.

In some cases, an increase in the level of methylation of the one or more genomic loci in the sample indicates that the mammal is responsive to the treatment, or a decrease in the level of methylation of the one or more genomic loci in the sample indicates the mammal is responsive to the treatment. In certain embodiments, a decrease in the level of methylation of at least one of the one or more genomic loci in the sample and an increase in the level of methylation of at least one of the one or more genomic loci in the sample indicates the mammal is responsive to the treatment. In certain embodiments, an increase in the level of methylation of the one or more genomic loci in the sample indicates the mammal is responsive to the treatment, or a decrease in the level of methylation of the one or more genomic loci in the sample indicates the mammal is not responsive to the treatment. In certain embodiments, a decrease in the level of methylation of at least one of the one or more genomic loci in the sample and an increase in the level of methylation of at least one of the one or more genomic loci in the sample indicates the subject is not responsive to the treatment.

This Example further provides algorithms for diagnosing and/or monitoring a mammal having necrotizing enterocolitis. In certain embodiments, the algorithm can be used to classify necrotizing enterocolitis of a mammal (e.g., a human infant).

In another aspect, this Example provides a kit for diagnosing, prognosing, and/or monitoring necrotizing enterocolitis in a mammal comprising a means for determining and/or detecting the methylation status of one or more genomic loci. In certain embodiments, the means comprises one or more primers and/or probes for determining and/or detecting the methylation status of the one or more genomic loci.

In certain embodiments, the one or more genomic loci are present within nucleic acids isolated from the sample. In certain embodiments, the one or more genomic loci are present within cell-free nucleic acids isolated from the sample.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Although methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, suitable methods and materials are described below. All publications, patent applications, patents, and other references mentioned herein are incorporated by reference in their entirety. In case of conflict, the present specification, including definitions, will control. In addition, the materials, methods, and examples are illustrative only and not intended to be limiting.

Other features and advantages of the invention will be apparent from the following detailed description, and from the claims.

Description of this Example

This Example provides methods for diagnosing, prognosing, monitoring, classifying, and/or treating necrotizing enterocolitis. For example, the methods described herein can include determining the methylation status of one or more genomic loci in a sample of a mammal (e.g., a human infant). In some cases, the methods described herein can include the use of an algorithm to diagnose, prognose, monitor, classify, and/or assist in the treatment of necrotizing enterocolitis. Non-limiting embodiments of this Example are described by the present specification and Examples.

Unless defined otherwise, all technical and scientific terms used in this Example generally have their ordinary meanings in the art, within the context of this Example and in the specific context where each term is used. The following references provide one of skill with a general definition of many of the terms used in this Example: Singleton et al., Dictionary of Microbiology and Molecular Biology (2nd ed. 1994); The Cambridge Dictionary of Science and Technology (Walker ed., 1988); The Glossary of Genetics, 5th Ed., R. Rieger et al. (eds.), Springer Verlag (1991); and Hale & Marham, The Harper Collins Dictionary of Biology (1991). Certain terms are discussed below, or elsewhere in the specification, to provide additional guidance to the practitioner in describing the compositions and methods of this Example and how to make and use them.

The terms “comprise(s),” “include(s),” “having,” “has,” “can,” “contain(s),” and variants thereof, as used herein, are intended to be open-ended transitional phrases, terms or words that do not preclude the possibility of additional acts or structures. The singular forms “a,” “an,” and “the” include plural references unless the context clearly dictates otherwise. The present Example also contemplates other embodiments “comprising,” “consisting of” and “consisting essentially of,” the embodiments or elements presented herein, whether explicitly set forth or not.

As used herein, the use of the word “a” or “an” when used in conjunction with the term “comprising” in the claims and/or the specification may mean “one,” but it is also consistent with the meaning of “one or more,” “at least one,” and “one or more than one.” Still further, the terms “having,” “including,” “containing,” and “comprising” are interchangeable, and one of skill in the art is cognizant that these terms are open ended terms.

The term “about” or “approximately” means within an acceptable error range for the particular value as determined by one of ordinary skill in the art, which will depend in part on how the value is measured or determined, i.e., the limitations of the measurement system. For example, “about” can mean within 3 or more than 3 standard deviations, per the practice in the art. Alternatively, “about” can mean a range of up to 20%, preferably up to 10%, more preferably up to 5%, and more preferably still up to 1% of a given value. Alternatively, particularly with respect to biological systems or processes, the term can mean within an order of magnitude, preferably within 5-fold, and more preferably within 2-fold, of a value.

In certain embodiments, the term “biomarker” refers to a marker (e.g., DNA methylation status) that allows detection of a disease and/or disorder in an individual, including detection of the disease or the disorder in its early stages. In certain embodiments, the term “biomarker” refers to a marker (e.g., DNA methylation status) that allows the characterization of a phenotype of a disease and/or a disorder in an individual. Early stage of a disease, as used herein, refers to the time period between the onset of the disease and the time point that signs or symptoms of the disease emerge. In certain non-limiting embodiments, the presence, absence, and/or level of a biomarker in a sample of a mammal (e.g., a human) is determined by comparing to a reference control.

The terms “reference sample,” “reference control,” “control,” or “reference,” as used interchangeably herein, refers to a control for a methylation status of a genomic locus that is to be detected in a sample of a mammal. In certain embodiments, a reference sample can be a sample from a healthy individual, e.g., an individual that does not have necrotizing enterocolitis. In certain embodiments, a reference sample can be a sample from a control individual that does not have the disease or phenotype to be detected by a biomarker disclosed herein. In certain embodiments, a control or reference can be the presence, absence, and/or a particular level of a methylation state of a genomic locus in a healthy individual. In certain embodiments, a reference can be a predetermined presence, absence, and/or particular level of a methylation state of a genomic locus that indicates a subject does not have necrotizing enterocolitis. In certain embodiments, a reference can be the methylation status of a locus in an individual having a disease or a phenotype, e.g., an individual that has necrotizing enterocolitis, where the methylation status of the locus is known to be not associated with the disease or the phenotype.

The term “a set of predicted values” refers to the methylation status of certain genomic loci for a sample. The status of those loci is not directly measured from that sample. Rather, it is inferred from measurements of other loci for that sample and/or measurements of other samples. The inference of the predicted values is based on mathematical/statistical models. The models are designed under the null hypothesis that the sample for which the methylation status of those loci is to be predicted has a normal phenotype.

The term “slightly invasive or non-invasive method” refers to a method that does not involve the removal of tissues by biopsy from the intestinal tract. In certain embodiments, slightly invasive or non-invasive methods, as described herein, include obtaining plasma or stool from a subject.

The term “patient” or “subject,” as used interchangeably herein, refers to any warm-blooded animal, e.g., human or non-human. Non-limiting examples of non-human subjects include mammals, non-human primates, dogs, cats, mice, rats, guinea pigs, rabbits, fowl, pigs, horses, cows, goats, sheep, etc. In certain embodiments, the subject is human.

The term “nucleic acid,” “nucleic acid molecule,” or “polynucleotide” includes any compound and/or substance that comprises a polymer of nucleotides. Each nucleotide is composed of a base, specifically a purine- or pyrimidine base (i.e., cytosine (C), guanine (G), adenine (A), thymine (T), or uracil (U)), a sugar (i.e., deoxyribose or ribose), and a phosphate group. In certain embodiments, the nucleic acid molecule is described by the sequence of bases, whereby said bases represent the primary structure (linear structure) of a nucleic acid molecule. The sequence of bases is typically represented from 5′ to 3′. These terms encompass deoxyribonucleic acid (DNA) including, e.g., complementary DNA (cDNA) and genomic DNA, ribonucleic acid (RNA), in particular messenger RNA (mRNA), synthetic forms of DNA or RNA, and mixed polymers comprising two or more of these molecules. The herein described nucleic acid molecule can contain naturally occurring or non-naturally occurring nucleotides. Examples of non-naturally occurring nucleotides include modified nucleotide bases with derivatized sugars or phosphate backbone linkages or chemically modified residues.

The term “isolated” (e.g., isolated genomic DNA) refers to a biological component that has been substantially separated or purified away from other biological components in the cell of the organism in which the component naturally occurs, e.g., other chromosomal and extra-chromosomal DNA and RNA, proteins, and organelles. Nucleic acids, e.g., DNA, that have been “isolated” include nucleic acids purified by standard purification methods.

The term “genomic locus” or “genomic DNA locus,” as used herein, refers to any fixed position in a genome. For example, a genomic locus can refer to a genomic element, a chromosomal region, a gene, a region of a gene, e.g., an exon or intron, a regulatory region of a gene, e.g., a promoter or enhancer, a CpG site, a CpG island, or a CpG island shore. For example, a genomic locus can include one or more CpG sites, e.g., between about 1 to about 100 CpG sites. In certain embodiments, a genomic locus can be of any particular length, e.g., between about 1 to about 10,000 nucleotides in length.

As used interchangeably herein, “methylation state,” “methylation profile,” “methylation status,” and “methylation level” refer to the presence, absence, percentage, and/or quantity of methylation at a particular nucleotide, or nucleotides, within a DNA region, e.g., a genomic locus. The methylation status of a particular DNA sequence (e.g., a genomic locus) can indicate the methylation state of every nucleotide in the sequence, indicate the methylation state of any of the nucleotides (e.g., cytosines) in the sequence, can indicate the methylation state of a subset of the nucleotides (e.g., of cytosines), can indicate the percentage or fraction of methylated cytosines at any particular stretch of nucleotides within the sequence or can indicate the average rate of methylation of all the cytosines (or a subset of the cytosines) present in a nucleic acid.

As used herein, a “methylated nucleic acid molecule” refers to a nucleic acid molecule that contains one or more methylated nucleotides that is/are methylated.

As used herein, a “CpG site” or “methylation site” is a nucleotide within a nucleic acid that is susceptible to methylation either by natural occurring events in vivo or by an event instituted to chemically methylate the nucleotide in vitro.

A “CpG island,” as used herein, describes a segment of a nucleic acid, e.g., DNA sequence, that have a high frequency of CpG dinucleotide repeats. See, e.g., Illingworth and Bird, FEBS Letters, 2009; 583:1713-1720. For example, Yamada et al. (Genome Research, 2004; 14:247-266) have described a set of standards for determining a CpG island: it must be at least 400 nucleotides in length, has a GC content greater than 50%, and an OCF/ECF ratio greater than 0.6. Others (Takai et al., Proc. Natl. Acad. Sci. U.S.A., 2002; 99:3740-3745) have defined a CpG island less stringently as a sequence of at least 200 nucleotides in length, having a greater than 50% GC content and an OCF/ECF ratio greater than 0.6.

A “CpG island shore,” as used herein, refers to methylation hotspots that are present a short distance, e.g., less than 2 kb, from CpG islands.

The term “methylome,” as used herein, refers to the amount or pattern of methylation at different sites or regions within a population of cells. The methylome can correspond to all of the genome, a subset of the genome (e.g., repeat elements in the genome), or a portion of the subset (e.g., those areas found to be associated with necrotizing enterocolitis). A methylome from plasma can be referred to a “plasma fluid methylome,” or a “plasma fluid DNA methylome.” The plasma fluid methylome is an example of a cell-free methylome that includes cell-free DNA (cfDNA). A methylome from stool can be referred to a “stool fluid methylome,” or a “stool fluid DNA methylome.” The stool fluid methylome is an example of a cell-free methylome that includes cell-free DNA (cfDNA).

As used herein, the term “increase” refers to alter positively by at least about 2%, including, but not limited to, alter positively by about 5%, by about 10%, by about 15%, by about 20%, by about 25%, by about 30%, by about 35%, by about 40%, by about 45%, by about 50%, by about 55%, by about 60%, by about 65%, by about 70%, by about 75%, by about 80%, by about 85%, by about 90%, by about 95% or by about 100%.

As used herein, the terms “reduce,” “reduction,” or “decrease” refers to alter negatively by at least about 2%, including, but not limited to, alter negatively by about 5%, by about 10%, by about 15%, by about 20%, by about 25%, by about 30%, by about 35%, by about 40%, by about 45%, by about 50%, by about 55%, by about 60%, by about 65%, by about 70%, by about 75%, by about 80%, by about 85%, by about 90%, by about 95% or by about 100%.

As described herein, this Example provides methods for diagnosing, monitoring, classifying, and/or treating necrotizing enterocolitis by analyzing the methylation status of one or more genomic loci in a sample (e.g., a plasma sample or a stool sample) of a mammal (e.g., a human infant). In certain embodiments, the methods can include using an algorithm described herein. In certain embodiments, the methods described herein can allow for the early diagnosis or screening of a subject with necrotizing enterocolitis, e.g., the subject does not have any symptoms, or only have early symptoms of necrotizing enterocolitis.

In certain embodiments, samples obtained for use in the methods described herein can include cfDNA, which carries DNA methylation information from the cell of origin. cfDNA can arise from cellular apoptosis and necrosis, and can be generated from active secretory processes, with the formation of extracellular vesicles. DNA signatures are highly tissue-specific, and include in vivo information relating to the tissue source of cfDNA. In certain embodiments, the methods described herein can include analyzing cfDNA in a sample (e.g., a plasma sample or a stool sample), to identify genetic phenotypes that are drivers and/or consequences of necrotizing enterocolitis.

The sample from the subject can be collected using any appropriate technique. For example, a blood sample, a plasma sample, or a stool sample can be collected using standard methods. In some cases, the sample can be collected from the subject before the subject has any symptom of necrotizing enterocolitis, i.e., a non-symptomatic subject. In certain embodiments, the sample can be collected from the non-symptomatic subject who is at high risk of necrotizing enterocolitis (e.g., a preterm baby). In certain embodiments, the sample can be collected from the subject who has previously received or is currently receiving a treatment for necrotizing enterocolitis. In certain embodiments, two or more samples (e.g., two or more, three or more, four or more, five or more, six or more or seven or more samples) can be obtained before and during the subject is receiving a treatment for necrotizing enterocolitis (e.g., serially obtained samples).

Diagnostic, Prognostic, Classification, and Monitoring Methods of this Example

This Example provides diagnostic and prognostic methods for diseases and/or disorders that are characterized by differential methylation of genomic loci. For example, this Example provides methods for diagnosing, prognosing, classifying, and/or monitoring necrotizing enterocolitis in a subject that includes analyzing the methylation status of certain genomic loci.

In certain embodiments, the analyzed genomic loci can include one or more genomic loci that exhibit differential methylation in a sample from a subject that has necrotizing enterocolitis compared to a reference sample. For example, the methods described herein can include assessing the methylation status of one or more genomic loci, e.g., about 5 or more, about 10 or more, about 50 or more, about 100 or more, about 500 or more, about 1,000 or more, about 5,000 or more, about 10,000 or more, about 25,000 or more, about 50,000 or more or about 100,000 or more genomic loci in a sample of a subject. In certain embodiments, the genomic loci can be selected from the genes, or a region within the genes, provided in Example Embodiment I. In certain embodiments, the one or more genomic loci can be one or more promoter regions of one or more genes, one or more exons of one or more genes, one or more introns of one or more genes, one or more CpG sites, one or more CpG islands, one or more CpG island shores, one or more enhancers of one or more genes, or a combination thereof. In certain embodiments, the genomic loci are present on a particular chromosome.

In certain embodiments, this Example provides methods for diagnosing, prognosing, and/or monitoring necrotizing enterocolitis in a subject by detecting the DNA methylation profiles associated with necrotizing enterocolitis. In certain embodiments, the methods described herein can include (a) obtaining a sample from the subject, (b) determining the methylation status of one or more genomic loci present in the sample, e.g., present within cfDNA in a plasma sample or a stool sample, (c) comparing the methylation status of the one or more genomic loci to a reference or a set of predicted values, and (d) diagnosing necrotizing enterocolitis in the subject. In certain embodiments, the difference in the methylation status of the one or more genomic loci in the sample compared to the reference or a set of predicted values indicates the presence of necrotizing enterocolitis in the subject. In certain embodiments, the difference in the methylation status also can indicate the severity of necrotizing enterocolitis.

In certain embodiments, the methods described herein for diagnosing, prognosing, and/or monitoring necrotizing enterocolitis in a subject can include (a) obtaining a sample from the subject, (b) determining the level of methylation of one or more genomic loci present in the sample, (c) comparing the level of methylation of the one or more genomic loci to a reference or a set of predicted values, and (d) diagnosing necrotizing enterocolitis in the subject. In certain embodiments, the difference in the level of methylation of the one or more genomic loci in the sample compared to the reference or a set of predicted values indicates the presence of necrotizing enterocolitis in the subject. In certain embodiments, the difference in the methylation level also can indicate the severity of necrotizing enterocolitis.

In certain embodiments, diagnosing necrotizing enterocolitis in the subject can include characterizing a phenotype of the necrotizing enterocolitis, wherein the difference in the methylation status of the one or more genomic loci in the sample compared to the reference or a set of predicted values indicates the phenotype of the necrotizing enterocolitis. In certain embodiment, the phenotype of the necrotizing enterocolitis can include the severity of the necrotizing enterocolitis, prognosis of the necrotizing enterocolitis, molecular expression profile of the necrotizing enterocolitis, responsiveness of the necrotizing enterocolitis to certain treatments, or any combinations thereof.

In certain embodiments, the methods described herein for determining if a subject is at risk of developing necrotizing enterocolitis in the subject can include (a) obtaining a sample from the subject, (b) determining the level of methylation of one or more genomic loci present in the sample, (c) comparing the level of methylation of the one or more genomic loci to a reference or a set of predicted values, and (d) determining that the subject is at risk of developing necrotizing enterocolitis, wherein the difference in the level of methylation of the one or more genomic loci in the sample compared to the reference or a set of predicted values indicates that the subject is at risk.

In certain embodiments, diagnosing, prognosing, and/or monitoring of a subject with necrotizing enterocolitis can be based on a higher or lower methylation level of the genomic locus in the sample of the subject relative to the methylation level in a reference sample, e.g., a sample from a subject that does not have necrotizing enterocolitis. In certain embodiments, a difference of greater than about 5%, greater than about 10%, greater than about 15%, greater than about 20%, greater than about 25%, greater than about 30%, greater than about 35%, greater than about 40%, greater than about 45%, greater than about 50%, greater than about 55%, greater than about 60%, greater than about 65%, greater than about 70%, greater than about 75%, greater than about 80%, greater than about 85%, greater than about 90% or greater than about 95% in the methylation (e.g., level, percentage and/or fraction) of the one or more genomic loci in a sample obtained from a subject compared to a control can be indicative that the subject has necrotizing enterocolitis or is at risk of developing necrotizing enterocolitis. In certain embodiments, the difference can be a decrease in methylation (e.g., level, percentage, and/or fraction) of the genomic loci in the sample of the subject. Alternatively, the difference can be an increase in methylation (e.g., level, percentage, and/or fraction) of the genomic loci in the sample of the subject. In certain embodiments, the difference can be a decrease in the methylation of a genomic locus and an increase in the methylation of a different genomic locus in the sample obtained from the subject. In certain embodiments, a decrease in the level of methylation of one or more genomic loci in the sample and the increase in the level of methylation of one or more different genomic loci in the sample can indicate the presence of necrotizing enterocolitis.

In certain embodiments, diagnosis of a subject with necrotizing enterocolitis can be based on the methylated or unmethylated state of a genomic locus, e.g., a CpG site. In certain embodiments, a genomic locus, e.g., a CpG site, in a sample from a subject diagnosed with necrotizing enterocolitis can be methylated and the genomic locus, e.g., the CpG site, in a reference sample can be unmethylated. In certain embodiments, a genomic locus in a sample from a subject diagnosed with necrotizing enterocolitis can be unmethylated and the genomic locus in a reference sample can be methylated.

Diagnostic, Prognostic, Classification, and Monitoring Methods Using an Algorithm of this Example

This Example also provides diagnostic and prognostic methods for diseases and/or disorders that are characterized by differential methylation of genomic loci by using an algorithm as described, for example, in Example Embodiment J. For example, this Example provides methods for diagnosing, prognosing, classifying, and/or monitoring necrotizing enterocolitis in a subject that includes analyzing the methylation status of certain genomic loci and/or genomic fractions.

Methods for Treating Necrotizing Enterocolitis

This Example also provides methods for treating a subject having necrotizing enterocolitis. For example, a mammal (e.g., a human infant) that was identified as having necrotizing enterocolitis as described herein (or identified as being at risk of developing necrotizing enterocolitis as described herein) can be administered one or more antibiotic therapies, can be fed intravenously as opposed to feeding by mouth, or a combination thereof to treat necrotizing enterocolitis. Examples of antibiotic therapies that can be used as described herein include, without limitation, ampicillin, gentamycin, vancomycin, cefepime, and metronidazole. In some cases, a mammal (e.g., a human infant) that was identified as having necrotizing enterocolitis as described herein (or identified as being at risk of developing necrotizing enterocolitis as described herein) can be treated using a surgical procedure to treat necrotizing enterocolitis. Examples of surgical procedures that can be used as described herein include, without limitation, exploratory laparotomy, resection of affected intestine and creation of stoma and reanastamosis.

