GWAS ask whether a phenotype might be associated with any of a very large number (usually millions) of common single nucleotide polymorphisms (SNPs) distributed across the genome. Unfortunately, use of such large numbers of SNPs requires stringent statistical corrections for false positive findings resulting in many negative studies. The strongest and most consistent GWAS findings for AUD are for SNPs in alcohol metabolizing genes in Asian samples. Another strong finding was for a meta-analysis of an alcohol consumption phenotype in 26,316 individuals from 12 European-ancestry population-based samples with replication genotyping in another 21,185 individuals for one SNP in the autism susceptibility candidate gene 2 (AUTS2).51 Furthermore, a GWAS for alcohol dependence in a discovery / replication dataset of 16,087 European and African ancestry individuals confirmed known risk variants and identified population specific novel variants but also cross population risk genes.52 There have been other major findings from GWAS for AUD, alcohol-related phenotypes and comorbid diseases, however other than for SNPs in the alcohol metabolizing genes there are few if any commonalities across studies,41 most likely due to insufficient sample sizes combined with small effects of individual variants.
