Abstract: We describe systems and methods for generating and training convolutional neural networks using biological sequences and relevance scores derived from structural, biochemical, population and evolutionary data. The convolutional neural networks take as input biological sequences and additional information and output molecular phenotypes. Biological sequences may include DNA, RNA and protein sequences. Molecular phenotypes may include protein-DNA interactions, protein-RNA interactions, protein-protein interactions, splicing patterns, polyadenylation patterns, and microRNA-RNA interactions, which may be described using numerical, categorical or ordinal attributes. Intermediate layers of the convolutional neural networks are weighted using relevance score sequences, for example, conservation tracks.

Abstract: Techniques for determining one or more cell composition percentages from expression data. The techniques include obtaining expression data for a biological sample, the biological sample previously obtained from a subject, the expression data including first expression data associated with a first set of genes associated with a first cell type; determining a first cell composition percentage for the first cell type using the expression data and one or more non-linear regression models including a first non-linear regression model, wherein the first cell composition percentage indicates an estimated percentage of cells of the first cell type in the biological sample, wherein determining the first cell composition percentage for the first cell type comprises: processing the first expression data with the first non-linear regression model to determine the first cell composition percentage for the first cell type; and outputting the first cell composition percentage.

Abstract: Amino acid sequences of antibodies can be generated using a generative adversarial network that includes a first generating component that generates amino acid sequences of antibody light chains and a second generating component generates amino acid sequences of antibody heavy chains. Amino acid sequences of antibodies can be produced by combining the respective amino acid sequences produced by the first generating component and the second generating component. The training of the first generating component and the second generating component can proceed at different rates. Additionally, the antibody amino acids produced by combining amino acid sequences from the first generating component and the second generating component may be evaluated according to complentarity-determining regions of the antibody amino acid sequences.

Abstract: The present disclosure describes methods for determining the functional consequences of mutations. The methods include the use of machine learning to identify and quantify features of all atom molecular dynamics simulations to obtain the disruptive severity of genetic variants on molecular function.

Abstract: Systems and methods for analyzing overlapping sequence information can obtain first and second overlapping sequence information for a polynucleotide, align the first and second sequence information, determine a degree of agreement between the first and second sequence information for a location along the polynucleotide, and determine a base call and a quality value for the location.

Abstract: Provided herein are methods, processes and apparatuses for non-invasive assessment of genetic variations.

Abstract: Systems, methods, and analytical approaches for short read sequence assembly and for the detection of insertions and deletions (indels) in a reference genome. A method suitable for software implementation is presented in which indels may be readily identified in a computationally efficient manner.

Abstract: The technology disclosed relates to determining pathogenicity of nucleotide variants. In particular, the technology disclosed relates to specifying a particular amino acid at a particular position in a protein as a gap amino acid, and specifying remaining amino acids at remaining positions in the protein as non-gap amino acids. The technology disclosed further relates to generating a gapped spatial representation of the protein that includes spatial configurations of the non-gap amino acids, and excludes a spatial configuration of the gap amino acid, and determining a pathogenicity of a nucleotide variant based at least in part on the gapped spatial representation, and a representation of an alternate amino acid created by the nucleotide variant at the particular position.

Abstract: The invention provides methods, systems, and computer readable medium for detecting ploidy of chromosome segments or entire chromosomes, for detecting single nucleotide variants and for detecting both ploidy of chromosome segments and single nucleotide variants. In some aspects, the invention provides methods, systems, and computer readable medium for detecting cancer or a chromosomal abnormality in a gestating fetus.

Abstract: Embodiments are directed to systems and methods for pathogen detection using next-generation sequencing (NGS) analysis of a sample. Embodiments may apply alignment algorithms (e.g., SNAP and/or RAPSearch alignment algorithms) to align individual sequence reads from a sample in a next-generation sequencing (NGS) dataset against reference genome entries in a classified reference genome database. Embodiments of the present invention may include classifying, filtering, and displaying results to a clinician that can then quickly and easily obtain the results of the sequencing to identify a pathogen or other genetic material in a sample that is being tested. A negative sample and a corresponding database can be used to remove contaminants from a list of candidate pathogens. Thus, embodiments are directed to a system that is configured to filter the results of a sequencing alignment and classify a sample quickly.

Abstract: The present disclosure provides methods of estimating a degree of ancestral relatedness between individuals. In an aspect, a method comprises receiving haplotype data comprising genetic markers shared among a population of individuals; dividing the haplotype data into segments based on the genetic markers; for each of the population of test individuals: (i) based on the genetic markers, matching segments of the haplotype data that are identical-by-descent between two individuals, (ii) for each of the matched segments: dividing the matched segment into discrete genomic intervals, scoring each of the discrete genomic intervals based on a degree of matching within or between the individuals, correcting the scores for consistency, and (iii) calculating a weighted sum over the discrete genomic intervals of the matched segment, based on the corrected scores and assigned weights; and (d) estimating the degree of ancestral relatedness between the individuals based on the weighted sums of the matched segments.

