This NSF-funded project develops a comprehensive privacy-preserving platform for genome imputation and analytics, enabling researchers to access high-quality reference genomes without exposing sensitive user data. The system combines trusted execution environments (TEEs) with data-oblivious computation, security monitoring, and related privacy technologies such as differential privacy to protect both the genomic data submitted by users and the controlled-access reference panels held by service providers. These research activities will produce deployable, confidential genome analytics services and foundational techniques for future privacy-preserving systems built on TEEs. 

The project is organized around three research thrusts (RTs):

• RT-1: Scalable TEE-based algorithms for genome imputation, phasing, and other emerging tasks with side-channel protection

• RT-2: Data transformation and monitoring tools to mitigate leakage of controlled-access reference datasets

• RT-3: Federated framework for securely leveraging reference datasets across multiple repositories

The broad societal impact of this project includes strengthening human subject protection in biomedical studies, enabling secure cross-border research collaboration, and making advanced genomic services more broadly accessible to the community.