Math/Stats Colloquium: Kelsey Grinde, Macalester College
Title: Statistical methods for genetic studies in admixed populations
Abstract: Tools for analyzing genetic data and uncovering the genetic causes of human disease are at the heart of government-sponsored precision medicine initiatives, genetic testing routinely conducted at health clinics, and direct-to-consumer genetic and ancestry testing offered by companies like 23andMe and AncestryDNA. However, the vast majority of genetic studies to date have been conducted in populations of European ancetry. Not coincidentally, many of the statistical methods originally developed for genetics studies were designed for the context of studying populations with so-called "homogenous" ancestry. Populations with mixed, diverse ancestry (also known as admixed populations) have been particularly sorely underrepresented in genetic studies to date, and a number of open questions remain in terms of updating/developing statistical methods for genetic studies in these populations. In this talk, I will discuss work in developing statistical methods for two important tasks: inferring genetic ancestry and identifying genetic variants associated with complex traits and diseases.