Bayesian Latent Variable Modeling of Genetic Pleiotropy Data

• Activity Number: 330
• Type: Invited
• Date/Time: Tuesday, August 2, 2016 : 10:30 AM to 12:20 PM
• Sponsor: JCGS-Journal of Computational and Graphical Statistics
• Author(s): Lei Sun* and Lizhen Xu and Radu V. Craiu and Andrew Paterson
• Companies: University of Toronto and University of Toronto and University of Toronto and Hospital for Sick Children
• Keywords: Bayesian inference ; Latent variable ; Marginal data augmentation ; Markov chain Monte Carlo ; Statistical genetics ; Pleiotropy

Abstract

Motivated by genetic studies of pleiotropy, where one is interested if a genetic marker is associated with multiple phenotypes/outcomes, we propose a Bayesian latent variable approach to jointly study multiple outcomes. We develop a model that can deal with both continuous and binary phenotypes, and account for serial and cluster correlations. We consider Bayesian estimation for the model parameters, and we develop a MCMC algorithm that builds upon hierarchical centering and parameter expansion techniques to efficiently sample from the posterior distribution. We evaluate the proposed method via extensive simulations and demonstrate its utility with an application to an association study of different complication outcomes related to Type 1 Diabetes. We also discuss related inferential challenges.