The sample for this study was drawn from the Collaborative Study on the Genetics of Alcoholism (COGA)17, 28, 29 (dbGaP Study Accession: phs000763.v1.p1). Probands (i.e., index individuals) were identified through alcohol treatment programs at seven U.S. sites. Probands and their families were invited to participate if the family was sufficiently large with two or more members in the COGA catchment areas. Comparison families were recruited from the same communities. The Institutional Review Boards at all data gathering sites approved this study, and written consent was obtained from all participants. Within the larger COGA sample, we selected parents and children (confirmed by genotyping) with relevant phenotypic and genome-wide genotypic data available. This was done separately by European ancestry (EA) and African ancestry (AA) groups, as defined by principal components in GWAS data calculated using SNPrelate30, to avoid population stratification in the downstream analyses incorporating genome-wide polygenic scores31. Analytic sample sizes differed across the phenotypes of interest and ranged from 4321 to 4806 in EAs (51% female, Mage (SD) = 30.72 (9.81) years) and 1616 to 1960 in AAs (52% female, Mage (SD) = 29.04 (8.76) years) and are specified in the results.
