System and method for privacy-preserving genomic data analysis
A method includes accessing genomic data of from a genomic database; generating, by a processor, a first hash by probabilistically and irreversibly encrypting a first portion of the genomic data encoding the first genomic sequence, the first hash projecting the first portion into reduced dimensions such that the first portion of the genomic data encoding the first genomic sequence becomes statistically improbable to recover outside the first processor; generating, by the processor, a first cryptogram by deterministically and reversibly encrypting a second portion of the genomic data encoding the first genomic sequence; generating, by the processor, a look-up table by using at least the first cryptogram as a key and the first portion of the genomic data encoding the first genomic sequence as the value, and transmitting data encoding the first hash and the first cryptogram to one or more processors that are different from the first processor.
1. A computer-implemented method, the method comprising:
accessing, by one or more processors in a data vault, genomic data of patients from a genomic database, the genomic data encoding strings of nucleotide values from genomic sequences of the patients as well as information of respective chromosomes on which corresponding genomic sequences reside;
generating, by the one or more processors in the data vault, a first hash by probabilistically and irreversibly encrypting a first portion of the genomic data encoding a first genomic sequence of each patient, the first hash projecting the first portion into reduced dimensions such that the first portion of the genomic data encoding the first genomic sequence of each patient becomes statistically improbable to recover outside the one or more processors in the data vault;
generating, by the one or more processors in the data vault, a first cryptogram by deterministically and reversibly encrypting a second portion of the genomic data encoding the first genomic sequence of each patient;
generating, by the one or more processors in the data vault, a look-up table by using at least the first cryptogram as a key and the first portion of the genomic data encoding the first genomic sequence of each patient as the value;
transmitting, by the one or more processors in the data vault, encrypted genomic data that includes the first hash and the first cryptogram to one or more processors outside the data vault without patient identifiable information;
receiving, from the one or more processors outside the data vault, processed results that include the encrypted genomic data of each patient; and
decoding, by the one or more processors in the data vault, the encrypted genomic data based on, at least in part, the look-up table to analyze the processed results according to the first portion of the genomic data encoding the first genomic sequence of each patient such that aggregate analysis of the patients are conducted without revealing patient identifiable information of each patient to the one or more processors outside the data vault.
2. The method of claim 1 , wherein probabilistically and irreversibly encrypting the first portion comprises probabilistically and irreversibly encrypting locus information that includes an identification (ID) of a particular chromosome on which the first genomic sequence resides as well a starting position of the first genomic sequence on the particular chromosome.
3. The method of claim 2 , wherein the information encoding the string of nucleotides values of the particular chromosome includes a reference genome sequence and an alteration sequence.
4. The method of claim 1 , wherein deterministically and reversibly encrypting the second portion comprises deterministically and reversibly encrypting strings of nucleotide values from the first genomic sequence.
5. The method of claim 1 , further comprising:
generating, by the one or more processors in the data vault, a second hash by deterministically encrypting the first portion of the genomic data.
6. The method of claim 5 , wherein generating the look-up table comprises:
using the second hash as well as the first cryptogram as a key and the first portion of the genomic data encoding the first genomic sequence as a value such that the first portion of the genomic data encoding the first genomic sequence is retrievable at the one or more processors in the data vault based on the combination of the second hash and the first cryptogram.
7. The method of claim 5 , wherein transmitting the information comprises transmitting the information encoding the first hash, the second hash, and the first cryptogram to one or more processors that are different from the one or more processors in the data vault and without patient identifiable information.
8. The method of claim 5 , further comprising:
generating, by the one or more processors in the data vault, a third hash by probabilistically and irreversibly encrypting the first portion of the genomic data encoding a second genomic sequence, the third hash projecting the first portion into reduced dimensions such that the first portion of the genomic data encoding the second genomic sequence becomes statistically improbable to recover outside the one or more processors in the data vault.
9. The method of claim 8 , further comprising:
generating, by the one or more processors in the data vault, a second cryptogram by deterministically and reversibly encrypting a second portion of the genomic data encoding the second genomic sequence.
10. The method of claim 9 , further comprising:
adding, by the one or more processors in the data vault, a new key-value entry to the look-up table using at least the third hash as a key in the new key-value entry and the second cryptogram as a value for the new key-value entry.
11. The method of claim 8 , wherein the first hash and the third hash both correspond to chromosomal information from the same patient.
12. The method of claim 8 , wherein the first hash and the second hash correspond to chromosomal information from different patients.
13. The method of claim 1 , wherein probabilistically and irreversibly encrypting comprises performing a locality-sensitive hashing.
14. The method of claim 1 , wherein deterministically and reversibly encrypting the string of nucleotide values comprises using a reversible hash to encrypt the string of nucleotide values.
