Excessive activation of brain reward systems by alcohol in OPRM1 118G carriers might be expected to confer susceptibility for alcohol use disorders. Although we in fact found evidence in support of this notion in an ethnically homogenous Caucasian sample 42, others have not. One study found that other polymorphisms within the same haplotype block, but not A118G, were associated with diagnoses of substance dependence 31, while a meta-analysis of all available data for A118G was negative 43. This apparent discrepancy closely parallels recent findings with the 5-HT transporter gene promoter length polymorphism (5-HTTLPR). A robust role of 5-HTTLPR variation has been found using an imaging based endophenotype that is closely related to negative affect 39,44, and these findings are paralleled by experimental data in a non-human primate model 45,46. Nevertheless, findings from 5-HTTLPR association studies in depression have not held up on meta-analysis 47,48. The reasons for these discrepancies are presently unknown, much debated, and critical to resolve. One important possibility is that current diagnostic categories pool genetically heterogeneous phenocopies, as has specifically been proposed in depression 49. If so,
