Statistical Kernel Machine Regression (KMR) has become popular recently as a practical and simple tool for dimension reduction as well as modeling non-linear effects of covariates. In this talk, I present an overview of KMR and its usage in gene and environmental studies, and genome-wide association studies. Specifically, I discuss about modeling the joint effect of an entire gene-set on an outcome of interest, accounting for gene-gene and gene-environment interactions in various settings based on real data applications. I also present some potential projects for interested students.
Return to Biostatistics Seminars