In some cases, the information provided by the methods described herein can be used by a clinician or physician in determining the most effective course of treatment (e.g., preventative or therapeutic) for the subject. A course of treatment refers to the measures taken for a patient after the prognosis or the assessment of increased risk for development of necrotizing enterocolitis is made. For example, when a subject is identified to have an increased risk of developing necrotizing enterocolitis, the physician can determine whether frequent monitoring of DNA methylation changes can be performed as a prophylactic measure. Also, when a subject is diagnosed with necrotizing enterocolitis (e.g., based on the presence of a DNA methylation pattern in a sample from a subject), it can be advantageous to follow such detection with a therapeutic treatment.

In some cases, this Example provides methods for assessing the efficacy of a therapeutic or prophylactic therapy for treating necrotizing enterocolitis in a subject, comprising determining the methylation status of one or more genomic loci present in a sample obtained from a subject prior to the therapy and determining methylation status of the one or more genomic loci present in a sample obtained from the subject at one or more time points during the therapeutic or prophylactic therapy, wherein the therapy is efficacious for treating necrotizing enterocolitis in a subject when there is a change in the presence and/or level of methylation of the one or more genomic loci in the second or subsequent samples, relative to the first sample. In certain embodiments, the first sample is obtained after therapeutic treatment has begun.

In certain embodiments, the methods for monitoring the response in a subject to prophylactic or therapeutic treatment of necrotizing enterocolitis can include measuring the methylation status and/or level of one or more genomic loci in a sample of a subject at a first time-point, administering a therapeutic agent, re-measuring the methylation status and/or level of the one or more genomic loci at a second time-point, comparing the results of the first and second measurements and optionally modifying the treatment regimen based on the comparison. In certain embodiments, the first time-point can be prior to an administration of the therapeutic agent, and the second time-point can be after said administration of the therapeutic agent. In certain embodiments, the first time-point can be prior to the administration of the therapeutic agent to the subject for the first time. In certain embodiments, the dose (defined as the quantity of therapeutic agent administered at any one administration) can be increased or decreased in response to the comparison. In certain embodiments, the dosing interval (defined as the time between successive administrations) can be increased or decreased in response to the comparison, including total discontinuation of treatment. In addition, the methods described herein can be used to determine the efficacy of the therapeutic treatment, wherein a change in the methylation status of certain genomic loci present in a sample of a subject can indicate that the therapeutic treatment regimen can be altered, reduced, and/or stopped.

Assays of this Example

This Example also provides assays and/or methods for determining the DNA methylation status and/or level of genomic loci that correlates with the presence, absence, and/or severity of necrotizing enterocolitis. In some cases, the assay method can include comparing the methylation status and/or level of genomic loci present in a sample from a subject that has necrotizing enterocolitis to the methylation status and/or level of genomic loci in a sample from a healthy subject to determine the methylation pattern, as described above, that correlates with the presence of necrotizing enterocolitis. In some cases, the assay methods can include comparing the methylation status and/or level of genomic loci in a sample from a subject that has necrotizing enterocolitis at an early stage to the methylation status and/or level of genomic loci in a sample from a subject that has necrotizing enterocolitis at a late stage to determine the methylation status and/or level that correlates with the different stages and/or severity of necrotizing enterocolitis.

DNA Isolation Techniques of this Example

In certain embodiments, the methods described herein can include isolating nucleic acid from a sample (e.g., a plasma sample or a stool sample) obtained from a subject. Any appropriate technique can be used to isolate nucleic acids from a sample. For example, isolation of DNA from a plasma sample can be performed by extraction methods using organic solvents such as a mixture of phenol and chloroform, followed by precipitation with ethanol (see, for example, J. Sambrook et al., “Molecular Cloning: A Laboratory Manual,” 1989, 2nd Ed., Cold Spring Harbor Laboratory Press: New York, N.Y.). Additional non-limiting examples include salting out DNA extraction (see, for example, P. Sunnucks et al., Genetics, 1996, 144:747-756; and S. M. Aljanabi and I. Martinez, Nucl. Acids Res. 1997, 25:4692-4693), the trimethylammonium bromide salts DNA extraction method (see, for example, S. Gustincich et al., BioTechniques, 1991, 11:298-302), and the guanidinium thiocyanate DNA extraction method (see, for example, J. B. W. Hammond et al., Biochemistry, 1996, 240:298-300). There are also numerous commercially available kits that can be used to extract DNA from biological fluids (e.g., plasma samples) or cells. For example, Qiagen's Gentra PureGene Cell Kit, QlAamp Circulating Nucleic Acid Kit, QiaAmp DNA Mini Kit, DNeasy Blood and Tissue Kit or QiaAmp DNA Blood Mini Kit (Qiagen, Hilden, Germany), GenomicPrep™ Blood DNA Isolation Kit (Promega, Madison, Wis.) and GFX™ Genomic Blood DNA Purification Kit (Amersham, Piscataway, N.J.) can be used to obtain DNA from a sample from a subject.

Methylation Detection Techniques of this Example

Various methylation analysis procedures are known in the art, and can be used with the methods described herein. These assays allow for determination of the methylation state of one genomic locus, e.g., one or more CpG sites or islands within a nucleic acid obtained from a sample. In addition, the methods can be used to quantify the methylation of a genomic locus. Such assays involve, among other techniques, DNA sequencing of bisulfite-treated DNA, PCR (for sequence-specific amplification), digital PCR and use of methylation-sensitive restriction enzymes.

In certain embodiments, methylation-specific PCR can be used to determine the methylation status of a genomic loci. Methylation-specific PCR is based on a chemical reaction of sodium bisulfite with DNA that converts unmethylated cytosines, e.g., of CpG dinucleotides, to uracil or UpG, followed by traditional PCR. Methylated cytosines will not be converted in this process, and primers can be designed to overlap the methylation site, e.g., CpG site, of interest, thereby allowing one to determine the methylation status of the methylation site as methylated or unmethylated. Additionally, restriction enzyme digestion of PCR products amplified from bisulfite-converted DNA may be used, e.g., by using the method described by Sadri & Hornsby (Nucl. Acids Res. 1996; 24:5058-5059) or COBRA (Combined Bisulfite Restriction Analysis) (Xiong & Laird, Nucleic Acids Res. 1997; 25:2532-2534).

In certain embodiments, whole genome bisulfite sequencing, which is a high-throughput genome-wide analysis of DNA methylation, can be used to determine the methylation status of multiple genomic loci. It is based on sodium bisulfite conversion of genomic DNA, as described above, which is then sequenced on a next-generation sequencing platform. The sequences obtained are then re-aligned to the reference genome to determine the methylation states of cytosines, e.g., of CpG dinucleotides, present within the analyzed genomic loci based on mismatches resulting from the conversion of unmethylated cytosines into uracil.

In certain embodiments, genome-wide DNA methylation profiling can be performed using commercially-available arrays, thereby allowing the interrogation of multiple genomic loci, e.g., multiple CpG sites. Non-limiting examples of such arrays include HumanMethylation BeadChips (Illumina, San Diego, Calif.) and Infinium MethylationEPIC kit (Illumina). Additional methods for analyzing the methylation state of multiple genomic loci are provided in Yong et al., Epigenetics & Chromatin 2016; 9:26, which is incorporated by reference herein.

Kits of this Example

This Example provides kits for diagnosing, monitoring, classifying, and/or treating a subject with necrotizing enterocolitis. The kits described herein can comprise a means for determining and/or detecting the methylation status of one or more genomic loci.

Kits described herein can include, without limitation, packaged probe and primer sets (e.g., TaqMan probe/primer sets), arrays/microarrays, which further contain one or more probes, primers, or other detection reagents for determining the methylation state and/or level of one or more genomic loci. For example, a kit described herein can include one or more probes or primers for detecting the methylation state of one or more genomic loci. In certain embodiments, the one or more genomic loci comprise a CpG site. In certain embodiments, one or more of the genomic loci do not comprise a CpG site. For example, about 5% or more, about 10% or more, about 15% or more, about 20% or more, about 25% or more, about 30% or more, about 35% or more, about 40% or more, about 45% or more, about 50% or more, about 55% or more, about 60% or more, about 65% or more or about 70% or more of the one or more genomic loci detected by the primers or probes can comprise one or more CpG sites.

In certain non-limiting embodiments, a primer and/or probe described herein can be at least about 10 nucleotides or at least about 15 nucleotides or at least about 20 nucleotides in length, and/or up to about 200 nucleotides or up to about 150 nucleotides or up to about 100 nucleotides or up to about 75 nucleotides or up to about 50 nucleotides in length.

In a further non-limiting embodiment, the oligonucleotide primers and/or probes can be immobilized on a solid surface or support, for example, on a nucleic acid microarray, wherein the position of each oligonucleotide primer and/or probe bound to the solid surface or support is known and identifiable.

In certain non-limiting embodiments, the kits described herein can additionally include other components such as a buffer, enzymes such as DNA polymerases or ligases, nucleotides such as deoxynucleotide triphosphates, positive control sequences, and/or negative control sequences necessary to carry out an assay or reaction to detect the methylation state of a genomic locus.

In certain embodiments, the kits described herein can include a container comprising one or more probes and/or primers for detecting the methylation state of one or more genomic loci. In certain embodiments, the kits further include instructions for use, e.g., the instructions can describe that a particular methylation status of a genomic locus is indicative of necrotizing enterocolitis in a subject. The instructions can be printed directly on the container (when present), or as a label applied to the container, or as a separate sheet, pamphlet, card or folder supplied in or with the container.

Reports, Programmed Computers, and Systems of this Example

In certain embodiments, a diagnosis and/or monitoring of necrotizing enterocolitis in a subject based on the methylation status of one or more genomic loci as described herein can be referred to herein as a “report.” A tangible report can optionally be generated as part of a testing process (which can be interchangeably referred to herein as “reporting,” or as “providing” a report, “producing” a report or “generating” a report).

Examples of tangible reports can include, without limitation, reports in paper (such as computer-generated printouts of test results) or equivalent formats and reports stored on computer readable medium (such as a CD, USB flash drive or other removable storage device, computer hard drive, or computer network server, etc.). Reports, particularly those stored on computer readable medium, can be part of a database, which can optionally be accessible via the internet (such as a database of patient records or genetic information stored on a computer network server, which can be a “secure database” that has security features that limit access to the report, such as to allow only the patient and the patient's medical practitioners to view the report while preventing other unauthorized individuals from viewing the report, for example). In addition to, or as an alternative to, generating a tangible report, reports can also be displayed on a computer screen (or the display of another electronic device or instrument).

A report can include, for example, an individual's medical history, or can just include size, presence, absence, or levels of one or more markers (for example, a report on computer readable medium such as a network server can include hyperlink(s) to one or more journal publications or websites that describe the medical/biological implications). Thus, for example, the report can include information of medical/biological significance as well as optionally also including information regarding the methylation status of relevant genomic loci, or the report can just include information regarding the methylation status of relevant genomic loci without other medical/biological significance.

A report can further be “transmitted” or “communicated” (these terms can be used herein interchangeably), such as to the individual who was tested, a medical practitioner (e.g., a doctor, nurse, clinical laboratory practitioner, genetic counselor, etc.), a healthcare organization, a clinical laboratory, and/or any other party or requester intended to view or possess the report. The act of “transmitting” or “communicating” a report can be by any means known in the art, based on the format of the report. Furthermore, “transmitting” or “communicating” a report can include delivering a report (“pushing”) and/or retrieving (“pulling”) a report. For example, reports can be transmitted/communicated by various means, including being physically transferred between parties (such as for reports in paper format) such as by being physically delivered from one party to another, or by being transmitted electronically or in signal form (e.g., via e-mail or over the internet, by facsimile, and/or by any wired or wireless communication methods known in the art) such as by being retrieved from a database stored on a computer network server, etc.

In certain embodiments, the disclosed subject matter provides computers (or other apparatus/devices such as biomedical devices or laboratory instrumentation) programmed to carry out the methods described herein, e.g., to perform the algorithm disclosed in this Example (see Example Embodiment J). In certain embodiments, the system can be controlled by the individual and/or their medical practitioner in that the individual and/or their medical practitioner requests the test, receives the test results back and (optionally) acts on the test results to reduce the individual's necrotizing enterocolitis risk or treat the individual, such as by implementing an necrotizing enterocolitis management system.

Example Embodiment I of Example 5: Whole Genome Bisulfite Sequencing of Laser-Captured Enterocytes from NEC-Affected Colon and Ileum

This Example Embodiment I identifies DNA methylation differences in enterocyte genomic DNA that may serve as biomarkers for NEC and also provides new insights into disease pathogenesis as well as tissue origin (colon or ileum). These experiments confirm that completely unbiased analysis of DNA methylation in an enterocyte cell-type-specific fashion can be performed. The goal of these studies was to differentiate patients with NEC from controls, thereby identifying putative diagnostic biomarkers that may be present in stool.

It is challenging to perform laser capture microdissection (LCM) followed by whole genome bisulfite sequencing (WGBS) in limited tissue samples. However, FIGS. 40 and 41 demonstrate that robust data can be generated using this approach. Specifically, a total of n=40 bisulfite sequencing libraries were prepared from individual tissue samples following laser capture microdissection of enterocytes. These consisted of n=12 from normal colon, n=10 from NEC colon, n=11 from normal ileum and n=7 from NEC ileum. There are differences that exist in the cellular make-up of the ileum and colon. NEC occurs commonly in the distal ileum, but also affects the colon, so the following was performed to determine whether differences exist in these cell types. FIG. 40 shows images of neonatal gut epithelial cells before and after LCM. The resulting DNA samples were subjected to WGBS in which a total, across all samples, of aligned read pairs 5,162,557,910 were sequenced. Sequencing data were aligned to the GRCh38 reference genome. As shown in FIG. 41, unsupervised clustering of the resulting WGBS data identified numerous DNA methylation changes associated with NEC when compared with control samples.

Note that, control tissue samples were in fact healed NEC tissue, obtained during surgical re-anastomosis. These samples were the only readily available source of human tissue for this study design, and the practice of using these controls is appropriate in the NEC field. When evaluating premature intestine, it is noted that the average premature infant is not having a bowel resection or a routine endoscopy as in older children/adults. Control bowel is never resected from a healthy infant, and it is very rare that surgeons will take non-necrotic margins from patients with NEC as they want to preserve every centimeter of bowel as possible. Bowel can be resected for non-inflamed conditions such as patients intestinal atresia, but to have those patients aged matched with an infant with NEC is nearly impossible and as those patients usually deliver at term and further, would require multi-center funding to collect these samples.

NEC samples were obtained across a corrected gestational span of approximately 5 weeks. Given the challenges associated with consenting subjects and obtaining samples, samples were combined for analysis into case versus control and corrected gestational age variables were ignore. Furthermore, because of the manner in which individuals had, prior to the beginning of this study, been consented and samples obtained, complete information regarding neonatal sex for all samples was not obtained. The focus, therefore, was only on autosomal data to avoid the complication of X-inactivation and minimize the impact of sex differences.

Analysis of the data identified differentially-methylated CpG sites that differ between NEC and control colon and NEC and control ileum. These are summarized in FIGS. 42A and 42B, respectively. 5352 genomic loci were identified on automsomes containing gene bodies (introns/exons) or promoter regions with a difference in average methylation rate of at least 0.1 between NEC and control colon samples. This compared with 2785 genomic loci on automsomes containing gene bodies or promoter regions with a difference in average methylation rate of at least 0.1 between NEC and ileum colon samples. The focus was on autosomal markers to avoid potentially confounding influences of neonatal sex caused by chromosome X-inactivation and, by including males and females, the number of samples available for comparison between NEC and control tissues was maximized. Differentially methylated single CpG sites were identified, and there were 38,843 of these between NEC and control colon with an adjusted p value (q value) of <0.05 (p value <0.00043) found. Notably, there were far fewer of these (n=652) at the same level of significance (q=0.05) when comparing NEC and control ileum.

Further analysis identified the top 30 CpG sites, determined by LCM-coupled WGBS, with the most highly significant differences (q value) between NEC and control colon and the top 30 CpG sites with the most highly significant differences between NEC and control ileum. These are shown in Table 12A and Table 12B. The lists of CpG sites represent examples of specific DNA methylation differences that may be detected in the relevant biological samples listed herein, demonstrating that DNA methylation differences such as those listed and many others exist and can be used to differentiate samples as described herein, for example, as described using procedures similar to those set forth in Example Embodiment J.

TABLE 12A
The top 30 CpG sites, determined by LCM-coupled WGBS,
with the most highly significant
differences (q value) between NEC and control colon.
				meth.diff
chr	coordinate	pvalue	qvalue	(NEC-Control)
chr17	11019015	1.20E−21	8.39E−15	40.2966432
chr2	86062716	2.67E−17	9.34E−11	32.2328187
chr4	115491856	5.48E−16	1.28E−09	40.7954545
chr14	52272903	1.95E−15	3.41E−09	72.8461538
chr10	21066470	1.40E−14	1.32E−08	35.8379475
chr3	175224788	1.33E−14	1.32E−08	37.8378378
chr5	24263140	1.51E−14	1.32E−08	56.26401
chr4	58967547	2.02E−14	1.57E−08	38.961039
chr2	79966760	3.05E−14	2.14E−08	55.4058265
chr4	22556420	3.43E−14	2.18E−08	45.3261808
chr5	17911285	3.99E−14	2.33E−08	33.6363636
chr2	119409733	4.60E−14	2.47E−08	53.9423547
chr20	38460167	9.79E−14	4.89E−08	42.204006
chr3	101804213	1.31E−13	5.91E−08	51.4418811
chr7	67694175	1.35E−13	5.91E−08	64.7450111
chr6	121426895	1.44E−13	5.94E−08	63.5005066
chr12	101416276	1.81E−13	7.02E−08	42.526497
chr6	78594584	2.13E−13	7.85E−08	50.2966259
chr11	12359400	2.47E−13	8.05E−08	38.5416667
chr6	81046717	2.53E−13	8.05E−08	31.9619471
chr15	63678136	2.37E−13	8.05E−08	55.5917004
chr1	167097455	3.50E−13	1.02E−07	66.4761905
chr15	81380112	4.30E−13	1.20E−07	44.6670538
chr2	150152356	4.63E−13	1.25E−07	56.978022
chr6	167440526	6.28E−13	1.47E−07	47.9513444
chr16	63305874	7.40E−13	1.67E−07	40.3446227
chr10	99591101	8.02E−13	1.70E−07	60.5306799
chr5	22585518	7.92E−13	1.70E−07	50.7375776
chr8	61783188	8.72E−13	1.80E−07	58.4869432
chr19	53612831	1.15E−12	2.24E−07	51.3878279

TABLE 12B
The top 30 CpG sites, determined by LCM-coupled WGBS, with the most
highly significant differences (q value) between NEC and control ileum.
chr	coordinate	pvalue	qvalue	meth.diff
chr2	70827641	2.11E−23	1.85E−16	−95.240642
chr1	100785163	4.60E−21	2.02E−14	−52.534562
chr1	100762620	2.21E−17	6.47E−11	−59.166667
chr1	143641173	6.27E−17	1.38E−10	34.0425532
chr6	73558525	9.22E−17	1.62E−10	84.6153846
chr11	10964709	5.72E−16	8.36E−10	39.4366197
chr17	53706550	4.09E−15	5.13E−09	−71.428571
chr1	100785162	1.64E−14	1.80E−08	−47.682119
chr10	53230713	2.29E−13	2.23E−07	36
chr8	27077249	3.38E−13	2.96E−07	94.1176471
chr5	36566389	5.11E−13	4.08E−07	31.1688312
chr5	60868950	6.00E−13	4.18E−07	−72.093023
chr7	29003503	6.20E−13	4.18E−07	35.7142857
chr1	100768520	1.11E−12	6.93E−07	−55.769231
chr11	47812794	1.21E−12	7.05E−07	−86.486486
chr4	147431181	1.47E−12	8.06E−07	−95.238095
chr1	67009294	1.81E−12	9.33E−07	−57.823129
chr2	45853090	2.60E−12	1.27E−06	94.4444444
chr13	33070399	3.26E−12	1.51E−06	−70.731707
chr14	68071519	4.29E−12	1.88E−06	−80.090909
chr13	25087158	4.95E−12	2.07E−06	−88.888889
chr1	100750518	9.64E−12	3.68E−06	−58.558559
chr14	22548745	9.48E−12	3.68E−06	73.9130435
chr19	7737210	1.05E−11	3.84E−06	−41.052632
chr2	54973902	1.29E−11	4.52E−06	30.6451613
chr5	61059457	1.71E−11	5.78E−06	−59.701493
chr1	123215253	1.80E−11	5.86E−06	35.5555556
chr5	116099451	1.91E−11	5.86E−06	−88.333333
chr6	33559051	1.94E−11	5.86E−06	−76.086957
chr3	78587902	2.03E−11	5.94E−06	−85.714286

Targeted Genome-Wide Analysis of NEC-Specific DNA Methylation in Tissue Sections from Neonatal Ileum and Colon

Targeted genome-wide bisulfite DNA sequencing from histopathological sections obtained from the tips of the villi or the crypts during NEC and control neonatal gut samples (both colon and ileum) was performed. This method delivered higher sequencing read depth per-dollar than WGBS in well-characterized regions of the human genome including promoters, exons, introns, CpG islands, CpG island shores and enhancers. This included about 5.5 million CpG sites. This approach quantified DNA methylation levels within genes involved in known biological pathways.

