Xiaoting Li

Genetic variation can impact gene expression either via cis-acting or via trans-acting mechanisms. Trans-acting loci affecting the regulatory activity of TFs have been referred to as activity quantitative trait loci (aQTLs). Mapping of trans-acting genetic variants to downstream genes is usually limited by statistical power. However, mapping the association between variants and the regulatory activity of TFs provides a complementary way to study how variants impact gene expression by modulating TF activity. The recent emergence of large collections of human functional genomics data has put human aQTL analysis within reach. Identifying aQTLs is important, both for understanding disease-causing variants and for gaining a better understanding of the cellular regulatory systems. To achieve this goal, this study has developed a generalized linear modeling (GLM) based method to systematically estimate activity levels of hundreds of human TFs in an individual-specific manner, and apply it to large human gene expression datasets. This will provide insight into how TFs regulate differential gene expression across tissues and across individuals. The inferred TF activity will be treated as a quantitative trait, and the genome-wide association mapping will be performed to identify genetic variants that are significantly associated with TF activity levels in each tissue. The identified aQTLs will provide insight into genetic determinants of TF regulatory network function.