FINDING RELATIVES IN A DATABASE
Determining relative relationships of people who share a common ancestor within at least a threshold number of generations includes: receiving recombinable deoxyribonucleic acid (DNA) sequence information of a first user and recombinable DNA sequence information of a plurality of users; processing, using one or more computer processors, the recombinable DNA sequence information of the plurality of users in parallel; determining, based at least in part on a result of processing the recombinable DNA information of the plurality of users in parallel, a predicted degree of relationship between the first user and a user among the plurality of users, the predicted degree of relative relationship corresponding to a number of generations within which the first user and the second user share a common ancestor.
This application is a continuation of and claims priority to U.S. patent application Ser. No. 15/264,493, entitled FINDING RELATIVES IN A DATABASE, filed Sep. 13, 2016, which is a continuation of U.S. patent application Ser. No. 13/871,744 (now abandoned), entitled FINDING RELATIVES IN A DATABASE, filed Apr. 26, 2013, which is a continuation of U.S. patent application Ser. No. 12/644,791, entitled FINDING RELATIVES IN A DATABASE, filed Dec. 22, 2009, which claims the benefit of U.S. Provisional Patent Application No. 61/204,195, entitled FINDING RELATIVES IN A DATABASE OF USERS, filed Dec. 31, 2008; all of the priority documents above are incorporated in their entireties by reference for all purposes.BACKGROUND OF THE INVENTION
Genealogy is the study of the history of families and the line of descent from ancestors. It is an interesting subject studied by many professionals as well as hobbyists. Traditional genealogical study techniques typically involve constructing family trees based on surnames and historical records. As gene sequencing technology becomes more accessible, there has been growing interest in genetic ancestry testing in recent years.
Existing genetic ancestry testing techniques are typically based on deoxyribonucleic acid (DNA) information of the Y chromosome (Y-DNA) or DNA information of the mitochondria (mtDNA). Aside from a small amount of mutation, the Y-DNA is passed down unchanged from father to son and therefore is useful for testing patrilineal ancestry of a man. The mtDNA is passed down mostly unchanged from mother to children and therefore is useful for testing a person's matrilineal ancestry. These techniques are found to be effective for identifying individuals that are related many generations ago (e.g., 10 generations or more), but are typically less effective for identifying closer relationships. Further, many relationships that are not strictly patrilineal or matrilineal cannot be easily detected by the existing techniques.
Various embodiments of the invention are disclosed in the following detailed description and the accompanying drawings.
The invention can be implemented in numerous ways, including as a process; an apparatus; a system; a composition of matter; a computer program product embodied on a computer readable storage medium; and/or a processor, such as a processor configured to execute instructions stored on and/or provided by a memory coupled to the processor. In this specification, these implementations, or any other form that the invention may take, may be referred to as techniques. In general, the order of the steps of disclosed processes may be altered within the scope of the invention. Unless stated otherwise, a component such as a processor or a memory described as being configured to perform a task may be implemented as a general component that is temporarily configured to perform the task at a given time or a specific component that is manufactured to perform the task. As used herein, the term ‘processor’ refers to one or more devices, circuits, and/or processing cores configured to process data, such as computer program instructions.
A detailed description of one or more embodiments of the invention is provided below along with accompanying figures that illustrate the principles of the invention. The invention is described in connection with such embodiments, but the invention is not limited to any embodiment. The scope of the invention is limited only by the claims and the invention encompasses numerous alternatives, modifications and equivalents. Numerous specific details are set forth in the following description in order to provide a thorough understanding of the invention. These details are provided for the purpose of example and the invention may be practiced according to the claims without some or all of these specific details. For the purpose of clarity, technical material that is known in the technical fields related to the invention has not been described in detail so that the invention is not unnecessarily obscured.
Because of recombination and independent assortment of chromosomes, the autosomal DNA and X chromosome DNA (collectively referred to as recombinable DNA) from the parents is shuffled at the next generation, with small amounts of mutation. Thus, only relatives will share long stretches of genome regions where their recombinable DNA is completely or nearly identical. Such regions are referred to as “Identical by Descent” (IBD) regions because they arose from the same DNA sequences in an earlier generation. The relative finder technique described below is based at least in part on locating IBD regions in the recombinable chromosomes of individuals.
