Secure top most significant genome variants search: iDASH 2017 competition

Abstract

BackgroundOne of the 3 tracks of iDASH Privacy & Security Workshop 2017 competition was to execute a whole genome variants search on private genomic data. Particularly, the search application was to find the top most significant SNPs (Single-Nucleotide Polymorphisms) in a database of genome records labeled with control or case. In this paper we discuss the solution submitted by our team to this competition.MethodsPrivacy and confidentiality of genome data had to be ensured using Intel SGX enclaves. The typical use-case of this application is the multi-party computation (each party possessing one or several genome records) of the SNPs which statistically differentiate control and case genome datasets.ResultsOur solution consists of two applications: (i) compress and encrypt genome files and (ii) perform genome processing (top most important SNPs search). We have opted for a horizontal treatment of genome records and heavily used parallel processing. Rust programming language was employed to develop both applications.ConclusionsExecution performance of the processing applications scales well and very good performance metrics are obtained. Contest organizers selected it as the best submission amongst other received competition entries and our team was awarded the first prize on this track.

The evolving privacy and security concerns for genomic data analysis and sharing as observed from the iDASH competition

“…To benchmark and evaluate these important topics, we organized, with the participation of community members from all over the world, eight annual iDASH competitions (2014–2021), aimed at tackling state-of-the-art privacy and security challenges. Each competition contained two to four different tracks (as shown in Supplementary Table ST1 ), and the iDASH consortium generated 40 publications 13–17 , 21–26 , 30 , 40–67 from 23 tracks. These papers have been cited 1491 times (max = 137, min = 3, median = 29.5, average = 37.3) as of April 2022, 68 demonstrating the impact of the competition on the field (a diagram of the total citations, as well as the citations per year published, is shown in Supplementary Figure SF1 ).…”

I-GWAS: Privacy-Preserving Interdependent Genome-Wide Association Studies

“…A federation of biocenters aiming to conduct a GWAS could rely on several privacy-preserving schemes, such as Secure Multiparty Computation (SMC) [17,19,38,75], Homomorphic Encryption (HE) [8,83,85], local DP-based approaches [20,58], and Trusted Execution Environment (TEE)-based solutions [9,12,15,52,64,68]. The works mentioned above ensure privacy while computing the results of a GWAS.…”

“…However, they are insufficient to prevent privacy leaks once the GWAS result is made accessible to an adversary. Inspired by existing TEE-based deployments for federated GWAS [9,12,15,52,68], we also opted for TEE-based implementation to compute GWAS statistics and to assess whether results can be safely released [64], but also to demonstrate that our algorithm can run in a secure environment, even if this is hosted by an untrustworthy third party. We have chosen Intel SGX [60] as a vehicle for our implementation without relying on any specific feature of SGX.…”

HySec-Flow: Privacy-Preserving Genomic Computing with SGX-based Big-Data Analytics Framework

“…Current benchmark experiments on data-intensive computing tasks [5] demonstrate that SGX provides data protection against attacks from the host operating system or even system administrators while introducing only moderate computation overhead; therefore, it is widely considered to be suitable for data-intensive computation, including the computing tasks involving personal human genomic data. Privacy-preserving algorithms have been developed for several genomic analysis tasks, including genetic testing and variant searching using human genomic profiles [6]- [8]. These tasks are relatively lightweight and do not require extensive memory that exceeds the limited Page Cache (EPC) available in an enclave.…”