Specifically, targeted genome-wide bisulfite sequencing was carried out on DNA extracted from histological tissue sections obtained from n=8 NEC colon, n=13 control colon, n=5 NEC ileum and n=9 control ileum samples. Bisulfite-converted DNA libraries underwent targeted capture using the SeqCap Epi CpGiant Enrichment Kit (Roche, Pleasanton, Calif.) and sequenced to a read depth of ˜50×. The Bismark Bisulfite Read Mapper was used to align the libraries against the GRCh38 reference genome and determine the methylation status for each CpG site. DNA methylation signatures were identified using the beta-binomial test implemented in the R packages methylSig and DSS. A total, across all samples, of aligned read pairs 3,126,811,295 was sequenced.

Note that, control tissue samples were in fact healed NEC tissue, obtained during surgical re-anastomosis. These samples were the only readily available source of human tissue for this study design, and the practice of using these controls is appropriate in the NEC field. NEC samples were obtained across a corrected gestational span of approximately 5 weeks. Given the challenges associated with consenting subjects and obtaining samples, samples were combined for analysis into case versus control, and corrected gestational age variables were ignore. Furthermore, because of the manner in which individuals had, prior to the beginning of this study, been consented and samples obtained, complete information regarding neonatal sex for all samples was not obtained. The focus, therefore, was only on autosomal data to avoid the complication of X-inactivation and minimize the impact of sex differences.

As shown in FIG. 43, unsupervised clustering of the resulting WGBS data identified numerous DNA methylation changes associated with NEC when compared with control samples. Numerous differentially-methylated CpG sites that differ between NEC and control colon and NEC and control ileum were identified. Specifically, 1144 genomic loci containing gene bodies (introns/exons) or promoter regions with a difference in average methylation rate of at least 0.1 between NEC and control colon samples were identified. This compared with 1388 genomic loci containing gene bodies or promoter regions with a difference in average methylation rate of at least 0.1 between NEC and ileum control samples. Differentially methylated single CpG sites were identified, and 327 of these between NEC and control colon with an adjusted p value (q value) of <0.05 (p value <0.00043) were found. Notably, there far fewer of these (n=282) at the same level of significance (q=0.05) when comparing NEC and control ileum.

The top 30 CpG sites, determined via targeted genome-wide analysis, with the most highly significant differences (q value) between NEC and control colon and the top 30 CpG sites with the most highly significant differences (q value) between NEC and control ileum were identified. These are shown in Table 13A and Table 13B. Again, the lists of CpG sites represent examples of specific DNA methylation differences that may be detected in the relevant biological samples listed herein, demonstrating that DNA methylation differences such as those listed and many others exist and can be used to differentiate samples as described herein, for example, as described using procedures similar to those set forth in Example Embodiment J.

TABLE 13A
The top 30 CpG sites, determined by genome-wide targeted bisulfite
sequencing of histological tissue sections, with the most highly
significant differences (q value) between NEC and control colon.
chr	coordinate	pvalue	qvalue	meth.diff
chr2	218381984	2.46E−12	2.19E−06	−35.326767
chr19	50604303	2.71E−11	1.45E−05	55.0837283
chr7	75985405	2.98E−11	1.45E−05	44.1860465
chr6	224755	9.29E−11	3.31E−05	−33.937735
chr16	85544194	2.51E−10	5.98E−05	−33.923694
chr2	156328627	2.64E−10	5.98E−05	−36.422369
chr15	25176679	3.75E−10	6.26E−05	−35.126778
chr3	183211486	1.34E−09	0.00015263	−34.514436
chr19	50604306	2.52E−09	0.00024029	49.2693651
chr11	76588776	2.79E−09	0.00025528	−37.664924
chr17	79086422	3.82E−09	0.00029847	−30.068785
chr1	226952071	5.07E−09	0.0003385	−32.523562
chr12	34310055	5.14E−09	0.0003385	−34.253185
chr5	157458062	5.01E−09	0.0003385	−40.56775
chr17	48156384	5.63E−09	0.00036127	−32.469752
chr19	50604335	6.15E−09	0.00038688	30.3593704
chr1	36913980	7.14E−09	0.00042878	43.6803937
chr4	87863958	7.14E−09	0.00042878	−36.179812
chr19	50604301	7.53E−09	0.00043353	54.0622169
chr4	87864026	9.08E−09	0.00048062	−35.344262
chr5	42924363	1.02E−08	0.00050657	35.5612968
chr1	1763942	1.27E−08	0.00057572	−30.307887
chr8	41623158	1.27E−08	0.00057572	33.8067593
chr17	4145468	1.66E−08	0.0006958	58.0597015
chr7	37389812	1.65E−08	0.0006958	−35.284934
chr19	941117	1.72E−08	0.00071385	44.6992481
chr11	1486034	1.83E−08	0.00073694	32.6567192
chr10	61869861	2.36E−08	0.00087391	31.7641412
chr8	124994839	2.61E−08	0.00093255	−30.431059
chr17	4145498	2.82E−08	0.00094855	51.9661017

TABLE 13B
The top 30 CpG sites, determined by genome-wide targeted bisulfite
sequencing of histological tissue sections, with the most highly
significant differences (q value) between NEC and control ileum.
chr	coordinate	pvalue	qvalue	meth.diff
chr5	36782610	1.46E−16	8.31E−10	75.8134902
chr9	136767037	2.02E−11	5.74E−05	37.8666667
chr5	36782611	9.68E−11	0.00011858	67.5714472
chr1	1759026	1.25E−10	0.00011858	76.0162602
chr16	23839083	1.16E−10	0.00011858	33.5234386
chr16	87999766	1.61E−10	0.00012579	74.1666667
chr6	106098949	2.69E−10	0.00016954	31.5360412
chr5	133047721	3.46E−10	0.00019645	−46.346154
chr14	22237077	8.18E−10	0.00033552	72.2095238
chr11	1238921	8.27E−10	0.00033552	36.6792929
chr17	8475916	7.38E−10	0.00033552	−34.309441
chr19	2248684	1.77E−09	0.00058746	−50
chr6	170110578	2.26E−09	0.00067503	52.8765352
chr8	41665339	4.23E−09	0.0011452	35.942029
chr5	36783444	7.87E−09	0.00178832	56.4516129
chr12	51392880	7.58E−09	0.00178832	40.3846154
chr5	36783343	1.04E−08	0.00228362	69.2049272
chr6	135722823	1.60E−08	0.00324704	36.3983488
chr10	90319722	1.68E−08	0.00328501	43.5909091
chr18	79225913	1.75E−08	0.00331339	70.713073
chr20	50891154	1.97E−08	0.00344344	51.6445352
chr5	16539524	1.88E−08	0.00344344	51.6109422
chr1	6078881	2.43E−08	0.00373276	69.2307692
chr2	218273094	2.37E−08	0.00373276	41.7998198
chr15	84235722	2.87E−08	0.00388368	36.9852941
chr15	41501738	2.65E−08	0.00388368	−30.643539
chr17	59826115	3.83E−08	0.00483427	39.0368981
chr1	5077162	4.31E−08	0.00519313	31.9605263
chr10	126392102	4.84E−08	0.00519313	46.9061462
chr2	39245224	4.57E−08	0.00519313	−42.878788

Targeted Genome-Wide Analysis of DNA Methylation in Whole Blood from NEC Cases and Controls

Blood samples from NEC patients (n=6) were compared with blood samples from control premature infants (n=6). A further samples from NEC cases (n=9) and from controls (n=7) were sequenced, and data is analyzed. These samples did not fully overlap with tissue samples analyzed. This was partly because of the availability of the tissue and blood samples and partly because of the challenges associated with gathering data from laser captured tissue, which meant that some tissue samples could not be assayed effectively. Data presented herein was from NEC samples that were surgical only. The goal was to identify DNA methylation changes in blood samples from affected individuals and advance the understanding of epigenomic dysregulation in the circulating hematopoietic cell compartment during NEC. As before, the SeqCap Epi CpGiant Enrichment Kit and related protocols as described above were used to generate an average of 42× read depth coverage across ˜80.5 Mb of the human genome. A total, across all samples, of aligned read pairs 1,671,368,776 was sequenced.

Analysis of the data identified differentially-methylated CpG sites that differ between NEC and control blood samples. These are summarized in FIG. 44. 1071 genomic loci containing gene bodies or promoter regions with a difference in average methylation rate of at least 0.1 between NEC and control blood samples were identified. Differentially methylated single CpG sites were identified, and 1,870 of these between NEC and control blood with an adjusted p value (q value) of <0.05 (p value <7.48×10⁻⁵) were found. Further analysis identified the top 30 CpG sites, determined by LCM-coupled WGBS, with the most highly significant differences (q value) between NEC and control blood samples. These are shown in Table 14. Again, the lists of CpG sites represent examples of specific DNA methylation differences that may be detected in the relevant biological samples listed herein, demonstrating that DNA methylation differences such as those listed and many others exist and can be used to differentiate samples as described herein, for example, as described using procedures similar to those set forth in Example Embodiment J.

TABLE 14
The top 30 CpG sites, determined by genome-wide targeted bisulfite
sequencing of neonatal whole blood samples, with the most highly
significant differences (q value) between NEC and control neonates.
chr	coordinate	pvalue	qvalue	meth.diff
chr14	96964828	8.37E−48	5.72E−41	−49.454564
chr14	96964827	1.25E−40	4.28E−34	−50.27117
chr8	142252127	9.27E−29	2.11E−22	−43.839856
chr18	673016	1.84E−25	3.15E−19	−48.047898
chr12	66884475	1.03E−23	1.41E−17	−49.206349
chr18	673017	1.72E−21	1.95E−15	−48.640963
chr19	6004094	2.42E−20	2.36E−14	40.8653846
chr12	20777963	5.47E−19	4.68E−13	−50
chr1	2153138	1.92E−18	1.45E−12	35.1955307
chr7	832620	3.60E−18	2.46E−12	32.9454859
chr18	35271680	6.99E−18	4.34E−12	46.5648855
chr4	165278238	1.54E−17	8.77E−12	−43.776491
chr5	57878551	1.94E−17	1.02E−11	34.1562161
chr1	223230217	5.11E−17	2.49E−11	35.7997186
chr1	223230191	8.28E−16	2.98E−10	33.6101736
chr7	51477426	7.43E−16	2.98E−10	50.3582517
chr4	165278239	8.78E−16	3.00E−10	−42.763623
chr1	97938476	2.45E−15	7.98E−10	−30.130003
chr18	35271887	6.70E−15	2.08E−09	50.5058426
chr5	10377076	3.27E−14	8.01E−09	33.8299617
chr2	144414421	3.28E−14	8.01E−09	−41.004454
chr1	223230234	3.45E−14	8.14E−09	36.5955035
chr2	151292174	8.76E−14	1.84E−08	−31.327801
chr17	768851	8.89E−14	1.84E−08	31.2066503
chr2	43040642	1.04E−13	2.10E−08	−31.634519
chr11	64870691	1.22E−13	2.28E−08	−30.076286
chr2	21228815	1.62E−13	2.84E−08	−38.551119
chr20	31473908	1.81E−13	2.94E−08	−41.565217
chr2	47101189	1.75E−13	2.94E−08	77.1785268
chr11	23021922	2.75E−13	4.28E−08	74.2857143

Targeted Genome-Wide Analysis of DNA Methylation in Stool Samples from NEC Cases and Controls

A discovery-based analysis of DNA methylation in stool samples from NEC affected neonates and controls was performed. The data was gathered from four cases and four controls. The SeqCap Epi CpGiant Enrichment Kit and related protocols as described above were used to generate bisulfite sequencing data.

Analysis of the stool data identified differentially-methylated CpG sites that differ between NEC and control samples. These are summarized in FIG. 45. 6579 genomic loci containing gene bodies or promoter regions with a difference in average methylation rate of at least 0.1 between NEC and control blood samples were identified. Differentially methylated single CpG sites were further identified, and 28,594 of these between NEC and control stool with an adjusted p value (q value) of <0.05 (p value <7.48×10⁻⁵) were found. Further analysis identified the top 30 CpG sites, determined by LCM-coupled WGBS, with the most highly significant differences (q value) between NEC and control stool samples. These are shown in Table 15. Again, the lists of CpG sites represent examples of specific DNA methylation differences that may be detected in the relevant biological samples listed herein, demonstrating that DNA methylation differences such as those listed and many others exist and can be used to differentiate samples as described herein, for example, as described using procedures similar to those set forth in Example Embodiment J.

TABLE 15
The top 30 CpG sites, determined by genome-wide targeted bisulfite
sequencing of neonatal stool samples, with the most highly
significant differences (q value) between NEC and control neonates.
chr	coordinate	pvalue	qvalue	meth.diff
chr4	148261442	3.30E−36	1.43E−29	53.5847129
chr4	148261415	1.45E−35	3.16E−29	51.6383654
chr7	155267735	1.49E−32	2.16E−26	45.4545455
chr1	207486726	2.65E−25	2.89E−19	57.2916667
chr3	176361389	6.81E−25	5.93E−19	−47.540984
chr17	58155314	6.60E−24	4.79E−18	−43.26874
chr1	25892797	2.10E−21	1.31E−15	69.6666667
chr10	129700127	5.96E−21	2.88E−15	−45
chr14	104186550	5.56E−21	2.88E−15	48.2352941
chr22	18080123	1.77E−19	7.69E−14	−54.285714
chr17	81159445	5.11E−19	2.02E−13	−50.833333
chr10	112356734	6.38E−19	2.31E−13	−86.792453
chr4	678098	1.12E−18	3.47E−13	54.8633185
chr12	130281547	1.20E−18	3.48E−13	52.0294024
chr4	109013654	1.47E−18	4.00E−13	82.8841608
chr1	151852673	3.06E−18	7.83E−13	59.6153846
chr11	71615502	4.61E−18	1.11E−12	41.3995726
chr5	110726839	1.20E−17	2.62E−12	43.7357306
chr11	68342746	1.31E−17	2.71E−12	71.9317678
chr16	29289124	4.63E−17	9.16E−12	51.3902427
chr10	129700126	5.04E−17	9.54E−12	−49.39158
chr13	113343082	2.81E−16	4.36E−11	40.3794643
chr17	1888879	3.46E−16	5.02E−11	−40.35225
chr7	95324592	3.41E−16	5.02E−11	50.8015087
chr11	17390274	3.74E−16	5.24E−11	40.4255319
chr16	75494614	3.99E−16	5.42E−11	39.8058252
chr9	2234921	4.66E−16	6.14E−11	63.5017422
chr1	182084066	1.13E−15	1.36E−10	−43.99529
chr10	102201061	1.70E−15	1.94E−10	53.0410357
chr3	176362257	4.72E−15	5.01E−10	−42.105263

Identification of Differentially Methylated Tissue Markers in Stool and Blood

One goal was to explore the overlap between DNA methylation differences identified in NEC versus control intestinal tissue and NEC versus control stool. It was hypothesized that differentially methylated CpG sites in intestinal tissue would be detectable in stool. Using the pilot data generated using the small number of stool samples from NEC patients and controls, the relationship between NEC-specific biomarkers identified in tissue with those identified in stool was explored. A total of n=41 autosomal CpG sites with adjusted p values of <0.05 and at least 20% methylation that are differentially methylated in both stool and tissue (ileum or colon) were identified. These were markers of interest because of their potential for translation into a stool-based assay for NEC (Table 16).

TABLE 16
Chromosome Coordinate
chr1       202007351
chr2       74177514
chr2       218273237
chr2       237905493
chr2       156328443
chr2       27199738
chr4       76692730
chr4       8205509
chr5       73813624
chr6       125914370
chr6       160091207
chr6       109294253
chr7       1941242
chr7       4706797
chr7       2735000
chr7       102614072
chr8       137636543
chr11      83456378
chr11      1238921
chr11      130166648
chr15      59401917
chr15      41502938
chr16      23187923
chr16      49969809
chr16      29820860
chr17      59840980
chr17      49294764
chr17      39098668
chr17      80774434
chr19      38267481
chr20      43559198
chr21      45482307
chr10      8378253
chr11      86676450
chr5       170105682
chr2       190791817
chr1       2315672
chr1       54475455
chr12      114666918
chr19      751200
chr8       122778505

Some overlap of NEC-specific differentially methylated sequences between stool and blood samples was identified (FIG. 46). Specifically, a total of n=145 were identified as shared CpG sites with difference of at least 20% methylation and q=<0.05. These markers were notable because of the potential to identity blood-based changes in the molecular phenotype of NEC in stool (Table 17).

TABLE 17
Chromosome Coordinate
chr1       223230217
chr17      4176389
chr19      2607848
chr7       135233855
chr4       1311577
chr11      118344406
chr2       240704772
chr13      40188158
chr21      45470957
chr4       123304818
chr16      89162563
chr1       116987097
chr17      76270311
chr6       139161125
chr1       203289852
chr1       167518392
chr8       266789
chr17      7549609
chr16      18558775
chr7       135233856
chr1       111593228
chr15      42156044
chr1       85298457
chr2       240704741
chr9       124840384
chr1       183195278
chr9       106921553
chr8       143461306
chr2       240704800
chr13      40188108
chr19      16330718
chr6       6614442
chr3       16301690
chr3       16301758
chr2       240704799
chr2       112182833
chr14      64468656
chr14      103140911
chr19      4153717
chr11      76232249
chr6       6614487
chr1       12453505
chr20      32359934
chr7       2076816
chr16      8644675
chr7       2076791
chr1       23949357
chr7       73299956
chr14      56205791
chr7       2076830
chr5       74642861
chr7       2076837
chr7       2076821
chr10      75428630
chr4       24974189
chr2       66444595
chr1       8211944
chr7       2076878
chr12      11724866
chr19      45379921
chr7       2048342
chr4       39481120
chr11      72135742
chr14      102210824
chr17      9361393
chr7       2077145
chr1       2084262
chr11      73272408
chr7       2076851
chr19      13835962
chr16      81495456
chr7       2076970
chr1       220882458
chr12      124506266
chr7       1152618
chr9       35042348
chr7       100431249
chr11      46522145
chr14      102210833
chr17      74531468
chr11      66509471
chr3       10357812
chr4       184775559
chr11      62806931
chr14      22851685
chr16      88975383
chr16      89764825
chr11      46522066
chr11      72135687
chr14      102210805
chr1       8211961
chr8       8374015
chr2       43540209
chr1       26554057
chr7       101718554
chr5       78527242
chr7       2076971
chr4       39481134
chr9       105065030
chr1       220882435
chr3       195154566
chr10      78256178
chr15      69462230
chr8       129988124
chr1       9089676
chr3       194599071
chr1       220882449
chr6       130993602
chr5       151298601
chr7       2076930
chr6       131572840
chr11      3154396
chr3       66442636
chr16      29607174
chr8       129988086
chr20      35093420
chr2       54673911
chr19      17022149
chr7       4714929
chr11      3154467
chr19      1423758
chr4       8200785
chr11      66335268
chr1       220882431
chr1       12714901
chr11      86673628
chr17      27632878
chr4       6939133
chr11      6576981
chr5       177508225
chr9       136917905
chr12      124487335
chr7       2076955
chr4       39481150
chr15      74422826
chr4       24974053
chr8       6970582
chr3       11556527
chr16      929814
chr11      71478431
chr9       136917853
chr2       239210589
chr1       1185856
chr16      29607169
chr13      27054217

Example Embodiment J of Example 5: Estimation of Abnormal Stool Methylome Variation in Targeted Regions for Diagnosis of Necrotizing Enterocolitis

Provided below is an algorithm that can be used to diagnose a subject with necrotizing enterocolitis. The presently disclosed subject matter provides that the methylome(s) of intestinal tissue of a mammal, or structures therein (e.g., small intestines), could be affected by certain abnormalities (e.g., necrotizing enterocolitis), and that the changes of these methylomes can lead to changes in the methylation patterns of the DNA fragments found in stool, which are released by intestinal tissue. An algorithm was developed to identify the changes of methylation patterns in the methylome of stool caused by intestinal tissue phenotypes. The main insight behind this algorithm was that the methylome of the DNA fragments in stool is a mixture of a variety of component methylomes of intestinal origin, and that the proportion of these different component methylomes in the mixture varies from subject to subject, even among the population with normal intestinal tissue phenotype. By constructing a model of stool methylome as a linear combination of various component methylomes of intestinal tissue origins, the algorithm can accurately predict the methylation patterns of a new stool sample under the hypothesis that it is from a normal individual. Consequently, the algorithm has high sensitivity for detecting abnormal methylation patterns in a stool sample caused by changes of the methylomes of some intestinal tissues when the sample is from an affected individual. The procedure can be applied with little modification to the diagnosis of NEC using other types of biopsy samples, such as plasma, provided that the DNA fragments from the tissues affected by NEC can be found in those biopsy samples.