Abstract: Provided herein are methods for determining fetal ploidy according to nucleic acid sequence reads. Nucleic acid sequence reads may be obtained from test sample nucleic acid comprising circulating cell-free nucleic acid from the blood of a pregnant female bearing a fetus. Fetal ploidy may be determined according to genomic section levels and a fraction of fetal nucleic acid in a test sample.

Abstract: A method for nucleic acid sequencing includes receiving a plurality of observed or measured signals indicative of a parameter observed or measured for a plurality of defined spaces; determining, for at least some of the defined spaces, whether the defined space comprises one or more sample nucleic acids; processing, for at least some of the defined spaces, the observed or measured signal to improve a quality of the observed or measured signal; generating, for at least some of the defined spaces, a set of candidate sequences of bases for the defined space using one or more metrics adapted to associate a score or penalty to the candidate sequences of bases; and selecting the candidate sequence leading to a highest score or a lowest penalty as corresponding to the correct sequence for the one or more sample nucleic acids in the defined space.

Abstract: Methods and devices to detect analyte in body fluid are provided. Embodiments include processing sampled data from analyte sensor, determining a single, fixed, normal sensitivity value associated with the analyte sensor, estimating a windowed offset value associated with the analyte sensor for each available sampled data cluster, computing a time varying offset based on the estimated windowed offset value, and applying the time varying offset and the determined normal sensitivity value to the processed sampled data to estimate an analyte level for the sensor.

Abstract: Methods are provided to improve the positive predictive value for cancer detection using cell-free nucleic acid samples. Various embodiments are directed to applications (e.g., diagnostic applications) of the analysis of the fragmentation patterns and size of cell-free DNA, e.g., plasma DNA and serum DNA, including nucleic acids from pathogens, including viruses. Embodiments of one application can determine if a subject has a particular condition. For example, a method of present disclosure can determine if a subject has cancer or a tumor, or other pathology. Embodiments of another application can be used to assess the stage of a condition, or the progression of a condition over time. For example, a method of the present disclosure may be used to determine a stage of cancer in a subject, or the progression of cancer in a subject over time (e.g., using samples obtained from a subject at different times).

Abstract: A method of estimating a parameter related to sequencing of a sample nucleic acid template includes: receiving signal data relating to nucleotide incorporation events resulting from a series of flows of nucleotides onto an array of wells including (i) a first well containing the sample nucleic acid template and (ii) a plurality of other sample-containing wells; determining sequence information for the sample nucleic acid template using signal data from the first well; and constructing a phase-state model for a set of nucleotide flows that contributed at least in part to the sequence information, wherein the model includes a signal correction parameter that is determined using signal data from the plurality of other sample-containing wells.

Abstract: A computer-implemented method for processing and/or analyzing nucleic acid sequencing data comprises receiving a first data input and a second data input. The first data input comprises untargeted sequencing data generated from a first nucleic acid sample obtained from a subject. The second data input comprises target-specific sequencing data generated from a second nucleic acid sample obtained from the subject. Next, with the aid of a computer processor, the first data input and the second data input are combined to produce a combined data set. Next, an output derived from the combined data set is generated. The output is indicative of the presence or absence of one or more polymorphisms of the first nucleic acid sample and/or the second nucleic acid sample.

Abstract: Processes and materials to detect cancer from a biopsy are described. In some cases, cell-free nucleic acids can be sequenced, and the sequencing result can be utilized to detect sequences derived from a neoplasm. Detection of somatic variants occurring in phase can indicate the presence of cancer in a diagnostic scan and a clinical intervention can be performed.

Abstract: A system for smart pooling includes a computing device configured to obtain a feature datum, identify a predictive prevalence value as a function of the feature datum, wherein identifying the predictive prevalence value further comprises receiving a predictive training set correlating the feature datum with a probabilistic outcome, training a predictive machine-learning model as a function of the predictive training set, and identifying the predictive prevalence value as a function of the trained predictive machine-learning model and the feature datum, and determine an enhanced well count, wherein determining the enhanced well count further comprises generating a pooling threshold, and determining the enhanced well count as a function of the pooling threshold and the predictive prevalence value.

Abstract: The present invention provides assay systems and methods for determining the percent fetal contribution of cell-free DNA in a maternal sample from a pregnant female with an egg donor pregnancy. Further provided, are assay systems and methods for determining a statistical likelihood of the presence or absence of a fetal aneuploidy in a maternal sample using a determined percent fetal cell-free DNA in the sample.