F93. PRIORITIZING CANDIDATE GENES IN GENOMIC RISK LOCI FOR ALCOHOL-RELATED BEHAVIOR WITH FUNCTIONAL AND STATISTICAL FINE-MAPPING

• Published in: European neuropsychopharmacology Vol. 87; p. 255
• Main Authors: Chen, Zeli, Schipper, Marijn, Posthuma, Danielle, Savage, Jeanne
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.10.2024
• ISSN: 0924-977X
• DOI: 10.1016/j.euroneuro.2024.08.504

Alcohol use disorder (AUD) can induce severe social, health and possible legal consequences that collectively lead to enormous public burden worldwide. It is known to have both strong genetic and environmental influences and has a reported heritability of approximately 50%. Genome-wide association studies (GWAS) have resulted in inconsistent SNP-level signals for different alcohol-related phenotypes. We aim to prioritize a robust set of candidate genes from the numerous risk loci underlying various alcohol-related behavioral phenotypes via a combination of fine-mapping tools, while taking into consideration the heterogeneity of alcohol-related behavior. We summarized over 1000 genomic risk loci from a variety of previous large-scale alcohol related GWASs. The loci of interest were fine-mapped using GWAS summary statistics of multiple alcohol-related behaviors we developed with UK Biobank data (N=410414). For European ancestry subjects, we applied functionally informed fine-mapping which sets prior probabilities for the SNPs of interest based on functional annotation data. For non-European subjects, a trans-ancestry fine-mapping approach was adopted, accounting for population specific genetic architecture. Both approaches provide SNP-level information with posterior inclusion probability, which is then used to pinpoint the effector genes via integration of SNP-to-gene information and functional convergence-based evidence of polygenic enrichments. Main phenotypes examined include AUD diagnosis, daily alcohol intake level, self-reported drinking frequency and problem drinking from the Alcohol Use Disorder Identification Test (AUDIT), as well as a latent class categorization of patterns of alcohol use and comorbid psychiatric disorders. We identified an extensive list of over 100 potential causal candidate genes, of which 13 risk genes have been identified for AUD, 26 risk genes for the AUDIT-problems scale, 7 genes are shared between the two phenotypes. We have repeated the association between alcohol linked behavior and genes previously reported for ethanol-metabolism, taste regulation and emotions and mood control. In addition, we have identified new effector genes such as TRMT10A, PSMD7, and ACVR2B. For further analysis, we intend to build the link between alcohol misuse and underlying biological pathways with cell-type specific gene-set enrichment tests. We would also like to validate the candidate genes via polygenic risk profiling for prediction of alcohol-related behavior. With this study, we summarize the risk loci that have been reported in past alcohol-related literature, to pinpoint the effector genes and biological pathways underlying alcohol consumption and pathological alcohol-related behavior. Our findings suggest problems-related phenotypes share many of the same effector genes while certain genes could be phenotype-specific, which may be explained by the heterogeneity of the disorder. The identification of genes specific to alcohol-related problems can promote deconstruction of the genetic architecture and individual differences underlying alcohol misuse, which contributes to early intervention and treatment in individuals.