Genetic studies in multi-ethnic cohorts offer great potential to elucidate the genetic factors influencing complex traits. A variety of statistical methods have recently been developed to overcome special challenges for whole genome association analysis of complex traits in large-scale multi-ethnic cohorts. With existing methodology, however, genetic ancestry differences among sampled individuals are often treated as a confounder to be adjusted for in an analysis to protect against spurious association.
In this talk, I will discuss leveraging genetic ancestry for improved complex trait mapping in multi-ethnic populations, such as African Americans and Hispanic/Latino populations, who have admixed ancestry derived from multiple continents. Mixed effects models for admixture mapping that incorporate both local and global ancestry for the identification of genetic loci influencing complex traits will be presented. The proposed mixed model admixture mapping methods have been developed for continuous and dichotomous traits, and are completely applicable to large-scale whole genome studies with multi-ethnic samples from a variety of study designs.
We demonstrate the utility of leveraging genetic ancestry for complex trait mapping in applications to the Hispanic Community Health Study / Study of Latinos (HCHS/SOL) and the Alzheimer's Disease Sequencing Project (ADSP), where our mixed model admixture mapping approach identifies novel loci that are not identified via genetic association mapping.