In a typical GWAS study, participants are recruited and provide written informed consent and blood or saliva samples for DNA extraction and genotyping using microarray (“00” indicates missing genotype call). Basic quality controls are performed to remove SNPs with low minor allele frequencies (MAF), high genotype missingness rate, or violation of Hardy Weinberg Equilibrium expectations (HWE) and remove samples with high genotype missingness. Since genetic factors often differ according to ancestry, principal component analysis (PCA) is performed on the data after quality controls with reference genomes — for example, the 1000 Genomes Project (165) — to infer the genetic ancestries of the study samples and remove genetic outliers (the results from different ancestry groups can then be combined by meta-analysis). Then, the remained samples and the data after quality control are imputed for millions more variants (imputed genotypes and SNPs [IMP], labeled in purple) using reference genomes (165–168). Imputation takes advantage of known patterns of linkage disequilibrium to provide useful data for many more variants than are genotyped directly. A study trait, in the context of either case-control status (for example, AUD) or continuous measurement (for example, AUD criterion counts), is assessed in the cohort. Regression models implemented in computational tools (169–175) are applied to test the association between each variant and the studied trait within the genetically inferred population group, adjusting for covariates including age, sex, and the top principal components of ancestry. Variants with P < 5 × 10^-8 are considered genome-wide significant (GWS) after multiple testing corrections for the number of independent genomic regions evaluated (176).