Ó³»­´«Ã½

Large-scale biobanks such as UK Biobank and the US All of Us program now store genomic information from millions of individuals, with each fully sequenced genome requiring around 300 GB of storage. To handle both the scale and the sensitivity of this data, these biobanks rely on public cloud sandbox models to enable large-scale research. However, this dependence on public cloud infrastructure introduces new privacy concerns. Homomorphic encryption offers a promising solution by allowing computations to be performed directly on encrypted data without ever exposing the raw information. In this talk, I will introduce the fundamentals of homomorphic encryption, focusing on the Brakerski Gentry Vaikuntanathan (BGV) scheme for integer-based homomorphic computation. I will explain how it works, showcase existing implementations in genotype-phenotype databases, and briefly discuss current limitations in the context of genomics research.