INTEGRATIVE PRIORITIZATION OF GENOMIC LOCI FOR ALCOHOL USE DISORDERS

Alcohol use disorder (AUD) is a persistent disease (10%) with a polygenic architecture that complicates the localization and identification of susceptibility loci. Prior studies suggest that AUD related phenotypes, such as... [ view full abstract ]

Alcohol use disorder (AUD) is a persistent disease (10%) with a polygenic architecture that complicates the localization and identification of susceptibility loci. Prior studies suggest that AUD related phenotypes, such as dependence and vulnerability are genetically influenced (h²=40-60%), but no single gene or robust set of gene have been identified. The integration of gene expression and functional experimentation evidence across species may facilitate human-based association analyses. Typically, applications of experimental results from humans and model-organisms in human genetic association analyses have been limited to post-genomewide association study screens of only the most-highly associated variants, which often fail to replicate. The present study investigates how highly connected genes related to specific features of AUD (e.g., craving and withdrawal) that are linked by their association to alcohol-related behaviors and processes in humans, Drosophila, mice and rats, contribute to individual differences in alcohol dependence severity (AD) in humans. We examine a new pipeline to assess the ability of genes identified from expression-QTL (eQTL) studies in model-organisms to prioritize genes for AD. eQTL-based gene sets were derived from the GeneWeaver, an Ontological Discovery Environment. Over 14,000 genes were initially identified, ranked, and systematically examined as joint predictors of alcohol dependence in several human studies (i.e., COGA, SAGE, and Australian GWAS) to identify sets of SNPs in an subset of genes that are highly connected to AD. In addition, we apply a novel Bayesian Hierarchical Model with annotation-specific effect-size priors to identify loci with the strongest enrichment for causality.