We meta-analyzed three published GWAS: the Million Veteran Program (MVP)19 GWAS of AUD (European [EUR] N = 202,004; Ncases = 34,658), with case status derived from International Classification of Diseases (ICD) codes of alcohol-related diagnoses from electronic health records (EHR) data, the Psychiatric Genomics Consortium (PGC) GWAS of alcohol dependence12 (cases based on DSM-IV diagnoses; EUR unrelated genotyped N = 28,757; Ncases = 8485) and the Collaborative Studies on Genetics of Alcoholism (COGA) GWAS of alcohol dependence (cases based on DSM-IV diagnoses; EUR unrelated genotyped N = 4849; Ncases = 2411)20. Genetically calculated principal component 1 (PC1) was added as additional covariate in analyses of individual GWAS data from individuals with European ancestry. The final meta-analysis in PGC was performed by combining summary statistics weighted on sample size from individual datasets. Appropriate PCs calculated using SNP data were also included as a covariate in MVP19 and COGA datasets.
