A Unified Framework for Testing Genetic Associations Integrating Environmental Exposures

• Activity Number: 211
• Type: Invited
• Date/Time: Monday, August 5, 2013 : 2:00 PM to 3:50 PM
• Sponsor: International Indian Statistical Association
• Author(s): Summer S. Han*+ and Philip S. Rosenberg and Nilanjan Chatterjee
• Companies: Stanford University and National Cancer Institute and National Cancer Institute
• Keywords: genetic associations ; genome-wide association study (GWAS) ; score test ; interaction ; environmental exposure ; additive model

Abstract

There has been considerable success in genome-wide association studies (GWAS) for identifying susceptibility loci for various complex traits; however, a large proportion of the heritability of these traits still remained unexplained. A major limitation of current methods in case-control studies is the lack of full consideration of environmental exposures. It's well known that environmental exposures play important roles in contributing to disease risk in many complex traits, and often modify genetic effects. Another fundamental limitation of current approaches is their reliance on a specific risk model-multiplicative risk model via logistic regression-which is not based on any biological explanation or evidence from data, but on statistical convention. In this study, we propose a unified test for identifying genetic associations, which integrates the effects of environmental exposures to allow for heterogeneous genetic effects by exposures; the proposed method will incorporate a class of underlying disease risk models-including multiplicative, super-multiplicative, additive and probit risk models. We illustrate our method by applying it to NCI lung cancer GWAS data.