Integrative Bayesian variable selection with gene-based informative priors for genome-wide association studies

Abstract

BackgroundGenome-wide Association Studies (GWAS) are typically designed to identify phenotype-associated single nucleotide polymorphisms (SNPs) individually using univariate analysis methods. Though providing valuable insights into genetic risks of common diseases, the genetic variants identified by GWAS generally account for only a small proportion of the total heritability for complex diseases. To solve this “missing heritability” problem, we implemented a strategy called integrative Bayesian Variable Selection (iBVS), which is based on a hierarchical model that incorporates an informative prior by considering the gene interrelationship as a network. It was applied here to both simulated and real data sets.ResultsSimulation studies indicated that the iBVS method was advantageous in its performance with highest AUC in both variable selection and outcome prediction, when compared to Stepwise and LASSO based strategies. In an analysis of a leprosy case–control study, iBVS selected 94 SNPs as predictors, while LASSO selected 100 SNPs. The Stepwise regression yielded a more parsimonious model with only 3 SNPs. The prediction results demonstrated that the iBVS method had comparable performance with that of LASSO, but better than Stepwise strategies.ConclusionsThe proposed iBVS strategy is a novel and valid method for Genome-wide Association Studies, with the additional advantage in that it produces more interpretable posterior probabilities for each variable unlike LASSO and other penalized regression methods.Electronic supplementary materialThe online version of this article (doi:10.1186/s12863-014-0130-7) contains supplementary material, which is available to authorized users.

Bayesian Genome- and Epigenome-Wide Association Studies with Gene Level Dependence

“…For example, a Dirichlet process model may be used for gene‐level intercepts within a regression model for marker probabilities that includes additional prior covariates (Lewinger et al, ). Incorporating additional gene‐level prior information, such as allowing greater dependence within known functional gene networks (Zhang et al, ), is also a promising direction of future work.…”

“…Moreover, markers that are statistically associated with a given phenotype may not affect the phenotype directly, especially if markers are correlated (e.g., linkage disequilibrium in genetic data). Our gene‐level model and other prior information can also be used in the context of model inclusion probabilities within a multi‐marker regression approach (Wilson et al, ; Duan and Thomas, ; Zhang et al, ), to select markers that have novel predictive power for a given phenotype and are therefore more likely to be causal. We have described a generally applicable approach to posterior computation in which sampling from the gene‐level prior is incorporated into the sampling scheme for the specified association model.…”

New Algorithm and Software (BNOmics) for Inferring and Visualizing Bayesian Networks from Heterogeneous Big Biological and Genetic Data

“…These often incorporate, explicitly or implicitly, a model complexity penalty to prevent overfitting. In addition, an initial search state (network structure prior) has to be specified—selection of an optimal (or biologically meaningful) prior has received much attention (e.g., Friedman et al, 1999; Steele et al, 2009; Keilwagen et al, 2010; Zhang et al, 2014), in part, because it may alleviate the scalability problem to a degree. The default BNOmics algorithm is completely agnostic in regard to the BN structure and prior and therefore is entirely data driven.…”

Synthetic data generation with probabilistic Bayesian Networks

“…Bayesian Networks (BNs)based dependency modeling is an established computational biology tool that has been rapidly gaining acceptance in big biological data analysis (Branciamore et al 2018;Cooper et al 2015;Gogoshin, Boerwinkle, and Rodin 2017;Jiang, Barmada, and Visweswaran. 2010;Lan et al 2016;Neapolitan, Xue, and Jiang 2014;Needham et al 2007;Pe'er 2005;Piatetsky-Shapiro and Tamayo 2003;Qi, Li, and Cheng 2014;Rodin et al 2005;Rodin et al 2012;Sedgewick et al 2019;Sherif, Zayed, and Fakhr 2015;Wang, Audenaert, and Michoel 2019;Yin et al 2015;Zeng, Jiang, and Neapolitan 2016;Ziebarth, Bhattacharya, and Cui 2013;Zhang and Shi 2017;Zhang et al 2017;Zhang et al 2014). Comprehensive treatments of BN methodology, and probabilistic networks (PNs) in general, can be found in numerous textbooks (Pearl 1988;Pearl 2009;Russell and Norvig.…”