Let i be any CpG site in human genome, z_i,j be the methylation level of CpG site i in a stool sample j, p_i,r,j be the proportion of the r^th component methylome m_r,j of intestinal tissue origin in stool sample j at site i, m_i,r,j be the methylation level of CpG i in methylome m_r,j. The hypothesis is:

Z_i,j=Σ_r=1^Rp_i,r,jm_i,r,j  (1)

where p_i,r,j, m_i,r,j>=0, m_i,r,j<=1, p_i,1,j+ . . . +p_i,R,j=1.

It is further assumed that there is a set of CpG sites S such that, for any CpG site i in S, and any stool j from a normal individual, it has m_I,r,j=m_l,r and p_I,r,j=p_r,j.

That is, it is assumed that in any stool sample from a normal individual, the proportions of different component methylomes in the mixture are the same for all CpG sites in S. It is also assumed that, by restricting to the set of CpG sites S, stool samples from all normal individuals have the same set of component methylomes. They are called restricted reference component methylomes (RRCM), and are labeled as m₁^S, . . . , m_R^S or simply m₁, . . . , m_R when there is no confusion. For any stool sample j from a normal individual, its methylome restricted to set of CpG sites in S can be expressed as a weighted average of the restricted reference component methylomes. More precisely, let z_j^S be the methylome of stool sample C restricted to S, then for some mixture vector p_j=[p_j,l . . . , p_j,R]^T, it has:

z_j^s=[m₁^S, . . . m_R^S]p_j  (2)

Finally, it is assumed that the set S is the union of two disjoint subsets C and T, where T is a union of K non-empty sets T_k such that T=U_k=1^KT_k where the index k represents the k^th type of abnormal intestinal tissue phenotype. T_k's do not need to be disjoint. Moreover, T_k itself is the union of two disjoint sets D_k and V_k. Either D_k or V_k could be empty, but not both. It is assumed that for any stool sample, including one from an abnormal individual, when restricted to CpG sites in C, its methylome can always be expressed as a weighted average of the restricted reference component methylomes. That is, it has: z_j^C=[m₁^C, . . . , m_R^C]p_j regardless whether j is from an abnormal individual. C is called the set of reference CpG sites. On the other hand, for a stool sample l from an abnormal individual, when restricted to CpG sites in S=CUT, its methylome can no longer be expressed as a weighted average of the restricted reference component methylomes. That is, it has: w₁^S≠[m₁^S, . . . , m_R^S]p_l for any mixture vector p_l. More specifically, for a stool sample l from an individual with the k^th type of abnormal phenotype, it has: 1), w_j^C=m₁^C, . . . , m_R^C]p_l, 2), if D_K is non-empty, then w_iD_K=[m_1,k^Dk, . . . , m_R,k^Dk]p_l such that [m₁^Dk, . . . , m_R^Dk]·[m_1,k^Dk, . . . , m_R,k^Dk], and 3), if V_k is non-empty, then w_l^Vk=[m₁^Vk, . . . , m_R^Vk]q_l such that p_l≠q_l. In other words, in a stool sample from the k^th type of abnormal individual, if the set D_k is not empty, the component methylomes of the sample l restricted to D_k are no longer the same as the reference component methylome restricted to D_k. If the set V_k is not empty, in this stool sample, the proportion of the reference component methylomes restricted to V_k is no longer the same as the proportion of the reference component methylome restricted to R.

T is called the target set of CpG sites, D_k is called the differential methylation target set, V_k is called the copy number variation target set, and T_k is called the target set for the k^th type of abnormal phenotype.

The main steps of the algorithm of this Example are:

• • 1) Identify the sets of reference CpG sites C, and T₁, . . . , T_K for the list of K types of abnormal individuals.
  • 2) Estimate the restricted reference component methylomes m_R, or R predictor methylomes n₁, . . . , n_R that are independent linear combinations of the reference component methylomes such that n_r=[m₁, . . . , m_R]q_r for R linearly independent mixture vectors q₁, . . . , q_R.
  • 3) (Optional) If the reference component methylomes are available, estimate the proportions of these components at the reference CpG sites C for the test stool samples.
  • 4) Predict the methylation level of the test stool samples at the target set T_k of CpG sites, under the hypothesis that the sample is from a normal individual.
  • 5) Compare the predicted methylation levels at D_k and V_k against the observed methylation levels, and reject the null hypothesis that a test sample is from a normal individual if the observed methylation levels are significantly different form the predicted levels.

The algorithm of this Example can be implemented in a variety of ways. For example, given the methyl-seq data for a set of stool samples from normal individuals, the presently disclosed EM algorithm or the data augmentation method can be applied to estimate the component methylomes, then use the maximum likelihood method to estimate the proportion of these component methylomes in the test sample. Below are exemplary simple implementations of the presently disclosed algorithm that use linear regression.

In the simplest implementation of the algorithm of this Example, it is assumed the restricted methylome of a stool sample from a normal individual can be approximated by a mixture of two restricted reference methylomes. It is further assumed that the estimations of these two reference component methylomes are available. For example, in the implementation below, for the genomic loci of interest, the stool methylome is approximated by the mixture of ileum and colon methylomes. The implementation of the algorithm includes the following steps:

1. Identify the Reference Set C, and the Target Sets T₁, . . . , T_K.

• • 1.1 Collect the methylation data for a set of colon-derived cell samples, a set of ileum-derived cell samples, and a set of stool samples, all from normal individuals. For each type of abnormal individuals, collect a set of colon-derived samples, a set of ileum-derived cell samples, and a set of stool samples from that type of abnormal individuals. All these samples should have matched age, race, and other relevant parameters. These are the training data.
  • 1.2 Let x_i,j be the observed methylation level of CpG site i in a normal colon-derived sample j, and y_i,l the observed methylation level of CpG site i in a normal ileum-derived cell sample l, s_x,i² the sample variance of x_i,j over all normal colon-derived samples, s_y,i² the sample variance of y_i,j over all normal ileum-derived cell samples. Identify the CpG sites S₀ such that for any i∈S₀, it has both s_x,i²<c₀ and s_y,i²<c₀ for some constant c₀. These are CpG sites with stable methylation levels in each type of normal cells.
  • 1.3 Let x_i,j be the observed methylation level of CpG site i in a colon-derived sample j, including normal and abnormal, and y_i,l the observed methylation level of CpG site i in a ileum-derived cell sample l, including normal and abnormal, s_x,i² the sample variance of x_i,j over all colon-derived samples, including normal and abnormal, s_y,i² the sample variance of y_i,j over all ileum-derived cell samples, including normal and abnormal. Identify the CpG sites S₁ such that for any i∈S₁, it has both s_x,i²<c₀ and s_y,i²<c₀ for some constant c₀, and that the statistical test for the difference between {x_i,j0: j0 is a normal colon—derived sample} and {x_i,jk: jk is an abnormal colon—derived sample of type k} is not significant for all abnormal types of colon-derived, and that the statistical test for the difference between {y_i,j0: j0 is a normal ileum—derived cell sample} and {y_i,jk: jk is an abnormal ileum—derived cell sample of type k} is not significant for all abnormal types of ileum-derived cell. These are CpG sites with stable methylation levels in each type of cells, and with no difference in methylation level between normal and any abnormal samples. Let x_i be the sample mean of x_i,j over all colon-derived samples, including normal and abnormal, y_i the sample mean of over all ileum-derived cell samples, including normal and abnormal. Identify the subset C₀ of S₁ such that for any i∈C₀, it has |x_i−y_i|>c₁ for some constant c₁. These are CpG sites that are stably methylated in each cell type, with no difference between the normal and abnormal samples of the same cell type, and differentially methylated between different types of cells.
  • 1.4 Let x^R0 be the vector of x_i for all i∈C₀, and y^C0 be the vector of y_i for all i∈C₀, where x_i is the mean methylation at site i in all colon-derived samples y_i the mean methylation at site i in all ileum-derived cell samples. Note that by the way the set C₀ is selected, there is no difference in the methylation level of any CpG sites in C₀ between normal and abnormal colon-derived samples, or between normal and abnormal ileum-derived cell samples. Let z_j^C0 be the observed methylation levels of CpG sites in C₀ for a stool sample j of the k^th abnormal type. (For convenience, the normal stool sample is called as sample of the 0th abnormal type). For each sample j belonging to the k^th abnormal type, regress z_j^C0 against x^C0 and y^C0, with the constraints that the intercept must be 0, and the coefficients must be non-negative and add to 1, and get the residual e_j^C0. Identify the subset C₀^k of C₀ such that for any CpG i in C₀^k, it has

$\frac{e_{i,k}^2}{s_{i,k}}<c_2,$

and e_i,k²<c₃ for some constants c₂ and c₃, where e_i,k² is the mean of the squared difference between estimated and observed methylation levels of CpG site i in all stool samples of the k^th abnormal type, and s_i,k² the sample variances of methylation levels of CpG site i in the same set of stool samples. Repeat the above procedure for each type of abnormal stool samples, the intersection of the subsets C=∩_k=0^KC₀^k is the reference set of CpG sites. These are CpG sites where their methylation levels in both normal and any type of abnormal stool samples can be accurately predicted by the reference component methylomes from normal individuals.

• • 1.5 Let T₀=S₀\ S₁. Let x^C and x^T0 be the vectors of x_i and x_h for all i∈C and h∈T₀ respectively, and y^C and y^T0 be the vectors of y_i and y_h for all i∈C and h∈T₀ respectively, where x_i, x_h, y_i, and y_h are mean methylation level of sites for a normal colon-derived or ileum-derived cell at sites i and h respectively. Let z_j^C and z_j^T0 and be the observed methylation levels of CpG sites in C and T₀ respectively for a normal stool sample j, w_lk^C and w_lk^T0 the observed methylation level of CpG sites in C and T₀ respectively for a stool sample l_k from an individual with the k^th type of abnormality, w_lg^C and w_lg^T0 the observed methylation level of CpG sites in C and T₀ respectively for a stool sample l_g from an individual with the g^th type of abnormality, where g≠k. For each j, l_k, and l_g, regress z_j^C, w_lk^C, and w_lg^C respectively against x^C and y^C, with the constraints that the intercept must be 0, and the coefficients must be non-negative and add to 1. Apply the fitted models respectively to x^T0 and y^T0 to predict z_j^T0, w_lk^T0, and w_lg^T0 respectively, and get the differences e_j^T0, e_lk^T0 and e_lg^T0 between the predicted values and observed values. Let e_i, e_i,k, and e_i,g be the means of the sets of differences {e_j^T0: j is a normal stool sample}, {e_lk^T0: l_k is a stool sample of th k^th abnormal type} and {e_lg^T0: l_g is a stool sample of the g^th abnormal type} for CpG site i respectively. Identify the subset T_k of T₀ such that for any i∈T_k, it has |e_i|<c_2,0, |e_I,k|>c_2,k, and |e_i,k−e_i,g|>c_3,k, for some constants C_2,0, C_2,k, and C_3,k, for all g≠k. T_k is the target set for the k^th type of the abnormal individual. These are the sites where the methylation of a normal stool sample can be accurately predicted, the observed methylation in a stool sample of the k^th abnormal type will deviate from the prediction, and deviation will be different from that of a stool sample of any other abnormal type.

2. Estimate Fraction of the New Stool Samples to be Tested

Recall that x^c and y^c are mean vectors of the methylation levels of the training colon-derived and training ileum-derived cell data for the CpG sites in the reference set C. For any new stool sample t to be tested, let z_t^C be the observed methylation levels of CpG sites in C. Regress z_t^C against x^C and y^C, with the constraints that the intercept must be 0, and the coefficients must be non-negative and add to 1. The estimated coefficients are the estimated fractions of the component methylomes for the stool sample t.

3. Test if the New Stool Samples are from the k^th Type of Abnormal Individual.

For the new stool sample t, let X^Tk and y^Tk be mean vectors of the methylation levels of the training colon-derived and training ileum-derived cells data for the CpG sites in the target set T_k identified in step 1 of this algorithm, apply the fitted regression models obtained from the step 2 of this algorithm to X^Tk and y^Tk to predict the methylation levels of CpG sites in T_k for sample t under the hypothesis that sample t is from a normal. Let n_k be the number of CpG sites in T_k. Define functions f_k(x₁, . . . , x_nk)=Σ_i(−1)^{I_(ei,k−ei)}x_i and f_k,g(x₁, . . . , x_nk)=Σ_i(−1)^{I_(ei,k−ei,g)}x_i, where I_(⋅)=I_(−∞,0)(⋅), that is, the indicator function for the interval (−∞, 0), e_i, e_i,k and e_i,g are estimations obtained from step 1.5 of the algorithm. It will be said the sample is from an individual with the k^th type of abnormal phenotype if f_k(e_1,t−e₁, . . . , e_nk,t−e_nk)>c_4,k, and f_k,g(e_1,t−e_1,g, . . . , e_nk,t−e_nk,g)>c_5,g for all g≠k, where e_i,t is the difference between the observed methylation level of the CpG site i∈T_k for sample t and the predicted value by the fitted model obtained from step 2, and g≠k is any type of abnormal phenotype that is different form the k^th type of abnormal phenotype.

Other ways of implementing the algorithm of this Example can be developed by modifying the simple implementation presented above. Specifically, it does not need to assume that there are only two component reference methylomes that make up the stool methylomes, nor does it need to approximate them by mixtures of the component methylomes. Instead, a set of predictor methylomes can be collected that are themselves mixtures of component reference genomes, as long as the number of the predictor methylomes is the same as the number of the reference component methylomes, and the mixture vectors of the predictor methylomes are linearly independent. For example, they can be methylomes of stool samples with known different proportion of colon-derived and ileum-derived cell DNAs.

In the algorithm of this Example, the difference between observed methylation levels in certain target regions and the predicted methylation levels as the test statistic to determine if in a stool sample the methylome has been affect by some type of intestinal tissue abnormality. To illustrate the advantage of this approach, it is assumed that the mixture vector p_j for the methylome of a normal stool sample j followed a Dirichlet's distribution with parameters α_i= . . . =α_R. Furthermore, for CpG site i, its methylation levels in the R reference vector p_j for component methylomes are m_i,r=(r−1)/(R−1). It can be shown that the methylation level of i in sample j then has a mean of 0.5, and a variance of

$\frac{R+1}{12\left(R-1\right)\left(R{\alpha }_1+1\right)}.$

If there is a methyl-seq library of sample j with a coverage of N for CpG site i, the variance of the measured methylation level z_i,j is

${\sigma }_1^2=\frac{1}{4N}+\frac{N-1}{N}\frac{R+1}{12\left(R-1\right)\left(R{\alpha }_1+1\right)}.$

In other words, if z_i,j is used as a test statistic to detect abnormal intestinal tissue using stool sample, under the null hypothesis, the test statistic has a variance of σ₁². However, in the presently disclosed algorithm, it is first estimated the mixture vector p_j, then predicted x_i,j by Σ_rm_i,rp_r,j. Note that in a methyl-seq data, it can get millions of CpG sites covered in each library, and that the variance of the coefficients in a linear regression model is inversely proportional to sample size. Thus it is possible to get highly accurate estimation of the mixture vector p_j, even if it is taken into account that adjacent CpG sites tend to have correlated methylation levels. Assuming an accurate estimate of Σ_rm_i,rp_r,j can be obtained, that is, the error of the estimation can be ignored, the variance of the difference z^i,j−Σ_rm_i,rp_r,j between the observed methylation level and the prediction will be

$\frac{1}{4N}-\frac{1}{N}\frac{R+1}{12\left(R-1\right)\left(R{\alpha }_1+1\right)}.$

In other words, under the null hypothesis, the test static z_i,j−Σ_rm_i,rp_r,j used in the presently disclosed algorithm has a much smaller variance than the other candidate test statistic z_i,j. This in turns means that the presently disclosed test will achieve a higher power at the same level of type I error.

Examples of Embodiments for Example 5

E1. A method for diagnosing, prognosing, classifying, and/or monitoring necrotizing enterocolitis in a mammal, comprising:

(a) obtaining a sample from the mammal;

(b) determining the methylation status and/or level of one or more genomic loci in the sample;

(c) comparing the methylation status and/or level of the one or more genomic loci to a reference or a set of predicted values; and

(d) diagnosing necrotizing enterocolitis in the mammal,

wherein the difference in the methylation status and/or level of the one or more genomic loci in the sample compared to the reference or a set of predicted values indicates the presence of necrotizing enterocolitis in the mammal.

E2. The method of embodiment E1, wherein an increase in the level of methylation of the one or more genomic loci in the sample indicates the presence of necrotizing enterocolitis in the mammal or a decrease in the level of methylation of the one or more genomic loci in the sample indicates the presence of necrotizing enterocolitis in the mammal.
E3. The method of embodiment E1, wherein a decrease in the level of methylation of at least one of the one or more genomic loci in the sample and an increase in the level of methylation of at least one of the one or more genomic loci in the sample indicates the presence of necrotizing enterocolitis in the mammal.
E4. A method of treating necrotizing enterocolitis in a mammal, comprising:

(a) obtaining a sample from the mammal;

(b) determining the methylation status and/or level of one or more genomic loci present in the sample;

(c) comparing the methylation status and/or level of the one or more genomic loci to a reference or a set of predicted values;

(d) diagnosing necrotizing enterocolitis in the mammal, wherein the difference in the methylation status and/or level of the one or more genomic loci in the sample compared to the reference or a set of predicted values indicates the presence of necrotizing enterocolitis in the mammal; and

(e) administering an antibiotic therapy.

E5. The method of any one of embodiments E1-E4, wherein the reference is the methylation status and/or level of the one or more genomic loci in a sample obtained from a mammal that does not have necrotizing enterocolitis.
E6. The method of any one of embodiments E1-E5, wherein said sample is a stool sample.
E7. A method of treating necrotizing enterocolitis comprising:

(a) measuring the methylation status and/or level of one or more genomic loci present in a sample from a mammal prior to a treatment of necrotizing enterocolitis;

(b) measuring the methylation status and/or level of one or more genomic loci present in a sample from the mammal during the treatment of necrotizing enterocolitis; and

(c) continuing the treatment if the difference in the methylation status and/or level of the one or more genomic loci between the samples from prior to and during the treatment of necrotizing enterocolitis indicates the mammal is responsive to the treatment.

E8. The method of embodiment E7, further comprising (d) administering a different treatment to the mammal if the difference in the methylation status and/or level of the one or more genomic loci between the samples from prior to and during the treatment of necrotizing enterocolitis indicates the mammal is not responsive to the treatment.
E9. The method of embodiment E7, wherein an increase in the level of methylation of the one or more genomic loci in the sample indicates the mammal is responsive to the treatment, or a decrease in the level of methylation of the one or more genomic loci in the sample indicates the mammal is responsive to the treatment.
E10. The method of embodiment E7, wherein a decrease in the level of methylation of at least one of the one or more genomic loci in the sample and an increase in the level of methylation of at least one of the one or more genomic loci in the sample indicates the mammal is responsive to the treatment.
E11. The method of embodiment E7, wherein an increase in the level of methylation of the one or more genomic loci in the sample indicates the mammal is not responsive to the treatment, or a decrease in the level of methylation of the one or more genomic loci in the sample indicates the mammal is not responsive to the treatment.
E12. The method of embodiment E7, wherein a decrease in the level of methylation of at least one of the one or more genomic loci in the sample and an increase in the level of methylation of at least one of the one or more genomic loci in the sample indicates the mammal is not responsive to the treatment.
E13. The method of any one of embodiments E7-E12, wherein said sample is a stool sample.
E14. The method of any one of embodiments E1-E13, wherein the one or more genomic loci comprise one or more CpG sites.
E15. The method of any one of embodiments E1-E14, wherein the one or more genomic loci are present within nucleic acids isolated from the sample.
E16. The method of any one of embodiments E1-E15, wherein the one or more genomic loci are present within cell-free nucleic acids isolated from the sample.
E17. A method of treating necrotizing enterocolitis, comprising;

(a) diagnosing necrotizing enterocolitis in a mammal by utilization of the algorithm disclosed in Example Embodiment J; and

(b) administering an antibiotic therapy to said mammal to treat said necrotizing enterocolitis.

