Aetiological stratification as a conceptual framework for gene-by-environment interaction research in psychiatry.

It has been argued that gene-by-environment interactions (G × E) research is unlikely to progress knowledge about psychiatric disorders, in contrast to genome-wide association (GWA) studies. However, G × E approaches are not alternatives for gene-hunting but a way to identify genetic and biological mechanisms within subgroups of patients exposed to a similar aetiological (environmental) factor via a process of 'aetiological stratification'. This is important as diagnostic categories targeted by GWA studies are inherently heterogeneous and lack biological validity. Aetiological stratification builds on examining possible phenotypic and/or molecular specificity associated with exposure to the environmental factor across multiple potentially relevant disorders, combined with efforts to identify an underlying biological substrate. G × E hypotheses within this framework investigate (1) which genes influence the degree to which individuals develop identified biological alterations that link environmental exposure to specific phenotypic and/or molecular characteristics within or across psychiatric disorders and (2) which genes are implicated in determining the development of psychopathology once this biological alteration has been brought about. As gene-hunting is not a goal in itself, the examination of pathway and/or polygenic risk scores may be more informative than the examination of individual markers, at the same time reducing multiple testing and the associated risk of spurious findings.