Intermediate phenotypes are traits that are presumed to be genetically determined and that are somehow related to one or more disease (Goldman and Ducci 2007). Endophenotypes (Gottesman and Gould 2003; Gould and Gottesman 2006) are a subset of intermediate phenotypes, for simplicity we will not use the term endophenotype in this review. Like animal models, some intermediate phenotypes specifically target one or more of the stages of substance abuse outlined in Table 1. Intermediate pheno-types may involve laboratory-based manipulations that can be performed in a similar manner in both mice and humans. These properties make intermediate phenotypes more amenable to integration with animal models, and thus of special importance for this review. In this manuscript, we will approach the topic of translational genetics form two perspectives. First we will examine studies in which a particular gene was initially identified in an animal model and was then examined in humans to see if polymorphisms in that same gene were related to analogous phenotypes. Second, we will explore the opposite scenario, in which genes that have been implicated by either candidate gene or GWAS studies in humans have been further explored using model organisms.