E18. The method of any one of embodiments E1-E17, wherein said mammal is a human.
E19. A kit for diagnosing, prognosing, and/or monitoring necrotizing enterocolitis in a mammal comprising a means for determining and/or detecting the methylation status of one or more genomic loci.
E20. The kit of embodiment E19, wherein the means comprises one or more primers and/or probes for determining and/or detecting the methylation status of the one or more genomic loci.

Abstract of this Example (Example 5)

This Example provides methods for diagnosing, prognosing, monitoring, classifying, and/or treating necrotizing enterocolitis. For example, algorithms, kits, and methods for diagnosing, prognosing, monitoring, classifying, and/or treating necrotizing enterocolitis are provided.

Example 6—Methods and Materials for Assessing and Treating Ovarian Cancer

Field of this Example

This Example relates to methods and materials involved assessing a mammal (e.g., a human) for and/or treating a mammal (e.g., human) having or developing ovarian cancer. For example, this Example provides methods and materials for using DNA methylation profiles of nucleic acid within a liquid biopsy (e.g., a plasma sample, a urine sample, a peritoneal fluid sample, or a cervical swab sample) to determine whether or not a mammal has, or is developing, ovarian cancer. This Example also provides methods, algorithms, and kits for diagnosing, prognosing, monitoring, classifying, and/or treating ovarian cancer.

Background of this Example

Ovarian cancer often remains undetected until it becomes incurable or challenging to treat. Early diagnosis is frequently impossible or dangerously invasive by current methods. There is therefore great interest in the development of practical and accurate methods for the non-invasive detection and phenotyping of ovarian cancer.

Summary of this Example

This Example provides methods and materials involved assessing a mammal (e.g., a human) for and/or treating a mammal (e.g., human) having or developing ovarian cancer. For example, this Example provides methods and materials for using DNA methylation profiles of nucleic acid within a liquid biopsy (e.g., a plasma sample, a urine sample, a peritoneal fluid sample, or a cervical swab sample) to determine whether or not a mammal has, or is developing, ovarian cancer. Determining if a mammal (e.g., a human) has, or is likely to develop, ovarian cancer by assessing DNA methylation profiles of nucleic acid within a liquid biopsy (e.g., a plasma sample, a urine sample, a peritoneal fluid sample, or a cervical swab sample) can aid in the identification of mammals (e.g., humans) that should be treated in a particular manner (e.g., by administering chemotherapy, by administering immunotherapy, and/or by performing a surgical treatment), for example, early in the disease process.

This Example also provides methods for diagnosing, prognosing, monitoring, classifying, and/or treating ovarian cancer. For example, the methods described herein can include determining the methylation status of one or more genomic loci in a sample (e.g., a plasma sample, a urine sample, a peritoneal fluid sample, or a cervical swab sample) of a mammal (e.g., a female human). This Example further provides algorithms and kits for diagnosing, prognosing, monitoring, classifying, and/or treating ovarian cancer.

In one aspect, this Example provides a method for diagnosing, prognosing, classifying, and/or monitoring ovarian cancer in a mammal (e.g., a human female) comprising: (a) obtaining a sample (e.g., a plasma sample, a urine sample, a peritoneal fluid sample, or a cervical swab sample) from the mammal; (b) determining the methylation status and/or level of one or more genomic loci in the sample; (c) comparing the methylation status and/or level of the one or more genomic loci to a reference or a set of predicted values; and (d) identifying the mammal as having ovarian cancer, wherein the difference in the methylation status and/or level of the one or more genomic loci in the sample compared to the reference or a set of predicted values indicates the presence of ovarian cancer in the mammal.

In another aspect, this Example provides a method of treating ovarian cancer in a mammal (e.g., a human female) comprising: (a) obtaining a sample (e.g., a plasma sample, a urine sample, a peritoneal fluid sample, or a cervical swab sample) from the mammal; (b) determining the methylation status and/or level of one or more genomic loci present in the sample; (c) comparing the methylation status and/or level of the one or more genomic loci to a reference or a set of predicted values; (d) identifying the mammal as having ovarian cancer, wherein the difference in the methylation status and/or level of the one or more genomic loci in the sample compared to the reference or a set of predicted values indicates the presence of ovarian cancer in the mammal; and (e) treating the mammal by administering a chemotherapy, by administering an immunotherapy, and/or by performing a surgical treatment.

In some cases, an increase in the level of methylation of the one or more genomic loci in a sample (e.g., a plasma sample, a urine sample, a peritoneal fluid sample, or a cervical swab sample) can indicate the presence of ovarian cancer in the mammal or a decrease in the level of methylation of the one or more genomic loci in a sample (e.g., a plasma sample, a urine sample, a peritoneal fluid sample, or a cervical swab sample) indicates the presence of the ovarian cancer in the mammal. In some cases, a decrease in the level of methylation of at least one of the one or more genomic loci in a sample and an increase in the level of methylation of at least one of the one or more genomic loci in the sample can indicate the presence of ovarian cancer in the mammal.

In some cases, the reference can be the methylation status and/or level of the one or more genomic loci in a sample (e.g., a plasma sample, a urine sample, a peritoneal fluid sample, or a cervical swab sample) obtained from a mammal (e.g., a human female) that does not have ovarian cancer.

In another aspect, this Example provides a method of treating ovarian cancer in a mammal (e.g., a human female) comprising: (a) measuring the methylation status and/or level of one or more genomic loci present in a sample (e.g., a plasma sample, a urine sample, a peritoneal fluid sample, or a cervical swab sample) from the mammal prior to a treatment of ovarian cancer; (b) measuring the methylation status and/or level of one or more genomic loci present in a sample (e.g., a plasma sample, a urine sample, a peritoneal fluid sample, or a cervical swab sample) from the mammal during the treatment of ovarian cancer; and (c) continuing the treatment if the difference in the methylation status and/or level of the one or more genomic loci between the samples from prior to and during the treatment of ovarian cancer indicates the subject is responsive to the treatment. In some cases, the method further comprises (d) administering a different treatment to the mammal if the difference in the methylation status and/or level of the one or more genomic loci between the samples from prior to and during the treatment of ovarian cancer indicates the subject is not responsive to the treatment.

In some cases, an increase in the level of methylation of the one or more genomic loci in the sample indicates that the mammal is responsive to the treatment, or a decrease in the level of methylation of the one or more genomic loci in the sample indicates the mammal is responsive to the treatment. In certain embodiments, a decrease in the level of methylation of at least one of the one or more genomic loci in the sample and an increase in the level of methylation of at least one of the one or more genomic loci in the sample indicates the mammal is responsive to the treatment. In certain embodiments, an increase in the level of methylation of the one or more genomic loci in the sample indicates the mammal is responsive to the treatment, or a decrease in the level of methylation of the one or more genomic loci in the sample indicates the mammal is not responsive to the treatment. In certain embodiments, a decrease in the level of methylation of at least one of the one or more genomic loci in the sample and an increase in the level of methylation of at least one of the one or more genomic loci in the sample indicates the subject is not responsive to the treatment.

This Example further provides algorithms for diagnosing and/or monitoring a mammal having ovarian cancer. In certain embodiments, the algorithm can be used to classify ovarian cancer of a mammal (e.g., a human female).

In another aspect, this Example provides a kit for diagnosing, prognosing, and/or monitoring ovarian cancer in a mammal comprising a means for determining and/or detecting the methylation status of one or more genomic loci. In certain embodiments, the means comprises one or more primers and/or probes for determining and/or detecting the methylation status of the one or more genomic loci.

In certain embodiments, the one or more genomic loci are present within nucleic acids isolated from the sample. In certain embodiments, the one or more genomic loci are present within cell-free nucleic acids isolated from the sample.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Although methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, suitable methods and materials are described below. All publications, patent applications, patents, and other references mentioned herein are incorporated by reference in their entirety. In case of conflict, the present specification, including definitions, will control. In addition, the materials, methods, and examples are illustrative only and not intended to be limiting.

Other features and advantages of the invention will be apparent from the following detailed description, and from the claims.

Description of this Example

This Example provides methods for diagnosing, prognosing, monitoring, classifying, and/or treating ovarian cancer. For example, the methods described herein can include determining the methylation status of one or more genomic loci in a sample of a mammal (e.g., a human female). In some cases, the methods described herein can include the use of an algorithm to diagnose, prognose, monitor, classify, and/or assist in the treatment of ovarian cancer. Non-limiting embodiments of this Example are described by the present specification and Examples.

Unless defined otherwise, all technical and scientific terms used in this Example generally have their ordinary meanings in the art, within the context of this Example and in the specific context where each term is used. The following references provide one of skill with a general definition of many of the terms used in this Example: Singleton et al., Dictionary of Microbiology and Molecular Biology (2nd ed. 1994); The Cambridge Dictionary of Science and Technology (Walker ed., 1988); The Glossary of Genetics, 5th Ed., R. Rieger et al. (eds.), Springer Verlag (1991); and Hale & Marham, The Harper Collins Dictionary of Biology (1991). Certain terms are discussed below, or elsewhere in the specification, to provide additional guidance to the practitioner in describing the compositions and methods of this Example and how to make and use them.

The terms “comprise(s),” “include(s),” “having,” “has,” “can,” “contain(s),” and variants thereof, as used herein, are intended to be open-ended transitional phrases, terms or words that do not preclude the possibility of additional acts or structures. The singular forms “a,” “an,” and “the” include plural references unless the context clearly dictates otherwise. The present Example also contemplates other embodiments “comprising,” “consisting of” and “consisting essentially of,” the embodiments or elements presented herein, whether explicitly set forth or not.

As used herein, the use of the word “a” or “an” when used in conjunction with the term “comprising” in the claims and/or the specification may mean “one,” but it is also consistent with the meaning of “one or more,” “at least one,” and “one or more than one.” Still further, the terms “having,” “including,” “containing,” and “comprising” are interchangeable, and one of skill in the art is cognizant that these terms are open ended terms.

The term “about” or “approximately” means within an acceptable error range for the particular value as determined by one of ordinary skill in the art, which will depend in part on how the value is measured or determined, i.e., the limitations of the measurement system. For example, “about” can mean within 3 or more than 3 standard deviations, per the practice in the art. Alternatively, “about” can mean a range of up to 20%, preferably up to 10%, more preferably up to 5%, and more preferably still up to 1% of a given value. Alternatively, particularly with respect to biological systems or processes, the term can mean within an order of magnitude, preferably within 5-fold, and more preferably within 2-fold, of a value.

In certain embodiments, the term “biomarker” refers to a marker (e.g., DNA methylation status) that allows detection of a disease and/or disorder in an individual, including detection of the disease or the disorder in its early stages. In certain embodiments, the term “biomarker” refers to a marker (e.g., DNA methylation status) that allows the characterization of a phenotype of a disease and/or a disorder in an individual. Early stage of a disease, as used herein, refers to the time period between the onset of the disease and the time point that signs or symptoms of the disease emerge. In certain non-limiting embodiments, the presence, absence, and/or level of a biomarker in a sample of a mammal (e.g., a human) is determined by comparing to a reference control.

The terms “reference sample,” “reference control,” “control,” or “reference,” as used interchangeably herein, refers to a control for a methylation status of a genomic locus that is to be detected in a sample of a mammal. In certain embodiments, a reference sample can be a sample from a healthy individual, e.g., an individual that does not have ovarian cancer. In certain embodiments, a reference sample can be a sample from a control individual that does not have the disease or phenotype to be detected by a biomarker disclosed herein. In certain embodiments, a control or reference can be the presence, absence, and/or a particular level of a methylation state of a genomic locus in a healthy individual. In certain embodiments, a reference can be a predetermined presence, absence, and/or particular level of a methylation state of a genomic locus that indicates a subject does not have ovarian cancer. In certain embodiments, a reference can be the methylation status of a locus in an individual having a disease or a phenotype, e.g., an individual that has ovarian cancer, where the methylation status of the locus is known to be not associated with the disease or the phenotype.

The term “a set of predicted values” refers to the methylation status of certain genomic loci for a sample. The status of those loci is not directly measured from that sample. Rather, it is inferred from measurements of other loci for that sample and/or measurements of other samples. The inference of the predicted values is based on some mathematical/statistical models. The models usually assume that the sample for which the methylation status of those loci is to be predicted has a normal phenotype. This assumption may be either correct or wrong, but its correctness is not required for the inference of the predicted values.

The term “slightly invasive or non-invasive method” refers to a method that does not involve the removal of tissues by biopsy from the ovaries. In certain embodiments, slightly invasive or non-invasive methods, as described herein, include obtaining plasma, urine, a peritoneal fluid sample, or a cervical swab from a subject.

The term “patient” or “subject,” as used interchangeably herein, refers to any warm-blooded animal, e.g., human or non-human. Non-limiting examples of non-human subjects include mammals, non-human primates, dogs, cats, mice, rats, guinea pigs, rabbits, fowl, pigs, horses, cows, goats, sheep, etc. In certain embodiments, the subject is human.

The term “nucleic acid,” “nucleic acid molecule,” or “polynucleotide” includes any compound and/or substance that comprises a polymer of nucleotides. Each nucleotide is composed of a base, specifically a purine- or pyrimidine base (i.e., cytosine (C), guanine (G), adenine (A), thymine (T), or uracil (U)), a sugar (i.e., deoxyribose or ribose), and a phosphate group. In certain embodiments, the nucleic acid molecule is described by the sequence of bases, whereby said bases represent the primary structure (linear structure) of a nucleic acid molecule. The sequence of bases is typically represented from 5′ to 3′. These terms encompass deoxyribonucleic acid (DNA) including, e.g., complementary DNA (cDNA) and genomic DNA, ribonucleic acid (RNA), in particular messenger RNA (mRNA), synthetic forms of DNA or RNA, and mixed polymers comprising two or more of these molecules. The herein described nucleic acid molecule can contain naturally occurring or non-naturally occurring nucleotides. Examples of non-naturally occurring nucleotides include modified nucleotide bases with derivatized sugars or phosphate backbone linkages or chemically modified residues.

The term “isolated” (e.g., isolated genomic DNA) refers to a biological component that has been substantially separated or purified away from other biological components in the cell of the organism in which the component naturally occurs, e.g., other chromosomal and extra-chromosomal DNA and RNA, proteins, and organelles. Nucleic acids, e.g., DNA, that have been “isolated” include nucleic acids purified by standard purification methods.

The term “genomic locus” or “genomic DNA locus,” as used herein, refers to any fixed position in a genome. For example, a genomic locus can refer to a genomic element, a chromosomal region, a gene, a region of a gene, e.g., an exon or intron, a regulatory region of a gene, e.g., a promoter or enhancer, a CpG site, a CpG island, or a CpG island shore. For example, a genomic locus can include one or more CpG sites, e.g., between about 1 to about 100 CpG sites. In certain embodiments, a genomic locus can be of any particular length, e.g., between about 1 to about 10,000 nucleotides in length.

As used interchangeably herein, “methylation state,” “methylation profile,” “methylation status,” and “methylation level” refer to the presence, absence, percentage, and/or quantity of methylation at a particular nucleotide, or nucleotides, within a DNA region, e.g., a genomic locus. The methylation status of a particular DNA sequence (e.g., a genomic locus) can indicate the methylation state of every nucleotide in the sequence, indicate the methylation state of any of the nucleotides (e.g., cytosines) in the sequence, can indicate the methylation state of a subset of the nucleotides (e.g., of cytosines), can indicate the percentage or fraction of methylated cytosines at any particular stretch of nucleotides within the sequence or can indicate the average rate of methylation of all the cytosines (or a subset of the cytosines) present in a nucleic acid.

As used herein, a “methylated nucleic acid molecule” refers to a nucleic acid molecule that contains one or more methylated nucleotides that is/are methylated.

As used herein, a “CpG site” or “methylation site” is a nucleotide within a nucleic acid that is susceptible to methylation either by natural occurring events in vivo or by an event instituted to chemically methylate the nucleotide in vitro.

A “CpG island,” as used herein, describes a segment of a nucleic acid, e.g., DNA sequence, that have a high frequency of CpG dinucleotide repeats. See, e.g., Illingworth and Bird, FEBS Letters, 2009; 583:1713-1720. For example, Yamada et al. (Genome Research, 2004; 14:247-266) have described a set of standards for determining a CpG island: it must be at least 400 nucleotides in length, has a GC content greater than 50%, and an OCF/ECF ratio greater than 0.6. Others (Takai et al., Proc. Natl. Acad. Sci. U.S.A., 2002; 99:3740-3745) have defined a CpG island less stringently as a sequence of at least 200 nucleotides in length, having a greater than 50% GC content and an OCF/ECF ratio greater than 0.6.

A “CpG island shore,” as used herein, refers to methylation hotspots that are present a short distance, e.g., less than 2 kb, from CpG islands.

The term “methylome,” as used herein, refers to the amount or pattern of methylation at different sites or regions within a population of cells. The methylome can correspond to all of the genome, a subset of the genome (e.g., repeat elements in the genome), or a portion of the subset (e.g., those areas found to be associated with ovarian cancer). A methylome from plasma can be referred to a “plasma fluid methylome,” or a “plasma fluid DNA methylome.” The plasma fluid methylome is an example of a cell-free methylome that includes cell-free DNA (cfDNA).

As used herein, the term “increase” refers to alter positively by at least about 2%, including, but not limited to, alter positively by about 5%, by about 10%, by about 15%, by about 20%, by about 25%, by about 30%, by about 35%, by about 40%, by about 45%, by about 50%, by about 55%, by about 60%, by about 65%, by about 70%, by about 75%, by about 80%, by about 85%, by about 90%, by about 95% or by about 100%.

As used herein, the terms “reduce,” “reduction,” or “decrease” refers to alter negatively by at least about 2%, including, but not limited to, alter negatively by about 5%, by about 10%, by about 15%, by about 20%, by about 25%, by about 30%, by about 35%, by about 40%, by about 45%, by about 50%, by about 55%, by about 60%, by about 65%, by about 70%, by about 75%, by about 80%, by about 85%, by about 90%, by about 95% or by about 100%.

As described herein, this Example provides methods for diagnosing, monitoring, classifying, and/or treating ovarian cancer by analyzing the methylation status of one or more genomic loci in a sample (e.g., a plasma sample, a urine sample, a peritoneal fluid sample, or a cervical swab sample) of a mammal (e.g., a human female). In certain embodiments, the methods can include using an algorithm described herein. In certain embodiments, the methods described herein can allow for the early diagnosis or screening of a subject with ovarian cancer, e.g., the subject does not have any symptoms, or only have early symptoms of ovarian cancer.

In certain embodiments, samples obtained for use in the methods described herein can include cfDNA, which carries DNA methylation information from the cell of origin. cfDNA can arise from cellular apoptosis and necrosis, and can be generated from active secretory processes, with the formation of extracellular vesicles. DNA signatures are highly tissue-specific, and include in vivo information relating to the tissue source of cfDNA. In certain embodiments, the methods described herein can include analyzing cfDNA in a sample (e.g., a plasma sample, a urine sample, a peritoneal fluid sample, or a cervical swab sample), to identify genetic phenotypes that are drivers and/or consequences of ovarian cancer.

The sample from the subject can be collected using any appropriate technique. For example, a blood sample, a plasma sample, a urine sample, a peritoneal fluid sample, or a cervical swab sample can be collected using standard methods. In some cases, the sample can be collected from the subject before the subject has any symptom of ovarian cancer, i.e., a non-symptomatic subject. In certain embodiments, the sample can be collected from the non-symptomatic subject who is at high risk of ovarian cancer. In certain embodiments, the sample can be collected from the subject who has previously received or is currently receiving a treatment for ovarian cancer. In certain embodiments, two or more samples (e.g., two or more, three or more, four or more, five or more, six or more or seven or more samples) can be obtained before and during the subject is receiving a treatment for ovarian cancer (e.g., serially obtained samples).

Diagnostic, Prognostic, Classification, and Monitoring Methods of this Example

This Example provides diagnostic and prognostic methods for diseases and/or disorders that are characterized by differential methylation of genomic loci. For example, this Example provides methods for diagnosing, prognosing, classifying, and/or monitoring ovarian cancer in a subject that includes analyzing the methylation status of certain genomic loci.

In certain embodiments, the analyzed genomic loci can include one or more genomic loci that exhibit differential methylation in a sample from a subject that has ovarian cancer compared to a reference sample. For example, the methods described herein can include assessing the methylation status of one or more genomic loci, e.g., about 5 or more, about 10 or more, about 50 or more, about 100 or more, about 500 or more, about 1,000 or more, about 5,000 or more, about 10,000 or more, about 25,000 or more, about 50,000 or more or about 100,000 or more genomic loci in a sample of a subject. In certain embodiments, the genomic loci can be selected from the genes, or a region within the genes, provided in Table 18 and/or Example Embodiment K. In certain embodiments, the one or more genomic loci can be one or more promoter regions of one or more genes, one or more exons of one or more genes, one or more introns of one or more genes, one or more CpG sites, one or more CpG islands, one or more CpG island shores, one or more enhancers of one or more genes, or a combination thereof. In certain embodiments, the genomic loci are present on a particular chromosome.

