The available data indicate that psychiatric disorders are highly ‘polygenic’ and we now expect hundreds or thousands of individual variants to be associated with each disorder. A promising strategy to deal with the small effect sizes and plethora of results is to adopt a pathway- and network-informed interpretation of GWAS hits. An analysis by Cao and Moult27 found that while only a small fraction of known drug targets are in GWAS loci (12 of 353 drug targets for 81 diseases), known drug targets are enriched three-fold in the nearest neighbour interactors (proteins that physically interact with a given protein) of genes in GWAS loci and are also enriched in second order interactors. This is supported by GWAS results in type 2 diabetes28 which found that pathways targeted by anti-diabetes drugs are enriched in genes from GWAS and their direct protein interactors. This pool of GWAS hits, their interacting partners and networks provides a resource for the identification of novel drug targets and for drug repositioning.