15. The method of claim 1 , wherein deterministically and reversibly encrypting the string of nucleotide values comprises encrypting using a key.
16. A computer system comprising at least one non-transitory storage device and at least one processor coupled thereto, the computer system configured to perform operations of:
accessing, by the at least one processor of the computer system, genomic data of patients from a genomic database, the genomic data encoding strings of nucleotide values from genomic sequences of the patients as well as information of respective chromosomes on which corresponding genomic sequences reside;
generating a first hash by probabilistically and irreversibly encrypting a first portion of the genomic data encoding a first genomic sequence of each patient, the first hash projecting the first portion into reduced dimensions such that the first portion of the genomic data encoding the first genomic sequence of each patient becomes statistically improbable to recover outside the at least one processor of the computer system;
generating a first cryptogram by deterministically and reversibly encrypting a second portion of the genomic data encoding the first genomic sequence of each patient;
generating a look-up table by using at least the first cryptogram as a key and the first portion of the genomic data encoding the first genomic sequence of each patient as the value;
transmitting encrypted genomic data that includes the first hash and the first cryptogram to one or more processors that are different from the at least one processor of the computer system without patient identifiable information;
receiving, from the one or more processors that are different from the computer system, processed results that include the encrypted genomic data of each patient; and
decoding the encrypted genomic data based on, at least in part, the look-up table to analyze the processed results according to the first portion of the genomic data encoding the first genomic sequence of each patient such that aggregate analysis of the patients are conducted without revealing patient identifiable information of each patient to the one or more processors that are different from the at least one processor of the computer system.
17. The computer system of claim 16 , wherein the operations further comprise:
generating a second hash by deterministically encrypting the first portion of the genomic data.
18. The computer system of claim 17 , wherein generating the look-up table comprise:
using the second hash as well as the first cryptogram as a key and the first portion of the genomic data encoding the first genomic sequence as a value such that the first portion of the genomic data encoding the first genomic sequence is retrievable at the computer system based on the combination of the second hash and the first cryptogram.
19. The computer system of claim 17 , wherein transmitting the information comprises transmitting the information encoding the first hash, the second hash, and the first cryptogram to one or more processors that are different from the computer system and without patient identifiable information.
20. The computer system of claim 16 , wherein the operations further comprise:
generating a third hash by probabilistically and irreversibly encrypting the first portion of a second genomic sequence, the third hash projecting the first portion into reduced dimensions such that the first portion of the genomic data encoding the second genomic sequence becomes statistically improbable to recover outside the at least one processor of the computer system.
21. The computer system of claim 20 , wherein the operations further comprise:
generating a second cryptogram by deterministically and reversibly encrypting a second portion of the genomic data encoding the second genomic sequence.
22. The computer system of claim 21 , wherein the operations further comprise:
adding, a new key-value entry to the look-up table using at least the third hash as a key in the new key-value entry and the second cryptogram as a value for the new key-value entry.
23. The computer system of claim 16 , wherein probabilistically and irreversibly encrypting comprises performing a locality-sensitive hashing.
24. The computer system of claim 16 , wherein deterministically and reversibly encrypting the string of nucleotide values comprises using a reversible hash to encrypt the string of nucleotide values.
25. The computer system of claim 16 , wherein deterministically and reversibly encrypting the string of nucleotide values comprises encrypting using a key.
26. A non-transitory computer-readable medium comprising software instructions that, when executed by a computer, causes the computer to perform operations of:
accessing, by one or more processors of the computer, genomic data of patients from a genomic database, the genomic data encoding strings of nucleotide values from genomic sequences of the patients as well as information of respective chromosomes on which corresponding genomic sequences reside;
generating a first hash by probabilistically and irreversibly encrypting a first portion of the genomic data encoding a first genomic sequence of each patient, the first hash projecting the first portion into reduced dimensions such that the first portion of the genomic data encoding the first genomic sequence of each patient becomes statistically improbable to recover outside the one or more processors of the computer;
generating a first cryptogram by deterministically and reversibly encrypting a second portion of the genomic data encoding the first genomic sequence of each patient;
generating a look-up table by using at least the first cryptogram as a key and the first portion of the genomic data encoding the first genomic sequence of each patient as the value;
transmitting encrypted genomic data that includes the first hash and the first cryptogram to one or more processors that are different from the one of more processors of the computer without patient identifiable information;
receiving, from the one or more processors that are different from the one or more processors of the computer, processed results that include the encrypted genomic data of each patient; and
decoding the encrypted genomic data based on, at least in part, the look-up table to analyze the processed results according to the first portion of the genomic data encoding the first genomic sequence of each patient such that aggregate analysis of the patients are conducted without revealing patient identifiable information of each patient to the one or more processors that are different from the one or more processors of the computer.