Other major findings from GWAS for AUD, alcohol-related phenotypes and comorbid diseases are listed and discussed in Rietschel and Treutlein [12]. A number of new chromosomal regions and candidate genes have been identified, however other than for the alcohol metabolizing genes there are few if any commonalities across studies. The initial expectation was that robust candidate gene associations with alcoholism, for example the widely replicated GABRA2 finding would be replicated in GWAS but this has not been the case. In a GWAS of nearly 2000 alcoholics and 2000 controls, five GABRA2 SNPs had only nominal association (p < 0.05) with AUD, with odds ratios of 1.1 – 1.2 [71]. Other commonly reported candidate loci for AUD (e.g. in DRD2, COMT, ADH1C, OPRM1) have not been replicated in GWAS of AUD [72]. It should also be noted that genetic variants that confer risk only in trauma-exposed individuals (G x E interactions) will not be picked up by GWAS.
