Bayesian Semiparametric Models for Testing Single and Multiple Genetic Pathway Effects in Prostate Cancer

• Activity Number: 521
• Type: Contributed
• Date/Time: Wednesday, August 4, 2010 : 10:30 AM to 12:20 PM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Sounak Chakraborty*+
• Companies: University of Missouri-Columbia
• Keywords: semiparametric regression ; genetic pathway ; reproducing kernel Hilbert space ; prostate cancer ; Bayes factor ; marginal likelihood

Abstract

In this paper, we propose Bayesian semiparametric regression models for continuous and discrete response data. In the proposed models, the multidimensional regression functions are estimated using the reproducing kernel Hilbert space (RKHS) theory. The developed model is then used to quantify the pathway effects of a single genetic pathway on prostate cancer. It is also extended to model multiple pathways jointly by using an additive model. Each pathway effect is estimated nonparametrically with the help of separate functions from RKHS. Based on Bayes factor and marginal likelihood we have proposed a scheme for testing single pathway effect and multiple pathway effects on the prostate cancer indicators. The consistency of the Bayes factors and our hypothesis testing schemes are validated through rigorous mathematical proofs. The proposed model is tested on simulated and real data sets.