In certain embodiments, this Example provides methods for diagnosing, prognosing, and/or monitoring ovarian cancer in a subject by detecting the DNA methylation profiles associated with ovarian cancer. In certain embodiments, the methods described herein can include (a) obtaining a sample from the subject, (b) determining the methylation status of one or more genomic loci present in the sample, e.g., present within cfDNA in a plasma sample, a urine sample, a peritoneal fluid sample, or a cervical swab sample, (c) comparing the methylation status of the one or more genomic loci to a reference or a set of predicted values, and (d) diagnosing ovarian cancer in the subject. In certain embodiments, the difference in the methylation status of the one or more genomic loci in the sample compared to the reference or a set of predicted values indicates the presence of ovarian cancer in the subject. In certain embodiments, the difference in the methylation status also can indicate the severity of ovarian cancer.

In certain embodiments, the methods described herein for diagnosing, prognosing, and/or monitoring ovarian cancer in a subject can include (a) obtaining a sample from the subject, (b) determining the level of methylation of one or more genomic loci present in the sample, (c) comparing the level of methylation of the one or more genomic loci to a reference or a set of predicted values, and (d) diagnosing ovarian cancer in the subject. In certain embodiments, the difference in the level of methylation of the one or more genomic loci in the sample compared to the reference or a set of predicted values indicates the presence of ovarian cancer in the subject. In certain embodiments, the difference in the methylation level also can indicate the severity of ovarian cancer.

In certain embodiments, diagnosing ovarian cancer in the subject can include characterizing a phenotype of the ovarian cancer, wherein the difference in the methylation status of the one or more genomic loci in the sample compared to the reference or a set of predicted values indicates the phenotype of the ovarian cancer. In certain embodiment, the phenotype of the ovarian cancer can include the severity of the ovarian cancer, prognosis of the ovarian cancer, molecular expression profile of the ovarian cancer, responsiveness of the ovarian cancer to certain treatments, or any combinations thereof.

In certain embodiments, the methods described herein for determining if a subject is at risk of developing ovarian cancer in the subject can include (a) obtaining a sample from the subject, (b) determining the level of methylation of one or more genomic loci present in the sample, (c) comparing the level of methylation of the one or more genomic loci to a reference or a set of predicted values, and (d) determining that the subject is at risk of developing ovarian cancer, wherein the difference in the level of methylation of the one or more genomic loci in the sample compared to the reference or a set of predicted values indicates that the subject is at risk.

In certain embodiments, diagnosing, prognosing, and/or monitoring of a subject with ovarian cancer can be based on a higher or lower methylation level of the genomic locus in the sample of the subject relative to the methylation level in a reference sample, e.g., a sample from a subject that does not have ovarian cancer. In certain embodiments, a difference of greater than about 5%, greater than about 10%, greater than about 15%, greater than about 20%, greater than about 25%, greater than about 30%, greater than about 35%, greater than about 40%, greater than about 45%, greater than about 50%, greater than about 55%, greater than about 60%, greater than about 65%, greater than about 70%, greater than about 75%, greater than about 80%, greater than about 85%, greater than about 90% or greater than about 95% in the methylation (e.g., level, percentage and/or fraction) of the one or more genomic loci in a sample obtained from a subject compared to a control can be indicative that the subject has ovarian cancer or is at risk of developing ovarian cancer. In certain embodiments, the difference can be a decrease in methylation (e.g., level, percentage, and/or fraction) of the genomic loci in the sample of the subject. Alternatively, the difference can be an increase in methylation (e.g., level, percentage, and/or fraction) of the genomic loci in the sample of the subject. In certain embodiments, the difference can be a decrease in the methylation of a genomic locus and an increase in the methylation of a different genomic locus in the sample obtained from the subject. In certain embodiments, a decrease in the level of methylation of one or more genomic loci in the sample and the increase in the level of methylation of one or more different genomic loci in the sample can indicate the presence of ovarian cancer.

In certain embodiments, diagnosis of a subject with ovarian cancer can be based on the methylated or unmethylated state of a genomic locus, e.g., a CpG site. In certain embodiments, a genomic locus, e.g., a CpG site, in a sample from a subject diagnosed with ovarian cancer can be methylated and the genomic locus, e.g., the CpG site, in a reference sample can be unmethylated. In certain embodiments, a genomic locus in a sample from a subject diagnosed with ovarian cancer can be unmethylated and the genomic locus in a reference sample can be methylated.

Diagnostic, Prognostic, Classification, and Monitoring Methods Using an Algorithm of this Example

This Example also provides diagnostic and prognostic methods for diseases and/or disorders that are characterized by differential methylation of genomic loci by using an algorithm as described, for example, in Example Embodiment L. For example, this Example provides methods for diagnosing, prognosing, classifying, and/or monitoring ovarian cancer in a subject that includes analyzing the methylation status of certain genomic loci and/or genomic fractions.

Methods for Treating Ovarian Cancer

This Example also provides methods for treating a subject having ovarian cancer. For example, a mammal (e.g., a human female) that was identified as having ovarian cancer as described herein (or identified as being at risk of developing ovarian cancer as described herein) can be administered one or more chemotherapies, one or more immunotherapies, or a combination thereof to treat ovarian cancer. Examples of chemotherapies that can be used as described herein to treat ovarian cancer include, without limitation, carboplatin, cisplatin, paclitaxel, and docetaxel. Examples of immunotherapies that can be used as described herein to treat ovarian cancer include, without limitation, bevacizumab and durvalumab. In some cases, a mammal (e.g., a human female) that was identified as having ovarian cancer as described herein (or identified as being at risk of developing ovarian cancer as described herein) can be treated using a surgical procedure to treat ovarian cancer. Examples of surgical procedures that can be used as described herein to treat ovarian cancer include, without limitation, surgeries to remove one or both ovaries with or without removing the uterus.

In some cases, the information provided by the methods described herein can be used by a clinician or physician in determining the most effective course of treatment (e.g., preventative or therapeutic) for the subject. A course of treatment refers to the measures taken for a patient after the prognosis or the assessment of increased risk for development of ovarian cancer is made. For example, when a subject is identified to have an increased risk of developing ovarian cancer, the physician can determine whether frequent monitoring of DNA methylation changes can be performed as a prophylactic measure. Also, when a subject is diagnosed with ovarian cancer (e.g., based on the presence of a DNA methylation pattern in a sample from a subject), it can be advantageous to follow such detection with a therapeutic treatment.

In some cases, this Example provides methods for assessing the efficacy of a therapeutic or prophylactic therapy for treating ovarian cancer in a subject, comprising determining the methylation status of one or more genomic loci present in a sample obtained from a subject prior to the therapy and determining methylation status of the one or more genomic loci present in a sample obtained from the subject at one or more time points during the therapeutic or prophylactic therapy, wherein the therapy is efficacious for treating ovarian cancer in a subject when there is a change in the presence and/or level of methylation of the one or more genomic loci in the second or subsequent samples, relative to the first sample. In certain embodiments, the first sample is obtained after therapeutic treatment has begun.

In certain embodiments, the methods for monitoring the response in a subject to prophylactic or therapeutic treatment of ovarian cancer can include measuring the methylation status and/or level of one or more genomic loci in a sample of a subject at a first time-point, administering a therapeutic agent, re-measuring the methylation status and/or level of the one or more genomic loci at a second time-point, comparing the results of the first and second measurements and optionally modifying the treatment regimen based on the comparison. In certain embodiments, the first time-point can be prior to an administration of the therapeutic agent, and the second time-point can be after said administration of the therapeutic agent. In certain embodiments, the first time-point can be prior to the administration of the therapeutic agent to the subject for the first time. In certain embodiments, the dose (defined as the quantity of therapeutic agent administered at any one administration) can be increased or decreased in response to the comparison. In certain embodiments, the dosing interval (defined as the time between successive administrations) can be increased or decreased in response to the comparison, including total discontinuation of treatment. In addition, the methods described herein can be used to determine the efficacy of the therapeutic treatment, wherein a change in the methylation status of certain genomic loci present in a sample of a subject can indicate that the therapeutic treatment regimen can be altered, reduced, and/or stopped.

Assays of this Example

This Example also provides assays and/or methods for determining the DNA methylation status and/or level of genomic loci that correlates with the presence, absence, and/or severity of ovarian cancer. In some cases, the assay method can include comparing the methylation status and/or level of genomic loci present in a sample from a subject that has ovarian cancer to the methylation status and/or level of genomic loci in a sample from a healthy subject to determine the methylation pattern, as described above, that correlates with the presence of ovarian cancer. In some cases, the assay methods can include comparing the methylation status and/or level of genomic loci in a sample from a subject that has ovarian cancer at an early stage to the methylation status and/or level of genomic loci in a sample from a subject that has ovarian cancer at a late stage to determine the methylation status and/or level that correlates with the different stages and/or severity of ovarian cancer.

DNA Isolation Techniques of this Example

In certain embodiments, the methods described herein can include isolating nucleic acid from a sample (e.g., a plasma sample, a urine sample, a peritoneal fluid sample, or a cervical swab sample) obtained from a subject. Any appropriate technique can be used to isolate nucleic acids from a sample. For example, isolation of DNA from a plasma sample can be performed by extraction methods using organic solvents such as a mixture of phenol and chloroform, followed by precipitation with ethanol (see, for example, J. Sambrook et al., “Molecular Cloning: A Laboratory Manual,” 1989, 2nd Ed., Cold Spring Harbor Laboratory Press: New York, N.Y.). Additional non-limiting examples include salting out DNA extraction (see, for example, P. Sunnucks et al., Genetics, 1996, 144:747-756; and S. M. Aljanabi and I. Martinez, Nucl. Acids Res. 1997, 25:4692-4693), the trimethylammonium bromide salts DNA extraction method (see, for example, S. Gustincich et al., BioTechniques, 1991, 11:298-302), and the guanidinium thiocyanate DNA extraction method (see, for example, J. B. W. Hammond et al., Biochemistry, 1996, 240:298-300). There are also numerous commercially available kits that can be used to extract DNA from biological fluids (e.g., plasma samples) or cells. For example, Qiagen's Gentra PureGene Cell Kit, QlAamp Circulating Nucleic Acid Kit, QiaAmp DNA Mini Kit, DNeasy Blood and Tissue Kit or QiaAmp DNA Blood Mini Kit (Qiagen, Hilden, Germany), GenomicPrep™ Blood DNA Isolation Kit (Promega, Madison, Wis.) and GFX™ Genomic Blood DNA Purification Kit (Amersham, Piscataway, N.J.) can be used to obtain DNA from a sample from a subject.

Methylation Detection Techniques of this Example

Various methylation analysis procedures are known in the art, and can be used with the methods described herein. These assays allow for determination of the methylation state of one genomic locus, e.g., one or more CpG sites or islands within a nucleic acid obtained from a sample. In addition, the methods can be used to quantify the methylation of a genomic locus. Such assays involve, among other techniques, DNA sequencing of bisulfite-treated DNA, PCR (for sequence-specific amplification), digital PCR and use of methylation-sensitive restriction enzymes.

In certain embodiments, methylation-specific PCR can be used to determine the methylation status of a genomic loci. Methylation-specific PCR is based on a chemical reaction of sodium bisulfite with DNA that converts unmethylated cytosines, e.g., of CpG dinucleotides, to uracil or UpG, followed by traditional PCR. Methylated cytosines will not be converted in this process, and primers can be designed to overlap the methylation site, e.g., CpG site, of interest, thereby allowing one to determine the methylation status of the methylation site as methylated or unmethylated. Additionally, restriction enzyme digestion of PCR products amplified from bisulfite-converted DNA may be used, e.g., by using the method described by Sadri & Hornsby (Nucl. Acids Res. 1996; 24:5058-5059) or COBRA (Combined Bisulfite Restriction Analysis) (Xiong & Laird, Nucleic Acids Res. 1997; 25:2532-2534).

In certain embodiments, whole genome bisulfite sequencing, which is a high-throughput genome-wide analysis of DNA methylation, can be used to determine the methylation status of multiple genomic loci. It is based on sodium bisulfite conversion of genomic DNA, as described above, which is then sequenced on a next-generation sequencing platform. The sequences obtained are then re-aligned to the reference genome to determine the methylation states of cytosines, e.g., of CpG dinucleotides, present within the analyzed genomic loci based on mismatches resulting from the conversion of unmethylated cytosines into uracil.

In certain embodiments, genome-wide DNA methylation profiling can be performed using commercially-available arrays, thereby allowing the interrogation of multiple genomic loci, e.g., multiple CpG sites. Non-limiting examples of such arrays include HumanMethylation BeadChips (Illumina, San Diego, Calif.) and Infinium MethylationEPIC kit (Illumina). Additional methods for analyzing the methylation state of multiple genomic loci are provided in Yong et al., Epigenetics & Chromatin 2016; 9:26, which is incorporated by reference herein.

Kits of this Example

This Example provides kits for diagnosing, monitoring, classifying, and/or treating a subject with ovarian cancer. The kits described herein can comprise a means for determining and/or detecting the methylation status of one or more genomic loci.

Kits of this Example can include, without limitation, packaged probe and primer sets (e.g., TaqMan probe/primer sets), arrays/microarrays, which further contain one or more probes, primers, or other detection reagents for determining the methylation state and/or level of one or more genomic loci. For example, a kit described herein can include one or more probes or primers for detecting the methylation state of one or more genomic loci. In certain embodiments, the one or more genomic loci comprise a CpG site. In certain embodiments, one or more of the genomic loci do not comprise a CpG site. For example, about 5% or more, about 10% or more, about 15% or more, about 20% or more, about 25% or more, about 30% or more, about 35% or more, about 40% or more, about 45% or more, about 50% or more, about 55% or more, about 60% or more, about 65% or more or about 70% or more of the one or more genomic loci detected by the primers or probes can comprise one or more CpG sites.

In certain non-limiting embodiments, a primer and/or probe described herein can be at least about 10 nucleotides or at least about 15 nucleotides or at least about 20 nucleotides in length, and/or up to about 200 nucleotides or up to about 150 nucleotides or up to about 100 nucleotides or up to about 75 nucleotides or up to about 50 nucleotides in length.

In a further non-limiting embodiment, the oligonucleotide primers and/or probes can be immobilized on a solid surface or support, for example, on a nucleic acid microarray, wherein the position of each oligonucleotide primer and/or probe bound to the solid surface or support is known and identifiable.

In certain non-limiting embodiments, the kits described herein can additionally include other components such as a buffer, enzymes such as DNA polymerases or ligases, nucleotides such as deoxynucleotide triphosphates, positive control sequences, and/or negative control sequences necessary to carry out an assay or reaction to detect the methylation state of a genomic locus.

In certain embodiments, the kits described herein can include a container comprising one or more probes and/or primers for detecting the methylation state of one or more genomic loci. In certain embodiments, the kits further include instructions for use, e.g., the instructions can describe that a particular methylation status of a genomic locus is indicative of ovarian cancer in a subject. The instructions can be printed directly on the container (when present), or as a label applied to the container, or as a separate sheet, pamphlet, card or folder supplied in or with the container.

Reports, Programmed Computers, and Systems of this Example

In certain embodiments, a diagnosis and/or monitoring of ovarian cancer in a subject based on the methylation status of one or more genomic loci as described herein can be referred to herein as a “report.” A tangible report can optionally be generated as part of a testing process (which can be interchangeably referred to herein as “reporting,” or as “providing” a report, “producing” a report or “generating” a report).

Examples of tangible reports can include, without limitation, reports in paper (such as computer-generated printouts of test results) or equivalent formats and reports stored on computer readable medium (such as a CD, USB flash drive or other removable storage device, computer hard drive, or computer network server, etc.). Reports, particularly those stored on computer readable medium, can be part of a database, which can optionally be accessible via the internet (such as a database of patient records or genetic information stored on a computer network server, which can be a “secure database” that has security features that limit access to the report, such as to allow only the patient and the patient's medical practitioners to view the report while preventing other unauthorized individuals from viewing the report, for example). In addition to, or as an alternative to, generating a tangible report, reports can also be displayed on a computer screen (or the display of another electronic device or instrument).

A report can include, for example, an individual's medical history, or can just include size, presence, absence, or levels of one or more markers (for example, a report on computer readable medium such as a network server can include hyperlink(s) to one or more journal publications or websites that describe the medical/biological implications). Thus, for example, the report can include information of medical/biological significance as well as optionally also including information regarding the methylation status of relevant genomic loci, or the report can just include information regarding the methylation status of relevant genomic loci without other medical/biological significance.

A report can further be “transmitted” or “communicated” (these terms can be used herein interchangeably), such as to the individual who was tested, a medical practitioner (e.g., a doctor, nurse, clinical laboratory practitioner, genetic counselor, etc.), a healthcare organization, a clinical laboratory, and/or any other party or requester intended to view or possess the report. The act of “transmitting” or “communicating” a report can be by any means known in the art, based on the format of the report. Furthermore, “transmitting” or “communicating” a report can include delivering a report (“pushing”) and/or retrieving (“pulling”) a report. For example, reports can be transmitted/communicated by various means, including being physically transferred between parties (such as for reports in paper format) such as by being physically delivered from one party to another, or by being transmitted electronically or in signal form (e.g., via e-mail or over the internet, by facsimile, and/or by any wired or wireless communication methods known in the art) such as by being retrieved from a database stored on a computer network server, etc.

In certain embodiments, the disclosed subject matter provides computers (or other apparatus/devices such as biomedical devices or laboratory instrumentation) programmed to carry out the methods described herein, e.g., to perform the algorithm of this Example (see Example Embodiment L). In certain embodiments, the system can be controlled by the individual and/or their medical practitioner in that the individual and/or their medical practitioner requests the test, receives the test results back and (optionally) acts on the test results to reduce the individual's ovarian cancer risk or treat the individual, such as by implementing an ovarian cancer management system.

Example Embodiment K of Example 6: Non-Invasive Molecular Phenotyping of Human Plasma for Ovarian Cancer

An epigenomic analysis was performed using liquid biopsy of plasma DNA samples obtained from human ovarian cancer patients (n=9) and human healthy controls (n=11). Bisulfite-converted DNA libraries underwent targeted capture using the SeqCap Epi CpGiant Enrichment Kit (Roche, Pleasanton, Calif.) and were sequenced to a read depth of ˜50×. This approach targets 80.5 Mb of the human genome and ˜5.5 million individual CpG sites. The Bismark Bisulfite Read Mapper 34 was used to align the libraries against the GRCh38 reference genome and determine the methylation status for each CpG site.

These analyses demonstrated the ability to generate high resolution epigenomic liquid biopsy data in cell free DNA samples obtained from plasma. Examples of ovarian cancer-specific differentially methylated loci are shown in Table 18. The list of CpG sites represents examples of specific DNA methylation differences that may be detected in the relevant biological samples listed herein, demonstrating that DNA methylation differences such as those listed and many others exist and can be used to differentiate samples as described herein, for example, as described using procedures similar to those set forth in Example Embodiment L.