In some embodiments, locating IBD regions includes sequencing the entire genomes of the individuals and comparing the genome sequences. In some embodiments, locating IBD regions includes assaying a large number of markers that tend to vary in different individuals and comparing the markers. Examples of such markers include Single Nucleotide Polymorphisms (SNPs), which are points along the genome with two or more common variations; Short Tandem Repeats (STRs), which are repeated patterns of two or more repeated nucleotide sequences adjacent to each other; and Copy-Number Variants (CNVs), which include longer sequences of DNA that could be present in varying numbers in different individuals. Long stretches of DNA sequences from different individuals' genomes in which markers in the same locations are the same or at least compatible indicate that the rest of the sequences, although not assayed directly, are also likely identical.
System 100 shown in this example includes genetic and other additional non-genetic information for many users. By comparing the recombinable DNA information to identify IBD regions between various users, the relative finder system can identify users within the database that are relatives. Since more distant relationships (second cousins or further) are often unknown to the users themselves, the system allows the users to “opt-in” and receive notifications about the existence of relative relationships. Users are also presented with the option of connecting with their newly found relatives.
At 204, a predicted degree of relationship between Alice and Bob is determined. In some embodiments, a range of possible relationships between the users is determined and a prediction of the most likely relationship between the users is made. In some embodiments, it is optionally determined whether the predicted degree of relationship at least meets a threshold. The threshold may be a user configurable value, a system default value, a value configured by the system's operator, or any other appropriate value. For example, Bob may select five generations as the maximum threshold, which means he is interested in discovering relatives with whom the user shares a common ancestor five generations or closer. Alternatively, the system may set a default value minimum of three generations, allowing the users to by default find relatives sharing a common ancestor at least three generations out or beyond. In some embodiments, the system, the user, or both, have the option to set a minimum threshold (e.g., two generations) and a maximum threshold (e.g., six generations) so that the user would discover relatives within a maximum number of generations, but would not be surprised by the discovery of a close relative such as a sibling who was previously unknown to the user.
At 206, Alice or Bob (or both) is notified about her/his relative relationship with the other user. In some embodiments, the system actively notifies the users by sending messages or alerts about the relationship information when it becomes available. Other notification techniques are possible, for example by displaying a list or table of users that are found to be related to the user. Depending on system settings, the potential relatives may be shown anonymously for privacy protection, or shown with visible identities to facilitate making connections. In embodiments where a threshold is set, the user is only notified if the predicted degree of relationship at least meets the threshold. In some embodiments, a user is only notified if both of the user and the potential relative have “opted in” to receive the notification. In various embodiments, the user is notified about certain personal information of the potential relative, the predicted relationship, the possible range of relationships, the amount of DNA matching, or any other appropriate information.
In some embodiments, at 208, the process optionally infers additional relationships or refines estimates of existing relationships between the users based on other relative relationship information, such as the relative relationship information the users have with a third user. For example, although Alice and Bob are only estimated to be 6th cousins after step 204, if among Alice's relatives in the system, a third cousin, Cathy, is also a sibling of Bob's, then Alice and Bob are deemed to be third cousins because of their relative relationships to Cathy. The relative relationships with the third user may be determined based on genetic information and analysis using a process similar to 200, based on non-genetic information such as family tree supplied by one of the users, or both.
In some embodiments, the relatives of the users in the system are optionally checked to infer additional relatives at 210. For example, if Bob is identified as a third cousin of Alice's, then Bob's relatives in the system (such as children, siblings, possibly some of the parents, aunts, uncles, cousins, etc.) are also deemed to be relatives of Alice's. In some embodiments a threshold is applied to limit the relationships within a certain range. Additional notifications about these relatives are optionally generated.
Upon receiving a notification about another user who is a potential relative, the notified user is allowed to make certain choices about how to interact with the potential relative.
