Transfer Learning in Genome-Wide Association Studies with Knockoffs

Abstract This paper presents and compares alternative transfer learning methods that can increase the power of conditional testing via knockoffs by leveraging prior information in external data sets collected from different populations or measuring related outcomes. The relevance this methodology is explored particular within context genome-wide association studies, where it be helpful to address pressing need for principled ways suitably account for, efficiently learn genetic variation associated diverse ancestries. Finally, we apply these analyze several phenotypes UK Biobank set, demonstrating helps discover more associations minority populations, potentially opening way development accurate polygenic risk scores.