TABLE 18
Examples of differentially methylated CpG sites between normal
plasma and ovarian cancer plasma.
				% Methylation
				Difference
Chromosome	Coordinate	pvalue	qvalue	(Case-Control)
chr1	3366144	2.64E−06	0.066643	−13.7369
chr1	3366307	2.84E−06	0.069386	−10.1065
chr1	6239344	4.43E−06	0.090961	5.92168
chr1	15048599	3.50E−06	0.076076	5.372909
chr1	110675458	2.43E−06	0.063373	−13.7349
chr1	110675513	1.29E−06	0.039991	−16.7303
chr1	110675581	2.10E−07	0.013145	−18.3871
chr1	110675587	2.28E−07	0.013199	−18.4758
chr1	110675617	5.25E−07	0.022788	−19.1298
chr1	125182646	2.22E−06	0.060231	−7.31516
chr1	125182662	1.30E−08	0.001258	−7.57844
chr1	125182668	3.62E−08	0.002919	−7.66042
chr1	125182688	9.75E−07	0.035115	−7.94721
chr1	143213094	3.34E−06	0.074525	−8.37223
chr1	143238114	4.78E−06	0.095	−8.73461
chr1	143264564	4.80E−06	0.095	−6.5326
chr1	227969662	1.05E−06	0.036894	−9.434
chr10	10266910	9.63E−11	2.33E−05	−21.8125
chr10	10266955	2.33E−08	0.002033	−19.4378
chr10	10267015	4.32E−06	0.089071	−13.0664
chr10	42079914	1.56E−06	0.045245	−7.42861
chr10	48465484	1.13E−06	0.037592	−12.8601
chr10	95184074	1.11E−07	0.007959	−13.0014
chr10	95184082	6.62E−07	0.026342	−13.3685
chr10	95184085	1.48E−07	0.009912	−13.4852
chr10	95184109	6.44E−07	0.026245	−14.1042
chr10	95184115	1.08E−06	0.037293	−14.1683
chr10	95184166	2.91E−06	0.069394	−13.8895
chr10	125895681	4.20E−13	2.03E−07	−17.2557
chr10	125895682	9.72E−07	0.035115	−17.1847
chr10	125895696	5.64E−17	6.36E−11	−15.4744
chr10	125895697	4.86E−17	6.36E−11	−15.3924
chr10	125895708	2.36E−18	7.99E−12	−14.9818
chr10	125895713	3.07E−13	1.73E−07	−14.0875
chr10	125895714	6.18E−09	0.000675	−14.0002
chr10	125895715	6.78E−14	4.59E−08	−13.9126
chr10	125895716	9.99E−13	3.76E−07	−13.8246
chr10	125895747	2.88E−06	0.069394	−12.216
chr11	68209486	2.29E−07	0.013199	−5.62643
chr12	5039987	2.60E−06	0.06663	−15.3903
chr12	5039991	3.79E−06	0.080237	−15.4107
chr12	5666506	3.02E−06	0.070988	−17.206
chr12	6215837	2.02E−06	0.056581	−17.01
chr12	40693719	3.37E−06	0.074525	−15.8434
chr12	40693738	1.53E−06	0.044858	−16.4129
chr12	40693739	4.27E−08	0.003364	−16.4424
chr12	40693782	5.05E−06	0.099323	−16.6923
chr12	132394139	5.25E−07	0.022788	−12.3098
chr12	132394165	3.36E−06	0.074525	−12.4282
chr12	132394178	2.81E−07	0.013996	−12.107
chr12	132394181	5.80E−07	0.024225	−12.062
chr12	132394190	5.73E−07	0.024225	−11.8616
chr12	132394229	3.18E−08	0.002629	−11.3937
chr14	32938088	1.29E−06	0.039991	−14.9034
chr14	64442345	1.96E−06	0.055266	−13.0625
chr15	52256671	3.33E−06	0.074525	−23.4156
chr16	12390340	1.87E−07	0.012171	−15.1675
chr16	34576408	4.73E−06	0.095	−6.18683
chr16	34576478	3.29E−07	0.01592	−5.61188
chr16	34582466	1.44E−06	0.043574	2.624713
chr16	34595076	1.46E−06	0.043794	−8.51925
chr16	34595124	7.80E−08	0.005739	−8.47079
chr16	46387382	1.16E−06	0.037643	−5.40415
chr16	46389677	2.91E−06	0.069394	−6.99576
chr16	46400320	3.79E−07	0.017592	−8.23138
chr16	46400330	6.60E−07	0.026342	−8.01991
chr17	18625364	3.72E−06	0.079288	−21.5684
chr17	20320695	3.01E−06	0.070988	−6.81262
chr17	20320736	3.32E−06	0.074525	−6.37515
chr17	26885700	1.09E−08	0.001083	15.43918
chr17	78921813	2.64E−06	0.066643	−10.1664
chr18	9102789	1.05E−06	0.036894	−12.8311
chr18	9102821	5.02E−08	0.003863	−12.1908
chr18	9102824	2.82E−08	0.002387	−12.1118
chr18	9102830	5.81E−09	0.000675	−11.9481
chr18	9102833	5.30E−09	0.000644	−11.853
chr18	9102851	1.11E−06	0.037345	−11.2436
chr18	31724111	7.16E−08	0.005389	−12.0506
chr18	31724113	2.62E−07	0.013662	−12.0828
chr18	31724118	1.15E−06	0.037643	−12.1459
chr18	31724148	7.49E−07	0.028817	−12.3979
chr18	31724187	2.85E−06	0.069386	−12.5095
chr18	31724276	2.68E−07	0.013746	−12.2512
chr18	311724334	4.78E−06	0.095	−12.4003
chr18	31724335	5.33E−09	0.000644	−12.3964
chr18	31724365	2.30E−07	0.013199	−12.2822
chr18	31724433	2.30E−06	0.060665	−11.7912
chr18	31724435	1.19E−06	0.037839	−11.7942
chr18	31724440	1.17E−06	0.037643	−11.7475
chr18	31724443	4.18E−06	0.086773	−11.7545
chr18	31724529	4.87E−07	0.021667	−12.3839
chr18	31724541	2.16E−06	0.058943	−12.531
chr18	31724583	3.87E−07	0.017721	−13.5801
chr18	31724606	2.80E−07	0.013996	−14.2326
chr18	31724622	2.41E−07	0.01339	−14.6741
chr19	16719929	2.29E−06	0.060665	−20.5149
chr19	16719994	2.31E−06	0.060665	−18.6129
chr19	52517362	1.24E−07	0.008539	−25.6849
chr2	48532708	3.54E−07	0.016656	−16.3139
chr2	128901652	7.77E−07	0.029333	−20.0871
chr2	128901658	2.75E−06	0.067887	−20.4302
chr2	128901702	1.61E−06	0.046109	−21.9965
chr2	128901778	4.36E−07	0.019674	−23.3685
chr2	128901800	2.49E−07	0.013611	−23.8407
chr2	128901810	1.44E−06	0.043574	−23.913
chr2	128901821	3.51E−06	0.076076	−24.1044
chr2	128901846	3.05E−06	0.071291	−24.2352
chr2	188034367	2.09E−06	0.058014	−21.0074
chr22	22297772	2.17E−09	0.000366	−13.8252
chr22	22297775	3.24E−09	0.000476	−14.1664
chr22	22297780	2.06E−08	0.001941	−14.7199
chr22	22297786	4.69E−09	0.000611	−15.3596
chr22	22297801	6.08E−09	0.000675	−14.9302
chr22	22297836	6.56E−13	2.78E−07	−19.0675
chr22	22297844	2.60E−14	2.20E−08	−19.3494
chr22	22297852	2.24E−11	6.88E−06	−18.6188
chr22	22297861	1.19E−11	4.02E−06	−18.1964
chr22	22297866	3.36E−10	7.10E−05	−16.991
chr22	22297870	4.69E−09	0.000611	−16.5603
chr22	22297874	5.61E−10	0.000112	−16.3023
chr3	12542803	1.18E−07	0.008329	−12.4785
chr3	49358049	3.19E−06	0.073473	−8.95098
chr3	49358057	7.50E−11	1.95E−05	−9.26347
chr3	49358073	3.59E−09	0.000506	−9.58429
chr3	49358074	7.49E−10	0.000141	−9.60497
chr3	49358080	7.12E−09	0.000753	−9.55828
chr3	49358081	9.92E−09	0.001018	−9.57239
chr3	49358086	2.60E−07	0.013662	−9.52519
chr3	49358088	3.45E−07	0.01646	−9.53955
chr3	49358089	3.05E−09	0.00047	−9.54489
chr3	49358098	2.03E−07	0.012966	−9.56184
chr3	49358099	1.48E−09	0.000263	−9.55482
chr3	49358118	2.34E−08	0.002033	−9.38281
chr3	49358119	2.77E−10	6.25E−05	−9.35371
chr3	49358123	1.23E−06	0.038963	−9.41791
chr3	49358124	2.79E−11	7.88E−06	−9.38576
chr3	49689148	2.30E−08	0.002033	−8.63847
chr3	49689161	2.56E−09	0.000413	−8.14538
chr3	49689191	6.20E−07	0.025594	−6.60995
chr3	52317368	3.13E−06	0.072601	−12.8742
chr4	2430180	1.30E−06	0.039991	−8.00714
chr4	8254522	2.98E−07	0.014615	−7.09407
chr4	20003976	9.39E−07	0.034543	−16.5665
chr4	52051137	2.23E−07	0.013199	−9.05182
chr4	140755696	7.90E−07	0.029375	−4.19519
chr4	151409307	3.89E−06	0.081301	−6.26016
chr4	151409311	2.51E−06	0.064797	−6.01374
chr4	151409314	3.36E−06	0.074525	−6.15508
chr4	151409322	1.51E−06	0.044837	−6.35103
chr4	151409329	4.75E−06	0.095	−6.48028
chr4	151409364	2.11E−06	0.058014	−6.9842
chr4	169030994	4.79E−06	0.095	−13.0722
chr5	117304877	1.49E−07	0.009912	−7.84561
chr5	117304880	7.31E−07	0.028432	−7.83696
chr5	117304901	2.56E−07	0.013662	−7.54501
chr5	117304913	2.41E−07	0.01339	−7.31788
chr6	14088969	2.31E−06	0.060665	−28.6201
chr6	112366877	6.99E−07	0.027505	−13.3154
chr6	155449253	2.68E−06	0.067096	−11.4461
chr6	162728535	1.54E−06	0.044858	−15.2105
chr6	167672244	1.72E−06	0.048802	3.420434
chr7	752348	3.61E−06	0.077419	−4.15708
chr7	752349	1.10E−06	0.037345	−4.13669
chr7	1870636	1.11E−06	0.037345	4.898359
chr7	73271542	3.47E−06	0.076076	−13.4436
chr7	74480244	3.86E−06	0.08125	−12.3172
chr7	158959750	1.06E−06	0.036894	5.219832
chr8	69711200	7.80E−07	0.029333	−16.5349
chr8	69711209	2.14E−07	0.013192	−15.9125
chr8	109688955	2.73E−06	0.067887	−24.1368
chr8	131038983	5.73E−07	0.024225	−11.9616
chr9	106038008	3.59E−06	0.077419	10.21198

Methylation density plots of data are shown in FIG. 47. These represent a high level view of the data, mapped across various genomic elements. The results confirm a consistent degree of hypomethylation in plasma from ovarian cancer patients. This likely reflects the generally hypomethylated state of the tumors.

Example Embodiment L of Example 6: Estimation of Abnormal Plasma Methylome Variation in Targeted Regions for Diagnosis of Ovarian Cancer

Provided below is an algorithm that can be used to diagnose a subject with ovarian cancer. The presently disclosed subject matter provides that the methylome(s) of ovaries of a mammal, or structures therein, could be affected by certain abnormalities (e.g., ovarian cancer), and that the changes of these methylomes can lead to changes in the methylation patterns of the DNA fragments found in plasma, which are released by ovary tissues. An algorithm was developed to identify the changes of methylation patterns in the methylome of plasma caused by ovary phenotypes. One insight behind this algorithm was that the methylome of the DNA fragments in plasma is a mixture of a variety of component methylomes of ovarian and other origins, and that the proportion of these different component methylomes in the mixture varies from subject to subject, even among the population with normal ovary phenotype. By constructing a model of plasma methylome as a linear combination of various component methylomes of ovary and other origins, the algorithm can accurately predict the methylation patterns of a new plasma sample under the hypothesis that it is from a normal individual. Consequently, the algorithm has high sensitivity for detecting abnormal methylation patterns in a plasma sample caused by changes of the methylomes of some ovarian/fallopian or other relevant tissue when the sample is from an affected individual.

The procedure can be applied with little modification to the diagnosis and phenotyping of ovarian cancer using other types of biopsy samples, such as cervical swabs, urine and peritoneal fluid, provided that the DNA fragments from the tissues affected by ovarian cancer can be found in those biopsy samples.

Let i be any CpG site in human genome, z_i,j be the methylation level of CpG site i in a plasma sample j, p_i,r,j be the proportion of the r^th component methylome m_r,j of ovarian/fallopian or other relevant tissue origin in plasma sample j at site i, m_i,r,j be the methylation level of CpG i in methylome m_r,j. The hypothesis is:

Z^i,j=Σ_r=1^Rp_i,r,jm_i,r,j  (1)

where p_i,r,j, m_i,r,j>=0, m_i,r,j<=1, p_i,1,j+ . . . p_i,R,j=1.

It is further assumed that there is a set of CpG sites S such that, for any CpG site i in S, and any plasma j from a normal individual, it has m_I,r,j=m_I,r and p_I,r,j=p_r,j.

That is, it is assumed that in any plasma sample from a normal individual, the proportions of different component methylomes in the mixture are the same for all CpG sites in S. It is also assumed that, by restricting to the set of CpG sites S, plasma samples from all normal individuals have the same set of component methylomes. They are called restricted reference component methylomes (RRCM), and are labeled as m₁^S, . . . , m_R^S or simply m₁, . . . , m_R when there is no confusion. For any plasma sample j from a normal individual, its methylome restricted to set of CpG sites in S can be expressed as a weighted average of the restricted reference component methylomes. More precisely, let z_j^S be the methylome of plasma sample C restricted to S, then for some mixture vector p_j=[p_j,l . . . , p_j,R]^T, it has:

z_j^s=[m₁^S, . . . m_R^S]p_j  (2)

Finally, it is assumed that the set S is the union of two disjoint subsets C and T, where T is a union of K non-empty sets T_k such that T=U_k=1^KT_k where the index k represents the k^th type of abnormal tissue phenotype. T_k's do not need to be disjoint. Moreover, T_k itself is the union of two disjoint sets D_k and V_k. Either D_k or V_k could be empty, but not both. It is assumed that for any plasma sample, including one from an abnormal individual, when restricted to CpG sites in C, its methylome can always be expressed as a weighted average of the restricted reference component methylomes. That is, it has: z_j^C=[m₁^C, . . . , m_R^C]p_j regardless whether j is from an abnormal individual. C is called the set of reference CpG sites. On the other hand, for a plasma sample l from an abnormal individual, when restricted to CpG sites in S=CUT, its methylome can no longer be expressed as a weighted average of the restricted reference component methylomes. That is, it has: w₁^S≠[m₁^S, . . . , m_R^S]p_l for any mixture vector p_l. More specifically, for a plasma sample l from an individual with the k^th type of abnormal phenotype, it has: 1), w_j^C=[m₁^C, . . . , m_R^C], 2), if D_K is non-empty w_lDK=[m_1,k^Dk, . . . , m_R,k^Dk]p_l such that [m₁^Dk, . . . , m_R^Dk]≠[m_1,k^Dk, . . . , m_R,k^Dk], and 3), if V_k is non-empty, then w_l^Vk=m₁^Vk, . . . , m_R^Vkυq_l such that p_l≠q_l. In other words, in a plasma sample from the k^th type of abnormal individual, if the set D_k is not empty, the component methylomes of the sample l restricted to D_k are no longer the same as the reference component methylome restricted to D_k. If the set V_k is not empty, in this plasma sample, the proportion of the reference component methylomes restricted to V_k is no longer the same as the proportion of the reference component methylome restricted to R.

T is called the target set of CpG sites, D_k is called the differential methylation target set, V_k is called the copy number variation target set, and T_k is called the target set for the k^th type of abnormal phenotype.

The main steps of the algorithm of this Example are:

• • 1) Identify the sets of reference CpG sites C, and T₁, . . . , T_K for the list of K types of abnormal individuals.
  • 2) Estimate the restricted reference component methylomes m₁, . . . , m_R, or R predictor methylomes n₁, . . . , n_R that are independent linear combinations of the reference component methylomes such that n_r=[m₁, . . . , m_R]q_r for R linearly independent mixture vectors q₁, . . . , q_R.
  • 3) (Optional) If the reference component methylomes are available, estimate the proportions of these components at the reference CpG sites C for the test plasma samples.
  • 4) Predict the methylation level of the test plasma samples at the target set T_k of CpG sites, under the hypothesis that the sample is from a normal individual.
  • 5) Compare the predicted methylation levels at D_k and V_k against the observed methylation levels, and reject the null hypothesis that a test sample is from a normal individual if the observed methylation levels are significantly different form the predicted levels.

The algorithm of this Example can be implemented in a variety of ways. For example, given the methyl-seq data for a set of plasma samples from normal individuals, the presently disclosed EM algorithm or the data augmentation method can be applied to estimate the component methylomes, then use the maximum likelihood method to estimate the proportion of these component methylomes in the test sample. Below are exemplary simple implementations of the presently disclosed algorithm that use linear regression.

In the simplest implementation of the algorithm of this Example, it is assumed the restricted methylome of a plasma sample from a normal individual can be approximated by a mixture of two restricted reference methylomes. It is further assumed that the estimations of these two reference component methylomes are available. For example, in the implementation below, for the genomic loci of interest, the plasma methylome is approximated by the mixture of leukocyte and ovarian/fallopian or other relevant tissue methylomes. The implementation of the algorithm includes the following steps:

1. Identify the Reference Set C, and the Target Sets T₁, . . . , T_K.

• • 1.1 Collect the methylation data for a set of leukocyte samples, a set of ovarian/fallopian or other relevant tissue/cell samples, and a set of plasma samples, all from normal individuals. For each type of abnormal individuals, collect a set of leukocyte-derived samples, a set of ovarian/fallopian or other relevant tissue/cell samples, and a set of plasma samples from that type of abnormal individuals. All these samples should have matched age, race, and other relevant parameters. These are the training data.
  • 1.2 Let x_i,j be the observed methylation level of CpG site i in a normal leukocyte-derived sample j, and y_i,l the observed methylation level of CpG site i in a normal ovarian/fallopian or other relevant tissue/cell sample l, s_x,i² the sample variance of x_i over all normal leukocyte-derived samples, s_y,i² the sample variance of y_i,j over all normal ovarian/fallopian or other relevant tissue/cell samples. Identify the CpG sites S₀ such that for any i∈S₀, it has both s_x,i²<c₀ and s_y,i²<c₀ for some constant c₀. These are CpG sites with stable methylation levels in each type of normal cells.
  • 1.3 Let x_i,j be the observed methylation level of CpG site i in a leukocyte-derived sample j, including normal and abnormal, and y_i,l the observed methylation level of CpG site i in a ovarian/fallopian or other relevant tissue/cell sample l, including normal and abnormal, s_x,i² the sample variance of x_i,l over all leukocyte-derived samples, including normal and abnormal, s_y,i² the sample variance of y_i,j over all ovarian/fallopian or other relevant tissue/cell samples, including normal and abnormal. Identify the CpG sites S₁ such that for any i∈S₁, it has both s_x,i²<c₀ and s_y,i²<c₀ for some constant c₀, and that the statistical test for the difference between {x_i,j0: j0 is a normal leukocyte—derived sample} and {x_i,jk: jk is an abnormal leukocyte—derived sample of type k} is not significant for all abnormal types of leukocyte-derived, and that the statistical test for the difference between {y_i,j0: j0 is a normal ovarian/fallopian or other relevant tissue/cell sample} and {y_i,jk: jk is an abnormal ovarian/fallopian or other relevant tissue/cell sample of type k} is not significant for all abnormal types of ovarian/fallopian or other relevant tissue/cell. These are CpG sites with stable methylation levels in each type of cells, and with no difference in methylation level between normal and any abnormal samples. Let x_i be the sample mean of x_i,j over all leukocyte-derived samples, including normal and abnormal, y_i the sample mean of y_i,j over all ovarian/fallopian or other relevant/cell samples, including normal and abnormal. Identify the subset C₀ of S₁ such that for any i∈C₀, it has |x_i−y_i|>c₁ for some constant c₁. These are CpG sites that are stably methylated in each cell type, with no difference between the normal and abnormal samples of the same cell type, and differentially methylated between different types of cells.
  • 1.4 Let x^R0 be the vector of x_i for all i∈C₀, and y^C0 be the vector of y_i for all i∈C₀, where x_i is the mean methylation at site i in all leukocyte-derived samples y_i the mean methylation at site i in all ovarian/fallopian or other relevant tissue/cell samples. Note that by the way the set C₀ is selected, there is no difference in the methylation level of any CpG sites in C₀ between normal and abnormal leukocyte-derived samples, or between normal and abnormal ovarian/fallopian or other relevant tissue/cell samples. Let z_j^C0 be the observed methylation levels of CpG sites in C₀ for a plasma sample j of the k^th abnormal type. (For convenience, the normal plasma sample is called as sample of the 0th abnormal type). For each sample j belonging to the k^th abnormal type, regress z_j^C0 against x^C0 and y^C0, with the constraints that the intercept must be 0, and the coefficients must be non-negative and add to 1, and get the residual e_j^C0. Identify the subset C₀^k of C₀ such that for any CpG i in C₀^k, it has

$\frac{e_{i,k}^2}{s_{i,k}}<c_2,$

and e_i,k²<c₃ for some constants c₂ and c₃, where e_i,k² is the mean of the squared difference between estimated and observed methylation levels of CpG site i in all plasma samples of the k^th abnormal type, and s_i,k² the sample variances of methylation levels of CpG site i in the same set of plasma samples. Repeat the above procedure for each type of abnormal plasma samples, the intersection of the subsets C=∩_k=0^KC₀^k is the reference set of CpG sites. These are CpG sites where their methylation levels in both normal and any type of abnormal plasma samples can be accurately predicted by the reference component methylomes from normal individuals.

