Bayesian Latent Variable Modeling of Longitudinal Family Data for Genetic Pleiotropy Studies

TL;DR

This paper introduces a Bayesian latent variable model for analyzing longitudinal family data in genetic pleiotropy studies, effectively handling multiple phenotypes and complex correlations, with demonstrated efficiency and applicability to diabetes-related GWAS.

Contribution

It presents a novel Bayesian approach with an efficient MCMC algorithm for joint modeling of multiple phenotypes in family data, including phenotype and model selection strategies.

Findings

Effective modeling of continuous and binary phenotypes

Accurate phenotype and model selection using Bayes factors and spike-and-slab priors

Successful application to type 1 diabetes GWAS data

Abstract

Motivated by genetic association studies of pleiotropy, we propose here a Bayesian latent variable approach to jointly study multiple outcomes or phenotypes. The proposed method models both continuous and binary phenotypes, and it accounts for serial and familial correlations when longitudinal and pedigree data have been collected. We present a Bayesian estimation method for the model parameters, and we develop a novel MCMC algorithm that builds upon hierarchical centering and parameter expansion techniques to efficiently sample the posterior distribution. We discuss phenotype and model selection in the Bayesian setting, and we study the performance of two selection strategies based on Bayes factors and spike-and-slab priors. We evaluate the proposed method via extensive simulations and demonstrate its utility with an application to a genome-wide association study of various complication phenotypes related to type 1 diabetes.