Putting Genes In Context: Applying Integrative 'Omics To Alcohol Use Disorder

Abstract

Alcohol misuse is the 4th leading cause of preventable mortality in the United States and contributes to over 3 million deaths worldwide each year. The use of alcohol in a pathologic fashion, alcohol use disorder, affects all populations, and the prevalence of this disorder has remained stable over the last several decades. The risk of developing alcohol use disorder has a substantial genetic component, with approximately 50% heritability. Some genetic risk variants are known, most notably variation in genes that are involved in the metabolism of alcohol – alcohol dehydrogenase (ADH) and aldehyde dehydrogenase (ALHD) genes. Our goal is to integrate ‘omic data from multiple resources to accelerate discovery of relevant genetic contributions to disease. By leveraging complementary data sources, we can narrow the search space and boost power. This successful approach is illustrated through our study of alcohol use disorder.As our first step, we mapped cis-methylation quantitative trait loci (cis-meQTL) and cis-expression quantitative trait loci (cis-eQTL) that correlate with DNA methylation and RNA expression in the alcohol metabolizing genes (ADH gene family and ALDH2) in pre-frontal cortex samples from 240 decedents using the Lieber Institute for Brain Development (LIBD) biobank. We found novel cis-meQTL SNPs for ADH1C (smallest P = 2.6×10-11) that associate with alcohol use disorder (smallest P = 2.3×10-5, surpassing Bonferroni correction) in our newly assembled alcohol use disorder GWAS meta-analysis. These novel SNPs are not in linkage disequilibrium with prior GWAS-identified SNPs, and this novel approach demonstrates the power for discovery that focuses on regulatory SNPs to narrow the search space for association testing. Integrative omics approach can be extended genome wide by focusing on genes highly expressed in the brain. By leveraging multi-omics data (genome-wide DNAm, RNAexp, and SNP genotypes) from multiple sources, we can focus on genetic variation at the intersection of these datasets that can accelerate the discovery of new genes for diseases and their susceptibility variants beyond the strategy of increasing sample sizes for meta-analysis. This proposed approach focuses on neurobiologically meaningful genes and variants and will expand our understanding of the etiology of alcohol use disorder and the neuropathology underlying this disease.