Semiparametric Bayesian Analysis of Gene-Environment Interactions with Error in Measurement of Environmental Covariates and Missing Genetic Data

• Activity Number: 13
• Type: Topic Contributed
• Date/Time: Sunday, August 1, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Iryna Lobach*+ and Bani K. Mallick and Raymond Carroll
• Companies: New York University School of Medicine and Texas A&M University and Texas A&M University
• Keywords: Errors in variables ; Gene-environment interactions ; Markov Chain Monte Carlo sampling ; Missing Data1 ; Pseudo-likelihood

Abstract

Case-control studies are widely used to detect gene-environment interactions in the etiology of complex diseases. Many variables of interest to biomedical researchers are difficult to measure on the individual level. Measurement error causes bias in parameter estimates, thus masking key features of data. We develop a Bayesian methodology for the case of measurement error an environmental covariate. A unique feature of our approach is that the analysis is based on a pseudo-likelihood function; hence conventional Bayesian techniques may not be technically correct. We propose an approach using MCMC sampling and a computationally simple method based on an asymptotic posterior distribution. Simulation experiments proved that our method produced parameter estimates that are nearly unbiased even for small sample sizes. An application of our method is illustrated on a Colorectal Adenoma Study.