Population genetics is a crucial tool for facilitating medical genetics and disease gene mapping studies. This proposal outlines two novel population genetic techniques that each facilitate disease gene mapping in different ways. First, it describes a probabilistic technique for detecting the proportion of homozygosity an individual is likely to have based on SNP array data. This is useful for prioritizing disease case sample individuals for whole genome sequencing and subsequent homozygosity mapping to identify recessive disease genes. Such a tool is useful for large outbred populations such as European Americans for which homozygous regions are likely to be short and therefore cannot be unambiguously detected using existing techniques for SNP array data. I will apply this homozygosity prioritization scheme to a dataset of individuals with autism to identify individuals that will be most informative to resequence and will perform homozygosity mapping on the sequence data to locate recessive disease genes.
The second aim i s to develop methods to permit combined admixture mapping and genome-wide association (GWA) of Latinos, a population for which current admixture mapping methods fail. Latino admixture mapping is challenging because suitable reference haplotypes for their Native American ancestry are lacking, and because existing techniques cannot model their complex three-way admixture. I will explore three methods for performing admixture mapping of Latinos, including using linear combinations of collections of Native American reference haplotypes and utilizing the information about Native American variation present in the Latinos themselves. To show that my approach works in practice, I will apply this tool to study the genetics of type 2 diabetes, a disease with higher prevalence among Latinos.
Insights from population genetics and statistical methods grounded in medical genetics have been very important in facilitating medical genetics and disease gene mapping studies. This pro- posal describes novel population genetic techniques with direct application to disease gene mapping, both in populations that have experienced founder events and in U.S. Latino populations that are admixed. The technology that I develop will be widely applicable to many disease gene mapping studies, but I will focus as proof-of-principle on two diseases: autism and type 2 diabetes.
|SIGMA Type 2 Diabetes Consortium; Williams, Amy L; Jacobs, Suzanne B R et al. (2014) Sequence variants in SLC16A11 are a common risk factor for type 2 diabetes in Mexico. Nature 506:97-101|