Their diagnostic distinction is therefore therapeutically pragmatic. More recently, the PGC Schizophrenia and BPD Workgroups tested for variants distinguishing the two.20 This was done in a variation on traditional case-control analysis, by conducting GWAS of schizophrenia vs BPD as opposed to cases vs controls. While no individual variants reached genome-wide significance, several were moderately significant (P < 10−5). More importantly, a polygenic score of schizophrenia vs BPD was seen to significantly predict schizophrenia vs BPD status in independent samples. Despite having the power to detect polygenic effects, more than likely, the study was underpowered to detect individual variant effects. However, it did suggest that a number of individual variants are disorder-specific. One caveat to this kind of study is that it assumes that the 2 disorders in question are themselves unitary, where they could in fact represent partially distinct genetic etiologies. Nevertheless, refinement and extension of analyses like this hold out hope of utilizing genetic information to assist in differential diagnosis, which might be especially useful in these 2 disorders, as they are sometimes difficult to distinguish cross-sectionally because of overlapping clinical features such as psychosis and agitation.