References of "Pascoal, Túlio 50032285"
     in
Bookmark and Share    
Full Text
See detailDyPS: Dynamic, Private and Secure GWAS (Summary) - GenoPri'21 Talk
Pascoal, Túlio UL; Decouchant, Jérémie; Boutet, Antoine et al

Presentation (2021, September 22)

Genome-Wide Association Studies (GWAS) identify the genomic variations that are statistically associated with a particular phenotype (e.g., a disease). GWAS results, i.e., statistics, benefit research and ... [more ▼]

Genome-Wide Association Studies (GWAS) identify the genomic variations that are statistically associated with a particular phenotype (e.g., a disease). GWAS results, i.e., statistics, benefit research and personalized medicine. The confidence in GWAS increases with the number of genomesanalyzed, which encourages federated computations where biocenters periodically include newly sequenced genomes. However, for legal and economical reasons, this collaboration can only happen if a release of GWAS results never jeopardizes the genomic privacy of data donors, if biocenters retain ownership and cannot learn each others’ data. Furthermore, given the reduced cost of sequencing DNA nowadays, there is now a need to update GWAS results in a dynamic manner, while enabling donors to withdraw consent at any time. Therefore, two challenges need to be simultaneously addressed to enable federated and dynamic GWAS: (i) the computation of GWAS statistics must rely on secure and privacy-preserving methods; and (ii) GWAS results that are publicly released should not allow any form of privacy attack. In this talk, we will introduce the problem we consider in more detail and present DyPS, the framework we have designed and recently presented at the Privacy Enhancing Technologies Symposium (PETS). We refer the reader to the full paper1 for the details we cannot cover in this short version. [less ▲]

Detailed reference viewed: 98 (16 UL)
Full Text
Peer Reviewed
See detailDyPS: Dynamic, Private and Secure GWAS
Pascoal, Túlio UL; Decouchant, Jérémie UL; Boutet, Antoine et al

in Proceedings on Privacy Enhancing Technologies (2021)

Genome-Wide Association Studies (GWAS) identify the genomic variations that are statistically associated with a particular phenotype (e.g., a disease). The confidence in GWAS results increases with the ... [more ▼]

Genome-Wide Association Studies (GWAS) identify the genomic variations that are statistically associated with a particular phenotype (e.g., a disease). The confidence in GWAS results increases with the number of genomes analyzed, which encourages federated computations where biocenters would periodically share the genomes they have sequenced. However, for economical and legal reasons, this collaboration will only happen if biocenters cannot learn each others’ data. In addition, GWAS releases should not jeopardize the privacy of the individuals whose genomes are used. We introduce DyPS, a novel framework to conduct dynamic privacy-preserving federated GWAS. DyPS leverages a Trusted Execution Environment to secure dynamic GWAS computations. Moreover, DyPS uses a scaling mechanism to speed up the releases of GWAS results according to the evolving number of genomes used in the study, even if individuals retract their participation consent. Lastly, DyPS also tolerates up to all-but-one colluding biocenters without privacy leaks. We implemented and extensively evaluated DyPS through several scenarios involving more than 6 million simulated genomes and up to 35,000 real genomes. Our evaluation shows that DyPS updates test statistics with a reasonable additional request processing delay (11% longer) compared to an approach that would update them with minimal delay but would lead to 8% of the genomes not being protected. In addition, DyPS can result in the same amount of aggregate statistics as a static release (i.e., at the end of the study), but can produce up to 2.6 times more statistics information during earlier dynamic releases. Besides, we show that DyPS can support a larger number of genomes and SNP positions without any significant performance penalty. [less ▲]

Detailed reference viewed: 297 (51 UL)