The broader application of polygenic risk score (PRS) is hindered by the limited transferability of PRS developed in Europeans to non-European populations. This disparity is a significant barrier for the equitable use of genomic information in healthcare. While many statistical methods have been developed to improve the performance of PRS in non-European populations, most of them focus on discrete genetic ancestry clusters and fail to account for the complexities of admixed individuals, including continuum variation of genetic ancestries and mosaic ancestral origins of alleles. For example, African Americans have mosaic genomic segments originated from European and African ancestries, and Hispanic/Latinos have mosaic genomic segments originated from European, African and Native American ancestries.
In this project, we will address this challenge by developing and applying a set of statistical methods using data from UK biobank. Specifically, we will 1) explore the global and local genetic structure of admixed populations in UK biobank; 2) identify genetic variants associated with complex traits and their interactions with environmental factors in admixed populations, with a focus on mental health outcomes with lifestyle and social-economic factors; 3) develop novel statistical and computational methods to calculate PRS in admixed populations by integrating continental GWAS summary statistics with ancestry-informed genotypes of admixed populations.