Upon receiving the notification, Alice decides that she would like to make a connection with the newly found relative. At 302, an invitation from Alice to Bob inviting Bob to make a connection is generated. In various embodiments, the invitation includes information about how Alice and Bob may be related and any personal information Alice wishes to share such as her own ancestry information. Upon receiving the invitation, Bob can accept the invitation or decline. At 304, an acceptance or a declination is received. If a declination is received, no further action is required. In some embodiments, Alice is notified that a declination has been received. If, however, an acceptance is received, at 306, a connection is made between Alice and Bob. In various embodiments, once a connection is made, the identities and any other sharable personal information (e.g., genetic information, family history, phenotype/traits, etc.) of Alice and Bob are revealed to each other and they may interact with each other. In some embodiments, the connection information is updated in the database.
In some embodiments, a user can discover many potential relatives in the database at once. Additional potential relatives are added as more users join the system and make their genetic information available for the relative finding process.
Many other user interfaces can be used in addition to or as alternatives of the ones shown above. For example, in some embodiments, at least some of the potential relatives are displayed in a family tree.
Determining the relationship between two users in the database is now described. In some embodiments, the determination includes comparing the DNA markers (e.g., SNPs) of two users and identifying IBD regions. The standard SNP based genotyping technology results in genotype calls each having two alleles, one from each half of a chromosome pair. As used herein, a genotype call refers to the identification of the pair of alleles at a particular locus on the chromosome. Genotype calls can be phased or unphased. In phased data, the individual's diploid genotype at a particular locus is resolved into two haplotypes, one for each chromosome. In unphased data, the two alleles are unresolved; in other words, it is uncertain which allele corresponds to which haplotype or chromosome.
The genotype call at a particular SNP location may be a heterozygous call with two different alleles or a homozygous call with two identical alleles. A heterozygous call is represented using two different letters such as AB that correspond to different alleles. Some SNPs are biallelic SNPs with only two possible states for SNPs. Some SNPs have more states, e.g. triallelic. Other representations are possible.
In this example, A is selected to represent an allele with base A and B represents an allele with base G at the SNP location. Other representations are possible. A homozygous call is represented using a pair of identical letters such as AA or BB. The two alleles in a homozygous call are interchangeable because the same allele came from each parent. When two individuals have opposite-homozygous calls at a given SNP location, or, in other words, one person has alleles AA and the other person has alleles BB, it is very likely that the region in which the SNP resides does not have IBD since different alleles came from different ancestors. If, however, the two individuals have compatible calls, that is, both have the same homozygotes (i.e., both people have AA alleles or both have BB alleles), both have heterozygotes (i.e., both people have AB alleles), or one has a heterozygote and the other a homozygote (i.e., one has AB and the other has AA or BB), there is some chance that at least one allele is passed down from the same ancestor and therefore the region in which the SNP resides is IBD. Further, based on statistical computations, if a region has a very low rate of opposite-homozygote occurrence over a substantial distance, it is likely that the individuals inherited the DNA sequence in the region from the same ancestor and the region is therefore deemed to be an IBD region.
In various embodiments, the effects of genotyping error are accounted for and corrected. In some embodiments, certain genotyped SNPs are removed from consideration if there are a large number of Mendelian errors when comparing data from known parent/offspring trios. In some embodiments, SNPs that have a high no-call rate or otherwise failed quality control measures during the assay process are removed. In some embodiments, in an IBD segment, an occasional opposite-homozygote is allowed if there is sufficient opposite-homozygotes-free distance (e.g., at least 3 cM and 300 SNPs) surrounding the opposite-homozygote.
There is a statistical range of possible relationships for the same IBDhalf and shared segment number. In some embodiments, the distribution patterns are determined empirically based on survey of real populations. Different population groups may exhibit different distribution patterns. For example, the level of homozygosity within endogamous populations is found to be higher than in populations receiving gene flow from other groups. In some embodiments, the bounds of particular relationships are estimated using simulations of IBD using generated family trees. Based at least in part on the distribution patterns, the IBDhalf, and shared number of segments, the degree of relationship between two individuals can be estimated.