• • 1.5 Let T₀=S₀\ S₁. Let x^C and x^T0 be the vectors of x_i and x_h for all i∈C and h∈T₀ respectively, and y^C and y^T0 be the vectors of y_i and y_h for all i∈C and h∈T₀ respectively, where x_i, x_h, y_i, and y_h are mean methylation level of sites for a normal leukocyte-derived or ovarian/fallopian or other relevant tissue/cell at sites i and h respectively. Let z_j^C and z_j^T0 and be the observed methylation levels of CpG sites in C and T₀ respectively for a normal plasma sample j, w_lk^C and w_lk^T0 the observed methylation level of CpG sites in C and T₀ respectively for a plasma sample l_k from an individual with the k^th type of abnormality, w_lg^C and w_lk^T0 the observed methylation level of CpG sites in C and T₀ respectively for a plasma sample l_g from an individual with the g^th type of abnormality, where g≠k. For each j, l_k, and l_g, regress z_j^C, w_lk^C, and w_lg^C respectively against x^C and y^C, with the constraints that the intercept must be 0, and the coefficients must be non-negative and add to 1. Apply the fitted models respectively to x^T0 and y^T0 to predict z_j^T0, w_lk^T0 and W_lg^T0 respectively, and get the differences e_j^T0, e_lk^T0 and e_lg^T0 between the predicted values and observed values. Let e_i, e_i,k, and e_i,g be the means of the sets of differences {e_j^T0: j is a normal plasma sample}, {e_lk^T0: l_k is a plasma sample of th k^th abnormal type} and {e_lg^T0: l_g is a plasma sample of the g^th abnormal type} for CpG site i respectively. Identify the subset T_k of T₀ such that for any i∈T_k, it has led <c_2,0, |e_i|<c_2,0, |e_i,k|>c_2,k, and |e_i,k−e_i,g|>c_3,k, for some constants c_2,0, c_2,k, and c_3,k, for all g≠k. T_k is the target set for the k^th type of the abnormal individual. These are the sites where the methylation of a normal plasma sample can be accurately predicted, the observed methylation in a plasma sample of the k^th abnormal type will deviate from the prediction, and deviation will be different from that of a plasma sample of any other abnormal type.

2. Estimate Fraction of the New Plasma Samples to be Tested

Recall that x^c and y^c are mean vectors of the methylation levels of the training leukocyte-derived and training ovarian/fallopian or other relevant tissue/cell data for the CpG sites in the reference set C. For any new plasma sample t to be tested, let z_t^C be the observed methylation levels of CpG sites in C. Regress z_t^C against x^C and y^C, with the constraints that the intercept must be 0, and the coefficients must be non-negative and add to 1. The estimated coefficients are the estimated fractions of the component methylomes for the plasma sample t.

3. Test if the New Plasma Samples are from the k^th Type of Abnormal Individual.

For the new plasma sample t, let x^Tk and y^Tk be mean vectors of the methylation levels of the training leukocyte-derived and training ovarian/fallopian or other relevant tissue/cell data for the CpG sites in the target set T_k identified in step 1 of this algorithm, apply the fitted regression models obtained from the step 2 of this algorithm to x^Tk and y^Tk to predict the methylation levels of CpG sites in T_k for sample t under the hypothesis that sample t is from a normal. Let n_k be the number of CpG sites in T_k. Define functions f_k(x₁, . . . , x_nk)=Σ_i(−1)^{I_(ei,k−ei)}x_i and f_k,g(x₁, . . . , x_nk)=Σ_i(−1)^{I_(ei,k−ei,g)}x_i, where I_(⋅)=I_(−∞,0)(⋅), that is, the indicator function for the interval (−∞, 0), e_i, e_i,k and e_i,g are estimations obtained from step 1.5 of the algorithm. It will be said the sample is from an individual with the k^th type of abnormal phenotype if f_k(e_1,t−e₁, . . . , e_nk,t−e_nk)>c_4,k, and f_k,g(e_1,t−e_1,g, . . . , e_nk,t−e_nk,g)>c_5,g for all g≠k, where e_i,t is the difference between the observed methylation level of the CpG site i∈T_k for sample t and the predicted value by the fitted model obtained from step 2, and g≠k is any type of abnormal phenotype that is different form the k^th type of abnormal phenotype.

Other ways of implementing the algorithm of this Example can be developed by modifying the simple implementation presented above. Specifically, it does not need to assume that there are only two component reference methylomes that make up the plasma methylomes, nor does it need to approximate them by mixtures of the component methylomes. Instead, a set of predictor methylomes can be collected that are themselves mixtures of component reference genomes, as long as the number of the predictor methylomes is the same as the number of the reference component methylomes, and the mixture vectors of the predictor methylomes are linearly independent. For example, they can be methylomes of plasma samples with known different proportion of leukocyte-derived and ovarian/fallopian or other relevant tissue/cell DNAs.

In the algorithm of this Example, the difference between observed methylation levels in certain target regions and the predicted methylation levels as the test statistic to determine if in a plasma sample the methylome has been affected by ovarian cancer. To illustrate the advantage of this approach, it is assumed that the mixture vector p₁ for the methylome of a normal plasma sample j followed a Dirichlet's distribution with parameters α₁= . . . =α_R. Furthermore, for CpG site i, its methylation levels in the R reference vector p_i for component methylomes are m_i,r=(r−1)/(R−1). It can be shown that the methylation level of i in sample j then has a mean of 0.5, and a variance of

$\frac{R+1}{12\left(R-1\right)\left(R{\alpha }_1+1\right)}.$

If there is a methyl-seq library of sample j with a coverage of N for CpG site i, the variance of the measured methylation level z_i,j is

${\sigma }_1^2=\frac{1}{4N}+\frac{N-1}{N}\frac{R+1}{12\left(R-1\right)\left(R{\alpha }_1+1\right)}.$

In other words, if z_i,j is used as a test statistic to detect and phenotype ovarian cancer using a plasma sample, under the null hypothesis, the test statistic has a variance of σ₁². However, in the algorithm of this Example, it is first estimated the mixture vector p_j, then predicted z_i,j by Σ_rm_i,rp_r,j. Note that in methyl-seq data, millions of CpG sites can be contained in each library, and that the variance of the coefficients in a linear regression model is inversely proportional to sample size. Thus it is possible to get highly accurate estimation of the mixture vector p_j, even if it is taken into account that adjacent CpG sites tend to have correlated methylation levels. Assuming an accurate estimate of Σ_rm_i,rp_r,j can be obtained, that is, the error of the estimation can be ignored, the variance of the difference z_i,j−Σ_rm_i,r p_r,j between the observed methylation level and the prediction will De

$\frac{1}{4N}-\frac{1}{N}\frac{R+1}{12\left(R-1\right)\left(R{\alpha }_1+1\right)}.$

In other words, under the null hypothesis, the test static z_i,j−Σ_r m_i,r p_r,j used in the presently disclosed algorithm has a much smaller variance than the other candidate test statistic z_i,j. This in turns means that the presently disclosed test will achieve a higher power at the same level of type I error.

Examples of Embodiments of Example 6

F1. A method for diagnosing, prognosing, classifying, and/or monitoring ovarian cancer in a mammal, comprising:

(a) obtaining a sample from the mammal;

(b) determining the methylation status and/or level of one or more genomic loci in the sample;

(c) comparing the methylation status and/or level of the one or more genomic loci to a reference or a set of predicted values; and

(d) diagnosing ovarian cancer in the mammal,

wherein the difference in the methylation status and/or level of the one or more genomic loci in the sample compared to the reference or the set of predicted values indicates the presence of ovarian cancer in the mammal.

F2. The method of embodiment F1, wherein an increase in the level of methylation of the one or more genomic loci in the sample indicates the presence of ovarian cancer in the mammal or a decrease in the level of methylation of the one or more genomic loci in the sample indicates the presence of ovarian cancer in the mammal.
F3. The method of embodiment F1, wherein a decrease in the level of methylation of at least one of the one or more genomic loci in the sample and an increase in the level of methylation of at least one of the one or more genomic loci in the sample indicates the presence of ovarian cancer in the mammal.
F4. A method of treating ovarian cancer in a mammal, comprising:

(a) obtaining a sample from the mammal;

(b) determining the methylation status and/or level of one or more genomic loci present in the sample;

(c) comparing the methylation status and/or level of the one or more genomic loci to a reference or a set of predicted values;

(d) diagnosing ovarian cancer in the mammal, wherein the difference in the methylation status and/or level of the one or more genomic loci in the sample compared to the reference or the set of predicted values indicates the presence of ovarian cancer in the mammal; and

(e) administering a chemotherapy, an immunotherapy, or both to said mammal.

F5. The method of any one of embodiments F1-F4, wherein the reference is the methylation status and/or level of the one or more genomic loci in a sample obtained from a mammal that does not have ovarian cancer.
F6. The method of any one of embodiments F1-F5, wherein said sample is a plasma sample.
F7. A method of treating ovarian cancer comprising:

(a) measuring the methylation status and/or level of one or more genomic loci present in a sample from a mammal prior to a treatment of ovarian cancer;

(b) measuring the methylation status and/or level of one or more genomic loci present in a sample from the mammal during the treatment of ovarian cancer; and

(c) continuing the treatment if the difference in the methylation status and/or level of the one or more genomic loci between the samples from prior to and during the treatment of ovarian cancer indicates the mammal is responsive to the treatment.

F8. The method of embodiment F7, further comprising (d) administering a different treatment to the mammal if the difference in the methylation status and/or level of the one or more genomic loci between the samples from prior to and during the treatment of ovarian cancer indicates the mammal is not responsive to the treatment.
F9. The method of embodiment F7, wherein an increase in the level of methylation of the one or more genomic loci in the sample indicates the mammal is responsive to the treatment, or a decrease in the level of methylation of the one or more genomic loci in the sample indicates the mammal is responsive to the treatment.
F10. The method of embodiment F7, wherein a decrease in the level of methylation of at least one of the one or more genomic loci in the sample and an increase in the level of methylation of at least one of the one or more genomic loci in the sample indicates the mammal is responsive to the treatment.
F11. The method of embodiment F7, wherein an increase in the level of methylation of the one or more genomic loci in the sample indicates the mammal is not responsive to the treatment, or a decrease in the level of methylation of the one or more genomic loci in the sample indicates the mammal is not responsive to the treatment.
F12. The method of embodiment F7, wherein a decrease in the level of methylation of at least one of the one or more genomic loci in the sample and an increase in the level of methylation of at least one of the one or more genomic loci in the sample indicates the mammal is not responsive to the treatment.
F13. The method of any one of embodiments F7-F12, wherein said sample is a plasma sample.
F14. The method of any one of embodiments F1-F13, wherein the one or more genomic loci comprise one or more CpG sites.
F15. The method of any one of embodiments F1-F14, wherein the one or more genomic loci are present within nucleic acids isolated from the sample.
F16. The method of any one of embodiments F1-F15, wherein the one or more genomic loci are present within cell-free nucleic acids isolated from the sample.
f17. A method of treating ovarian cancer, comprising;

(a) diagnosing ovarian cancer in a mammal by utilization of the algorithm disclosed in Example Embodiment L; and

(b) administering a chemotherapy, an immunotherapy, or both to said mammal to treat said ovarian cancer.

F18. The method of any one of embodiments F1-F17, wherein said mammal is a human.
F19. A kit for diagnosing, prognosing, and/or monitoring ovarian cancer in a mammal comprising a means for determining and/or detecting the methylation status of one or more genomic loci.
F20. The kit of embodiment F19, wherein the means comprises one or more primers and/or probes for determining and/or detecting the methylation status of the one or more genomic loci.

Abstract of this Example (Example 6)

Example 6 provides methods for diagnosing, prognosing, monitoring, classifying, and/or treating ovarian cancer. For example, algorithms, kits, and methods for diagnosing, prognosing, monitoring, classifying, and/or treating ovarian cancer are provided.

Other Embodiments

It is to be understood that while the invention has been described in conjunction with the detailed description thereof, the foregoing description is intended to illustrate and not limit the scope of the invention, which is defined by the scope of the appended claims. Other aspects, advantages, and modifications are within the scope of the following claims.

Claims

1. A computer-implemented method, comprising:

obtaining, by a computing system, initial sequence data that describes sequences of an initial set of nucleic acids from a biological sample of a person, the initial set of nucleic acids including nucleic acids originating from a plurality of different tissues of the person;

filtering, by the computing system, the initial sequence data to generate filtered sequence data that describes sequences of a filtered subset of nucleic acids from the biological sample, wherein the filtering includes (i) selecting target nucleic acids from the initial set of nucleic acids based on at least one of a methylation characteristic or a copy number characteristic of the target nucleic acids and (ii) enriching the target nucleic acids in the filtered subset;

determining, by the computing system, a methylation profile for the filtered subset of nucleic acids from the biological sample;

processing, by the computing system, the methylation profile for the filtered subset of nucleic acids to determine a likelihood that the person has a specified medical condition; and

outputting, by the computing system, an indication of the likelihood that the person has the specified medical condition.

2. The computer-implemented method of claim 1, further comprising identifying a pre-defined set of genomic regions;

wherein selecting target nucleic acids from the initial set of nucleic acids comprises comparing nucleic acid sequences from the initial set of nucleic acids to sequences from the pre-defined set of genomic regions; and

wherein enriching the target nucleic acids in the filtered subset comprises discarding nucleic acid sequences from the initial sequence data that are not among the sequences from the pre-defined set of genomic regions, while retaining nucleic acid sequences from the initial sequence data that are among the sequences from the pre-defined set of genomic regions.

3. The computer-implemented method of claim 2, wherein at least a first subset of the pre-defined set of genomic regions are defined based on the regions in the first subset exhibiting a minimum level of stability with respect to at least one of the methylation characteristic or the copy number characteristic in a population of individuals.

4. The computer-implemented method of claim 3, wherein at least a second subset of the pre-defined set of genomic regions are defined based on the regions in the second subset exhibiting at least a minimum difference with respect to the methylation characteristic or the copy number characteristic between individuals who have the specified medical condition and individuals who do not have the specified medical condition.

5. The computer-implemented method of claim 1, wherein the biological sample comprises plasma, and the initial set of nucleic acids comprises cell-free DNA in the plasma.

6. The computer-implemented method of claim 1, further comprising:

identifying a set of restricted reference component methylomes in the initial set or filtered subset of nucleic acids;

identifying a set of reference component methylomes;

determining a proportion of the reference component methylomes at a reference set of CpG sites in the initial set or filtered subset of nucleic acids;

generating predictions of methylation levels at a target set of CpG sites in the initial set or filtered subset of nucleic acids;

comparing the predictions of methylation levels at the target set of CpG sites to observed methylation levels; and

determining whether the person likely has or does not have the specified medical condition based on the comparison.

7. The computer-implemented method of claim 1, wherein the biological sample comprises a stool sample or cerebrospinal fluid.

8. The computer-implemented method of claim 1, further comprising:

determining, by the computing system, a copy number profile for the filtered subset of nucleic acids from the biological sample; and

processing, by the computing system, the copy number profile along with the methylation profile for the filtered subset of nucleic acids to determine the likelihood that the person has the specified medical condition.

9. The computer-implemented method of claim 1, wherein the initial set of nucleic acids were treated to facilitate detection of methylated sites before sequencing.

10. The computer-implemented method of claim 1, wherein the specified medical condition is ovarian cancer, endometriosis, necrotizing enterocolitis, fetal aneuploidy, preeclampsia, or a brain condition.

11. The computer-implemented method of claim 1, wherein the methylation profile for the filtered subset of nucleic acids indicates, for each of a plurality of genomic loci, a methylation level of the locus.

12. The computer-implemented method of claim 1, wherein the genomic loci is a CpG site, CpG island, differentially methylated region (DMR), promoter region, enhancer region, or CpG island shore.

13. The computer-implemented method of claim 1, wherein determining the likelihood that the person has the specified medical condition comprises determining a probability that the person has the specified medical condition.

14. The computer-implemented method of claim 1, wherein determining the likelihood that the person has the specified medical condition comprises generating a binary indication that the person either likely has the specified medical condition or likely does not have the specified medical condition.

15. The computer-implemented method of claim 1, wherein processing the methylation profile comprises providing data representing the methylation profile as input to a machine-learning model, and obtaining the likelihood, or a value from which the likelihood is derived, as an output of the machine-learning model.

16. The computer-implemented method of claim 15, wherein the machine-learning model comprises at least one of a classifier, an artificial neural network, a support vector machine, a decision tree, or a regression model.

17. The computer-implemented method of claim 15, wherein the machine-learning model defines reference or predicted methylation profiles against which the methylation profile for the filtered subset are compared to determine the likelihood that the person has the specified medical condition.

18. The computer-implemented method of claim 1, wherein the determined likelihood that the person has the specified medical condition is used by a medical provider to assess whether to perform additional diagnostic testing on the person.

19. The computer-implemented method of claim 1, wherein the determined likelihood that the person has the specified medical condition is used by a medical provider to at least one of diagnose the person or treat the person for the specified medical condition.

20. The computer-implemented method of claim 1, wherein outputting the indication of the likelihood that the person has the specified medical condition comprises at least one of presenting the indication on an electronic display, audibly playing the indication through a speaker, storing the indication in a memory of a computing system for subsequent retrieval, or transmitting the indication in an electronic message to one or more users.

21. The computer-implemented method of claim 1, wherein enriching the target nucleic acids in the filtered subset comprises generating the filtered subset so that a fraction of the target nucleic acids that occur in the filtered subset is greater than a fraction of the target nucleic acids that occur in the initial set of nucleic acids.

22. The computer-implemented method of claim 1, wherein the filtered subset consists exclusively of the target nucleic acids.

23. The computer-implemented method of claim 1, wherein the filtered subset comprises the target nucleic acids and non-targeted nucleic acids.

24-25. (canceled)

26. A computer-implemented method, comprising:

obtaining, by a computing system, initial sequence data that describes sequences of an initial set of nucleic acids from a biological sample of a person, the initial set of nucleic acids including nucleic acids originating from a plurality of different tissues of the person;

filtering, by the computing system, the initial sequence data to identify a first subset of sequences from the initial sequence data that correspond to a first pre-defined set of genomic regions;

filtering, by the computing system, the initial sequence data to identify a second subset of sequences from the initial sequence data that correspond to a second pre-defined set of genomic regions;

processing, by the computing system, data that includes an observed methylation profile of the first subset of sequences to generate a predicted methylation profile of the second subset of sequences;

comparing, by the computing system, an observed methylation profile of the second subset of sequences to the predicted methylation profile of the second subset of sequences to determine whether the person has a specified medical condition, wherein the person is deemed to have the specified medical condition if a difference between the observed methylation profile of the second subset of sequences and the predicted methylation profile of the second subset of sequences meets a minimum difference criterion; and

outputting, by the computing system, an indication of whether the person was determined to have the specified medical condition.

27. The computer-implemented method of claim 26, wherein:

the first pre-defined set of genomic regions are regions that exhibit a minimum level of stability with respect to at least one of a methylation characteristic or a copy number characteristic in a population of individuals; and

the second pre-defined set of genomic regions are regions that exhibit a minimum difference with respect to at least one of the methylation characteristic or the copy number characteristic between a first sub-population of individuals who have the specified medical condition and a second sub-population of individuals who do not have the specified medical condition.

28. The computer-implemented method of claim 26, wherein the first pre-defined set of genomic regions is a first reference set of genomic regions, and the second pre-defined set of genomic regions is a first target set of genomic regions;

the method further comprising: selecting the first reference set of genomic regions as the first pre-defined set of genomic regions from a database that includes a plurality of reference sets of genomic regions, wherein different ones of the plurality of reference sets of genomic regions correspond to different medical conditions; and selecting the first target set of genomic regions as the second pre-defined set of genomic regions from the database, wherein the database further includes a plurality of target sets of genomic regions, wherein different ones of the plurality of target sets of genomic regions correspond to different medical conditions.

29. The computer-implemented method of claim 26, wherein the specified medical condition is preeclampsia, endometriosis, ovarian cancer, necrotizing enterocolitis, or a brain condition.

30-31. (canceled)

32. A computer-implemented method, comprising:

obtaining, by a computing system, initial sequence data that describes sequences of an initial set of nucleic acids from a biological sample of a person, the initial set of nucleic acids including nucleic acids originating from a plurality of different tissues of the person;

filtering, by the computing system, the initial sequence data to identify a target subset of sequences from the initial sequence data that correspond to a pre-defined set of genomic regions;

comparing, by the computing system, an observed methylation profile of the target subset of sequences to a pre-defined methylation profile to determine whether the person has a specified medical condition, wherein the person is deemed to have the specified medical condition if a difference between the observed methylation profile of the target subset of sequences and the pre-defined methylation profile meets a minimum difference criterion; and

outputting, by the computing system, an indication of whether the person was determined to have the specified medical condition.

33-34. (canceled)