System and method for genomic data analysis
A method includes accessing genomic data from a first cohort and a second cohort of patients that are encrypted to comprise a probabilistic and irreversible hash of each patient's genomic sequence data; based on the probabilistic and irreversible hashes, determining one or more variants residing in a particular locale indicating where the one or more variants reside; comparing a first number of variants determined to reside in the particular locale for the first cohort of patients with a second number of variants determined to reside in the particular locale for the second cohort of patients; and in response to determining that the first number of variants determined to reside in the particular locale for the first cohort of patients and the second number of variants determined to reside in the particular locale for the second cohort of patients differ by more than a threshold value, identifying the particular locale.
1. A computer-implemented method, the method comprising:
accessing, by one or more processors, encrypted genomic data from a first cohort and a second cohort of patients, wherein the genomic data for each patient includes chromosome location information that is structured with a probabilistic and irreversible hash and genomic sequence information that is structured with a deterministic and reversible hash, and wherein a phenotypic trait is present in the second cohort but absent from the first cohort;
generating location bins based on the probabilistic and irreversible hash that projects the chromosome location information from the first and second cohorts of patients into reduced dimensions, wherein genomic variants located near one another belong to a particular location bin;
identifying, based on the location bins and a comparison of the genomic data and a reference genome, one or more genomic variants in the particular location bin, wherein the one or more genomic variants represent variations from the reference genome;
comparing a first number of genomic variants in the particular location bin for the first cohort of patients with a second number of genomic variants in the particular location bin for the second cohort of patients, wherein a difference between the first number and the second number is associated with the phenotypic trait;
determining that the first number of genomic variants and the second number of genomic variants differ by more than a threshold value in the particular location bin; and
providing, for downstream genomic analysis, information indicative of the genomic variants in the particular location bin.
2. The computer-implemented method of claim 1 , wherein identifying the one or more genomic variants in the particular location bin comprises:
determining that the one or more genomic variants reside on a particular chromosome and at a particular starting position on the particular chromosome.
3. The computer-implemented method of claim 1 , further comprising:
comparing each of the first number of genomic variants with each of the second number of genomic variants to identify one or more specific genomic variants that are uniquely present in one but not both of the first cohort and the second cohort of patients.
4. The computer-implemented method of claim 3 , further comprising:
statistically associating, from the one or more specific genomic variants, a subset of genomic variants with the phenotypic trait.
5. The computer-implemented method of claim 4 , wherein statistically associating the subset of genomic variants with the phenotypic trait further comprises:
tabulating, for each particular location bin, the difference between the first number of genomic variants and the second number of genomic variants;
ranking, based on the tabulated differences, the location bins; and
selecting the location bins with more than a median of the tabulated differences.
6. The computer-implemented method of claim 4 , wherein the statistically associating is not performed for genomic variants across substantially all positions on a particular chromosome.
7. The computer-implemented method of claim 1 , wherein accessing the encrypted genomic data comprises:
accessing the probabilistic and irreversible hash of the chromosome location information of the patients.
8. The computer-implemented method of claim 1 , wherein the probabilistic and irreversible hash prohibits rehashing the chromosome location information.
9. The computer-implemented method of claim 1 , wherein the genomic sequence information from the first cohort and the second cohort of patients are encrypted to conceal identities of the patients.
10. The computer-implemented method of claim 1 , wherein accessing the encrypted genomic data comprises:
accessing the deterministic and reversible hash of the genomic sequence information of the patients.
11. A computer system comprising one or more processors configured to perform operations of:
accessing, by the one or more processors, encrypted genomic data from a first cohort and a second cohort of patients, wherein the genomic data for each patient includes chromosome location information that is structured with a probabilistic and irreversible hash and genomic sequence information that is structured with a deterministic and reversible hash, and wherein a phenotypic trait is present in the second cohort but absent from the first cohort;
generating location bins based on the probabilistic and irreversible hash that projects the chromosome location information from the first and second cohorts of patients into reduced dimensions, wherein genomic variants located near one another belong to a particular location bin;
identifying, based on the location bins and a comparison of the genomic data and a reference genome, one or more genomic variants in the particular location bin, wherein the one or more genomic variants represent variations from the reference genome;
comparing a first number of genomic variants in the particular location bin for the first cohort of patients with a second number of genomic variants in the particular location bin for the second cohort of patients, wherein a difference between the first number and the second number is associated with the phenotypic trait;
determining that the first number of genomic variants and the second number of genomic variants differ by more than a threshold value in the particular location bin; and
providing, for downstream genomic analysis, information indicative of the genomic variants in the particular location bin.
