Privacy-preserving genotype imputation with fully homomorphic encryption

Gürsoy, Gamze; Chielle, Eduardo Ottobelli; Brannon, Charlotte M.; Maniatakos, Michail; Gerstein, Mark

doi:10.1016/j.cels.2021.10.003

Cited by 15 publications

(4 citation statements)

References 41 publications

(43 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Among those works that applied FL, MPC, MHE, and DP to omic data analysis, most of them used either the cryptographic techniques (41)(42)(43) or the DP notion (31,41,(44)(45)(46)(47)(48) to provide formal privacy guarantees for the participants in the research of singlenucleotide polymorphisms (SNPs), genome-wide association study (GWAS), and differential gene expression analysis (49), which are relatively narrow and specific problems in genomics studies, and whose data are obtained by postprocessing the raw sequencing data (section S1.3). In addition, the methods in those articles could only be shown to be applicable to statistical solutions or traditional ML solutions in GWAS.…”

Section: Introductionmentioning

confidence: 99%

PPML-Omics: A privacy-preserving federated machine learning method protects patients’ privacy in omic data

Zhou,

Chen,

et al. 2024

Sci. Adv.

View full text Add to dashboard Cite

Modern machine learning models toward various tasks with omic data analysis give rise to threats of privacy leakage of patients involved in those datasets. Here, we proposed a secure and privacy-preserving machine learning method (PPML-Omics) by designing a decentralized differential private federated learning algorithm. We applied PPML-Omics to analyze data from three sequencing technologies and addressed the privacy concern in three major tasks of omic data under three representative deep learning models. We examined privacy breaches in depth through privacy attack experiments and demonstrated that PPML-Omics could protect patients’ privacy. In each of these applications, PPML-Omics was able to outperform methods of comparison under the same level of privacy guarantee, demonstrating the versatility of the method in simultaneously balancing the privacy-preserving capability and utility in omic data analysis. Furthermore, we gave the theoretical proof of the privacy-preserving capability of PPML-Omics, suggesting the first mathematically guaranteed method with robust and generalizable empirical performance in protecting patients’ privacy in omic data.

show abstract

Section: Introductionmentioning

confidence: 99%

PPML-Omics: A privacy-preserving federated machine learning method protects patients’ privacy in omic data

Zhou,

Chen,

et al. 2024

Sci. Adv.

View full text Add to dashboard Cite

show abstract

“…For example, it has been shown that privacy-enhancing genome-wide association studies (GWAS) can be possible [62,70,13,44]. It has also been shown that secure genotype imputation is feasible and scalable using homomorphic encryption [71,20,36]. Homomorphic encryption was also used for genomic variant querying [19], regression analysis for rare disease variants [68], and inference using genetic variants in machine learning applications [60].…”

Section: Introductionmentioning

confidence: 99%

Ultra-secure storage and analysis of genetic data for the advancement of precision medicine

Blindenbach,

Kang,

Hong

et al. 2024

Preprint

Self Cite

View full text Add to dashboard Cite

Cloud computing provides the opportunity to store the ever-growing genotype-phenotype data sets needed to achieve the full potential of precision medicine. However, due to the sensitive nature of this data and the patchwork of data privacy laws across states and countries, additional security protections are proving necessary to ensure data privacy and security. Here we present SQUiD, asecurequeryabledatabase for storing and analyzing genotype-phenotype data. With SQUiD, genotype-phenotype data can be stored in a low-security, low-cost public cloud in the encrypted form, which researchers can securely query without the public cloud ever being able to decrypt the data. We demonstrate the usability of SQUiD by replicating various commonly used calculations such as polygenic risk scores, cohort creation for GWAS, MAF filtering, and patient similarity analysis both on synthetic and UK Biobank data. Our work represents a new and scalable platform enabling the realization of precision medicine without security and privacy concerns.

show abstract

“…Among those works that applied FL, MPC, MHE and DP to omic data analysis, most of them utilized either the cryptographic techniques [42], [43], [44] or the DP notion [33], [42], [45], [46], [47], [48], [49] to provide formal privacy guarantees for the participants in the research of SNPs, GWAS and differential gene expression analysis [50], which are relatively narrow and specific problems in genomics studies, and whose data is obtained by post-processing the raw sequencing data ( Supplementary Section 1.3 ). In addition, the methods in those articles could only be shown to be applicable to statistical solutions or traditional machine learning solutions in GWAS.…”

Section: Introductionmentioning

confidence: 99%

PPML-Omics: a Privacy-Preserving federated Machine Learning method protects patients’ privacy in omic data

Zhou

Chen

et al. 2022

Preprint

View full text Add to dashboard Cite

Modern machine learning models towards various tasks with omic data analysis give rise to threats of privacy leakage of patients involved in those datasets. Despite the advances in different privacy technologies, existing methods tend to introduce too much noise, which hampers model accuracy and usefulness. Here, we built a secure and privacy-preserving machine learning (PPML) system by combining federated learning (FL), differential privacy (DP) and shuffling mechanism. We applied this system to analyze data from three sequencing technologies, and addressed the privacy concern in three major tasks of omic data, namely cancer classification with bulk RNA-seq, clustering with single-cell RNA-seq, and the integration of spatial gene expression and tumour morphology with spatial transcriptomics, under three representative deep learning models. We also examined privacy breaches in depth through privacy attack experiments and demonstrated that our PPML-Omics system could protect patients' privacy. In each of these applications, PPML-Omics was able to outperform state-of-the-art systems under the same level of privacy guarantee, demonstrating the versatility of the system in simultaneously balancing the privacy-preserving capability and utility in omic data analysis. Furthermore, we gave the theoretical proof of the privacy-preserving capability of PPML-Omics, suggesting the first mathematically guaranteed model with robust and generalizable empirical performance.

show abstract

Privacy-preserving genotype imputation with fully homomorphic encryption

Cited by 15 publications

References 41 publications

PPML-Omics: A privacy-preserving federated machine learning method protects patients’ privacy in omic data

PPML-Omics: A privacy-preserving federated machine learning method protects patients’ privacy in omic data

Ultra-secure storage and analysis of genetic data for the advancement of precision medicine

PPML-Omics: a Privacy-Preserving federated Machine Learning method protects patients’ privacy in omic data

Contact Info

Product

Resources

About