Despite the paucity of findings from GWAS on substance addiction, the present analyses indicate that common SNPs on existing GWA platforms capture substantial genetic information regarding the vulnerability to substance dependence in subjects of European ancestry. The bivariate findings also confirm the expectation that different phenotypic representations of generalized substance dependence are influenced by the same SNPs. Our ability to find largely overlapping SNP effects across substances of abuse may be attributable to our analytic approach. As described and demonstrated elsewhere (36-39), relative to GWAS or candidate studies that look for the independent effects of individual SNPs, GCTA aggregates variance across all SNPs and typically explains a greater proportion of variance because the effects of individually non-significant SNPs are included in the estimate. Our results suggest that GWA studies on these phenotypes that have larger samples will discover additional significant genetic associations, as has occurred for other phenotypes (40, 41). It should be noted that additional research is needed to confirm these effects in African populations and to determine whether the observed effects can be attributed to shared genetic factors across different ethnic backgrounds.
