Exploring the genetic determinants of the portability of GWAS associations and polygenic risk score performance across populations (360G-Wellcome-222334_Z_21_Z)

Over the last 10-15 years, large datasets containing genetic, physical and medical information for thousands of people (such as the UK Biobank) have allowed researchers to detect many genetic mutations that increase disease risk or explain variation in other physical traits. An important limitation of most of the studies conducted so far is that they mainly use data from people of European ancestry. This is significant, because there is increasing evidence that genetic differences between populations limit the validity and applicability of findings from European samples in other populations. We propose to address one aspect of this problem by exploring genetic variation that affects the accuracy of a set of relevant mutations in predicting a physical trait or disease within a single population. This is an important first step in understanding the causes of the poor portability of findings across populations. We expect that our work will identify relevant new genetic mutations, or interactions between them, which impact particular traits and diseases (as well as corroborate previous findings) and that this information will then improve our ability to understand the causes of, and predict, disease risk in understudied populations.