GWA studies have shown replication of SNPs associated with AD in the COGA candidate gene studies (Bierut et al., 2010; Edenberg et al., 2010); however, in an effort to create SNPs that captured unique information by pruning them based on LD, some of the replicated SNPs were not included in the model. An expanded candidate gene sum score incorporated more SNPs that met nominal significance levels in the COGA and SAGE GWAS samples (Supporting Information Table S1), but did not have a significantly different AUC compared with the candidate gene sum score composed of pruned SNPs. In these data, we have previously demonstrated that the missense SNP rs1229984 is associated with AD at p < 5×10−8 (Bierut et al., 2012). This variant, previously well-recognized for its protective influence on alcoholism in Asians, has also been found to exert an influence on alcoholism risk in Caucasians and African-Americans. However, it is fairly uncommon in non-Asian samples (< 5%) and is poorly captured by content on commercially available GWAS platforms, due to lack of LD with neighboring SNPs. We assessed the discriminatory
