BACKGROUND: Alcohol use disorder (AUD) significantly affects over 200 health conditions, causing about 3 million deaths annually worldwide and is approximately 50 % heritable.

METHODS: We conducted a genome-wide association study (GWAS) on self-reported alcohol consumption and alcohol-related challenges (AUDIT score) in a large East Slavs cohort (N = 41,575). Genetic correlations with diverse phenotypes were assessed, and a polygenic risk score (PRS) for alcohol use disorder (AUD) was built and tested in an independent clinical cohort. GWAS and PRS associations were validated across various genetic ancestries.

RESULTS: The East Slavs GWAS identified a highly significant association (p = 2.5 ×10-18) between the rs1229984 SNP and AUD. Transancestral modeling revealed significant associations in Ashkenazi Jews, Tatars, Siberian, and other populations. Additional subthreshold associations (p < 10-6) were found in SNPs within KIF26, OCA2, DLGAP2, and miRNA AL161421 gene regions. SNP-based heritability was estimated at hg = 6.4 % (SE = 1.1 %). Genetic correlation analysis revealed the strongest positive associations with psychiatric traits. We trained the PRS using external summary statistics and individual-level genomic data from our cohort (R2 = 0.013). It outperformed other external PRSs included in the analysis. Validation in the independent clinical cohort showed an AUC of 0.6 (95 % CI = 0.56-0.64), and integrating PRS with non-genetic models increased the AUC by 1.33 %, resulting in 0.762 (p = 8 × 10⁻⁵).

CONCLUSIONS: Our findings suggest a genetic basis for AUD involving genes related to alcohol metabolism and the reward system. The applicability of the PRS across diverse genetic ancestries supports its integration into non-genetic prediction models, enhancing AUD prediction accuracy in diverse populations.