12. The computer system of claim 11 , wherein identifying the one or more genomic variants in the particular location bin comprises:
determining that the one or more genomic variants reside on a particular chromosome and at a particular starting position on the particular chromosome.
13. The computer system of claim 11 , wherein the operations further comprise:
comparing each of the first number of genomic variants with each of the second number of genomic variants to identify one or more specific genomic variants that are uniquely present in one but not both of the first cohort and the second cohort of patients.
14. The computer system of claim 13 , further comprising:
statistically associating, from the one or more specific genomic variants, a subset of genomic variants with the phenotypic trait.
15. The computer system of claim 14 , wherein statistically associating the subset of genomic variants with the phenotypic trait further comprises:
tabulating, for each particular location bin, the difference between the first number of genomic variants and the second number of genomic variants;
ranking, based on the tabulated differences, the location bins; and
selecting the location bins with more than a median of the tabulated differences.
16. The computer system of claim 14 , wherein the statistically associating is not performed for genomic variants across substantially all positions on a particular chromosome.
17. The computer system of claim 11 , wherein accessing the encrypted genomic data comprises:
accessing the probabilistic and irreversible hash of the chromosome location information of the patients.
18. The computer system of claim 11 , wherein the probabilistic and irreversible hash prohibits rehashing the chromosome location information.
19. The computer system of claim 11 , wherein the genomic sequence information from the first cohort and the second cohort of patients are encrypted to conceal identities of the patients.
20. The computer system of claim 11 , wherein accessing the encrypted genomic data comprises:
accessing the deterministic and reversible hash of the genomic sequence information of the patients.
21. A non-transitory computer-readable storage device encoded with computer program instructions that, when executed by one or more computers, cause the one or more computers to perform operations comprising:
accessing, by the one or more computers, encrypted genomic data from a first cohort and a second cohort of patients, wherein the genomic data for each patient includes chromosome location information that is structured with a probabilistic and irreversible hash and genomic sequence information that is structured with a deterministic and reversible hash, and wherein a phenotypic trait is present in the second cohort but absent from the first cohort;
generating location bins, by the one or more computers, based on the probabilistic and irreversible hash that projects the chromosome location information from the first and second cohorts of patients into reduced dimensions, wherein genomic variants located near one another belong to a particular location bin;
identifying, by the one or more computers and based on the location bins and a comparison of the genomic data and a reference genome, one or more genomic variants in the particular location bin, wherein the one or more genomic variants represent variations from the reference genome;
comparing, by the one or more computers, a first number of genomic variants in the particular location bin for the first cohort of patients with a second number of genomic variants in the particular location bin for the second cohort of patients, wherein a difference between the first number and the second number is associated with the phenotypic trait;
determining, by the one or more computers, that the first number of genomic variants and the second number of genomic variants differ by more than a threshold value in the particular location bin; and
providing, by the one or more computers and for downstream genomic analysis, information indicative of the genomic variants in the particular location bin.
22. The non-transitory computer-readable storage device of claim 21 , wherein the operations further comprise:
comparing each of the first number of genomic variants with each of the second number of genomic variants to identify one or more specific genomic variants that are uniquely present in one but not both of the first cohort and the second cohort of patients.
23. The non-transitory computer-readable storage device of claim 22 , further comprising:
statistically associating, from the one or more specific genomic variants, a subset of genomic variants with the phenotypic trait.
24. The non-transitory computer-readable storage device of claim 23 , wherein statistically associating the subset of genomic variants with the phenotypic trait further comprises:
tabulating, for each particular location bin, the difference between the first number of genomic variants and the second number of genomic variants;
ranking, based on the tabulated differences, the location bins; and
selecting the location bins with more than a median of the tabulated differences.