The amount of IBD sharing is used in some embodiments to identify different population groups. For example, for a given degree of relationship, since Ashkenazi tend to have much more IBD sharing than non-Ashkenazi Europeans, users may be classified as either Ashkenazi or non-Ashkenazi Europeans based on the number and pattern of IBD matches.
In some embodiments, instead of, or in addition to, determining the relationship based on the overall number of IBD segments and percent DNA shared, individual chromosomes are examined to determine the relationship. For example, X chromosome information is received in some embodiments in addition to the autosomal chromosomes. The X chromosomes of the users are also processed to identify IBD. Since one of the X chromosomes in a female user is passed on from her father without recombination, the female inherits one X chromosome from her maternal grandmother and another one from her mother. Thus, the X chromosome undergoes recombination at a slower rate compared to autosomal chromosomes and more distant relationships can be predicted using IBD found on the X chromosomes.
In some embodiments, analyses of mutations within IBD segments can be used to estimate ages of the IBD segments and refine estimates of relationships between users.
In some embodiments, the relationship determined is verified using non-DNA information. For example, the relationship may be checked against the users' family tree information, birth records, or other user information.
In some embodiments, the efficiency of IBD region identification is improved by comparing a user's DNA information with the DNA information of multiple other users in parallel and using bitwise operations.
A bitwise operation is performed on the encoded arrays to determine whether a section of DNA such as the section between locations 806 and 810 includes opposite-homozygous calls. In this example, a bitwise OR operation is performed to generate a result array 824. Any user with no opposite-homozygous calls between beginning location 806 and ending location 816 results in an entry value of 0 in array 824. The corresponding DNA segment, therefore, is deemed as an IBD region for such user and Alice. In contrast, users with opposite-homozygotes result in corresponding entry values of 1 in array 824 and they are deemed not to share IBD with Alice in this region. In the example shown, user 1 shares IBD with Alice while other users do not.
In some embodiments, phased data is used instead of unphased data. These data can come directly from assays that produce phased data, or from statistical processing of unphased data. IBD regions are determined by matching the SNP sequences between users. In some embodiments, sequences of SNPs are stored in dictionaries using a hash-table data structure for the ease of comparison.
In some embodiments, relative relationships found using the techniques described above are used to infer characteristics about the users that are related to each other. In some embodiments, the inferred characteristic is based on non-genetic information pertaining to the related users. For example, if a user is found to have a number of relatives that belong to a particular population group, then an inference is made that the user may also belong to the same population group. In some embodiments, genetic information is used to infer characteristics, in particular characteristics specific to shared IBD segments of the related users. Assume, for example, that Alice has sequenced her entire genome but her relatives in the system have only genotyped SNP data. If Alice's genome sequence indicates that she may have inherited a disease gene, then, with Alice's permission, Alice's relatives who have shared IBD with Alice in the same region that includes the disease gene may be notified that they are at risk for the same disease.
Although the foregoing embodiments have been described in some detail for purposes of clarity of understanding, the invention is not limited to the details provided. There are many alternative ways of implementing the invention. The disclosed embodiments are illustrative and not restrictive.
1. A method for determining relative relationships of people who share a common ancestor within at least a threshold number of generations, comprising:
- receiving recombinable deoxyribonucleic acid (DNA) sequence information of a first user and recombinable DNA sequence information of a plurality of users;
- processing, using one or more computer processors, the recombinable DNA sequence information of the plurality of users in parallel;
- determining, based at least in part on a result of processing the recombinable DNA information of the plurality of users in parallel, a predicted degree of relationship between the first user and a user among the plurality of users, the predicted degree of relative relationship corresponding to a number of generations within which the first user and the second user share a common ancestor.
Filed: Sep 12, 2018
Publication Date: Jan 10, 2019
Inventors: Lawrence Hon (Mountain View, CA), Serge Saxonov (Mountain View, CA), Brian Thomas Naughton (Mountain View, CA), Joanna Louise Mountain (Mountain View, CA), Anne Wojcicki (Mountain View, CA), Linda Avey (Mountain View, CA)
Application Number: 16/129,645