TrustGWAS: A full-process workflow for encrypted GWAS using multi-key homomorphic encryption and pseudorandom number...

Area of Science:

• Genetics
• Bioinformatics
• Computational Biology

Background:

• Genome-wide association studies (GWAS) are crucial for identifying genetic variants associated with diseases.
• Effective sample size is key to GWAS statistical power, but privacy concerns hinder data sharing.
• Existing cryptographic solutions for GWAS are incomplete, particularly for principal component analysis (PCA).

Purpose of the Study:

• To develop a comprehensive, privacy-preserving cryptographic solution for large-scale genome-wide association studies (GWAS).
• To enable secure cross-institutional collaboration for genetic data analysis.
• To provide a tool that supports essential GWAS analytical steps without compromising data security.

Main Methods:

• TrustGWAS utilizes pseudorandom number perturbations for secure PCA.
• A multi-key homomorphic encryption scheme is employed for other GWAS modules.
• The platform supports quality control, linkage disequilibrium pruning, PCA, and various regression models.

Main Results:

• TrustGWAS successfully replicates gold-standard GWAS results comparable to PLINK.
• The system can analyze 100,000 individuals and 1 million variants within 50 hours.
• Gene loci associated with fasting blood glucose were identified, aligning with previous findings.

Conclusions:

• TrustGWAS offers a complete, secure solution for large-scale GWAS, overcoming privacy challenges.
• The platform facilitates collaborative genetic research by ensuring data confidentiality.
• This work advances secure methods for genetic discovery and analysis.