First, hypothesis-driven candidate gene studies had poor statistical power by contemporary standards. Even for a relatively large candidate gene study with power-favorable multiple comparison correction, power would have been poor to detect the genetic effects typical for GWAS of human diseases. As the genetic effects for schizophrenia may be smaller than for other human diseases (Purcell et al., 2009, Shi et al., 2009, Stefansson et al., 2009), nearly all hypothesis-driven candidate gene studies were under-powered. Given what we now know about the genetic architecture of schizophrenia, a typical candidate gene study requires sample sizes ~11,000 cases plus controls for a single marker, 17,500 subjects for 10 markers, and 24,000 subjects for 100 markers (Supplemental Methods). Future association studies of schizophrenia should use similarly realistic power calculations.
