Granted Patent B1

US 12,580,048 · App. 16/844,758 · Granted Mar 17, 2026

Inventors: Chuong Do (Mountain View, CA); Eric Yves Jean-Marc Durand (Sceaux, FR); John Michael Macpherson (Santa Ana, CA)

Assignee: 23andMe Genomics LLC

G16B40/00G06N5/04G06N7/01G06N20/00

Abstract

Claims (50)

1 . A pipelined ancestry prediction system, the system comprising:

one or more processors and one or more memories coupled to the one or more processors, wherein the one or more memories are configured to provide the one or more processors with instructions which when executed cause the one or more processors to:

obtain a first set of phased haplotype data of an individual attributed to a first parent and a second set of phased haplotype data of the individual attributed to a second parent;

divide the first set of phased haplotype data into a first sequence of haplotype windows and the second set of phased haplotype data into a second sequence of haplotype windows corresponding to the first sequence of haplotype windows, each haplotype window comprising a plurality of genetic markers;

determine an initial ancestry classification for each haplotype window based on the plurality of genetic markers in the haplotype window, each haplotype window being classified as associated with a specific geographical region at an initial confidence level, the first and second sequences of haplotype windows having a plurality of different initial ancestry classifications; and

combine a plurality of initial ancestry classifications of a plurality of haplotype windows having initial confidence levels below a required threshold to form an updated ancestry classification for a portion of the individual's genome, wherein:

the updated ancestry classification is for a broader geographical region comprising geographical regions of the plurality of initial ancestry classifications,

the updated ancestry classification has a total confidence level generated from the initial confidence levels of the plurality of initial ancestry classifications, and

the portion of the individual's genome includes the plurality of haplotype windows.

2 . The system of claim 1 , wherein the one or more memories are further configured to provide the one or more processors with instructions which when executed cause the one or more processors to:

recalibrate probabilities that portions of the individual's genome associate with specific ancestries to generate the initial ancestry classifications.

3 . The system of claim 2 , wherein the recalibration is determined based on reference data of actual unadmixed individuals and/or simulated admixed individuals.

4 . The system of claim 1 , wherein the initial ancestry classifications are generated by a learning machine trained on a training set comprising genetic information of a plurality of individuals with known ancestries.

5 . The system of claim 4 , further comprising training the learning machine using the training set.

6 . The system of claim 4 , wherein the learning machine includes a neural network.

7 . The system of claim 4 , wherein the learning machine includes a support vector machine (SVM).

8 . The system of claim 1 , wherein the one or more memories are further configured to provide the one or more processors with instructions which when executed cause the one or more processors to:

store the reclassified ancestry classification to a database, output the reclassified ancestry classification to another application, or both.

9 . The system of claim 8 , wherein the one or more memories are further configured to provide the one or more processors with instructions which when executed cause the one or more processors to:

store the total confidence level associated with the reclassified ancestry classification to the database, output the total confidence level associated with the reclassified ancestry classification to the other application, or both.

10 . The system of claim 1 , further comprising a user interface configured to display the reclassified ancestry classification to the individual.

11 . The system of claim 10 , wherein the user interface displays the reclassified ancestry classification with hierarchical information of different geographical regions pertaining to the individual's ancestry composition.

12 . The system of claim 10 , wherein the user interface is further configured to receive a request from the individual to display the individual's ancestry composition.

13 . A method, implemented using a computer comprising system memory and one or more processors, the method comprising:

obtaining, by the one or more processors, a first set of phased haplotype data of an individual attributed to a first parent and a second set of phased haplotype data of the individual attributed to a second parent;

dividing, by the one or more processors, the first set of phased haplotype data into a first sequence of haplotype windows and the second set of phased haplotype data into a second sequence of haplotype windows corresponding to the first sequence of haplotype windows, each haplotype window comprising a plurality of genetic markers;

determining, by the one or more processors, an initial ancestry classification for each haplotype window based on the plurality of genetic markers in the haplotype window, each haplotype window being classified as associated with a specific geographical region at an initial confidence level, the first and second sequences of haplotype windows having a plurality of different initial ancestry classifications; and

combining, by the one or more processors, a plurality of initial ancestry classifications of a plurality of haplotype windows having initial confidence levels below a required threshold to form an updated ancestry classification for a portion of the individual's genome, wherein:

the updated ancestry classification is for a broader geographical region comprising geographical regions of the plurality of initial ancestry classifications,

the updated ancestry classification has a total confidence level generated from the initial confidence levels of the plurality of initial ancestry classifications, and

the portion of the individual's genome includes the plurality of haplotype windows.

14 . The method of claim 13 , further comprising:

generating the initial ancestry classifications by recalibrating probabilities that portions of the individual's genome associate with specific ancestries.

15 . The method of claim 14 , wherein the recalibration is determined based on reference data of actual unadmixed individuals and/or simulated admixed individuals.

16 . The method of claim 13 , wherein the initial ancestry classifications are generated by a learning machine trained on a training set comprising genetic information of a plurality of individuals with known ancestries.

17 . The method of claim 16 , further comprising training the learning machine using the training set.

18 . The method of claim 16 , wherein the learning machine includes a neural network or support vector machine (SVM).

19 . The method of claim 13 , wherein the total confidence level is generated by summing the initial confidence levels of the plurality of initial ancestry classifications.

20 . A non-transitory machine-readable medium having stored thereon program code that, when executed by one or more processors of a computer system, cause the computer system to perform operations for classifying ancestry of genomic data, said program code comprising code for:

obtaining a first set of phased haplotype data of an individual attributed to a first parent and a second set of phased haplotype data of the individual attributed to a second parent;

dividing the first set of phased haplotype data into a first sequence of haplotype windows and the second set of phased haplotype data into a second sequence of haplotype windows corresponding to the first sequence of haplotype windows, each haplotype window comprising a plurality of genetic markers;

determining an initial ancestry classification for each haplotype window based on the plurality of genetic markers in the haplotype window, each haplotype window being classified as associated with a specific geographical region at an initial confidence level, the first and second sequences of haplotype windows having a plurality of different initial ancestry classifications;

combine a plurality of initial ancestry classifications of a plurality of haplotype windows having initial confidence levels below a required threshold to form an updated ancestry classification for a portion of the individual's genome, wherein:

the updated ancestry classification is for a broader geographical region comprising geographical regions of the plurality of initial ancestry classifications,

the updated ancestry classification has a total confidence level generated from the initial confidence levels of the plurality of initial ancestry classifications, and

the portion of the individual's genome includes the plurality of haplotype windows.

21 . The system of claim 1 , wherein the first and second sets of phased genotype data of the individual comprise hundreds of thousands of genetic markers, and wherein each haplotype window comprises 100 genetic markers.

22 . The system of claim 1 , wherein the one or more memories are further configured to provide the one or more processors with instructions which when executed cause the one or more processors to: repeat the combining using the updated ancestry classification as a classification of the plurality of initial ancestry classifications until the total confidence level meets the required threshold.

23 . The method of claim 13 , further comprising: repeating the combining using the updated ancestry classification as a classification of the plurality of initial ancestry classifications until the total confidence level meets the required threshold.

24 . The non-transitory machine-readable medium of claim 20 , said program code further comprising code for: repeating the combining using the updated ancestry classification as a classification of the plurality of initial ancestry classifications until the total confidence level meets the required threshold.

Assignments (5)

Continuity (4)

References Cited (400)