An additional implication of these findings is that we should expect to have a better chance of finding and replicating genetic association analyses (using common SNPs) across phenotypes indicative of a known underlying dimension of risk. The current findings suggest that the same SNPs affect each of these three phenotypes, and therefore genetic associations from studies across these three phenotypes can be meaningfully compared. While the current study is limited to DSM criteria, the evidence suggests the possibility of replication success across studies using non-DSM phenotypes, but that likelihood is dependent upon the extent to which the selected phenotypes reflect common biological variation. Non-DSM phenotypes were beyond the scope of the current manuscript and additional research comparing additive genetic effects across DSM and non-DSM-based phenotypes is necessary in order to obtain a definitive conclusion on expectations of replication using broad phenotypes and endophenotypes that may not necessarily harmonize. Findings of common underlying genetic contributions across different phenotypic representations of the vulnerability to substance dependence tested here may also be extended to non-genetic studies to inform models of substance etiology and
