To maximize contribution of genetic variants to the risk of AD and to reduce heterogeneity, a subsample of COGA families was selected for a family-based association study. These families consisted of primarily self-reported European American. Families were prioritized based on (i) largest number of alcohol dependent family members with DNA, (ii) the largest number of family members with DNA and electrophysiological data, and (iii) the largest number of family members with DNA. The final sample consisted of 118 large European American families, with 2,322 individuals with available DNA (Kang et al., 2012; Wang et al., 2013). Genotyping was performed at the Genome Technology Access Center at Washington University School of Medicine in St. Louis using the Illumina Human OmniExpress array 12.VI as well as at the Center for Inherited Diseases using the Illumina 1M array. Further genotyping details, including SNP and sample cleaning, are available in Wang et al. (2013). The average number of genotyped individuals in a family was 20, and there was an average 5.9 members meeting criteria for DSM-IV AD. A total of 684 individuals were classified as alcohol dependent and 964 as unaffected, and 327 individuals endorsed the withdrawal criterion, while 1,459 did not.
