Genome‐wide association study of schizophrenia in Ashkenazi Jews

Schizophrenia is a common, clinically heterogeneous disorder associated with lifelong morbidity and early mortality. Several genetic variants associated with schizophrenia have been identified, but the majority of the heritability remains unknown. In this study, we report on a case-control sample of Ashkenazi Jews (AJ), a founder population that may provide additional insights into genetic etiology of schizophrenia. We performed a genome-wide association analysis (GWAS) of 592 cases and 505 controls of AJ ancestry ascertained in the US. Subsequently, we performed a meta-analysis with an Israeli AJ sample of 913 cases and 1640 controls, followed by a meta-analysis and polygenic risk scoring using summary results from Psychiatric GWAS Consortium 2 schizophrenia study. The U.S. AJ sample showed strong evidence of polygenic inheritance (pseudo-R(2) ∼9.7%) and a SNP-heritability estimate of 0.39 (P = 0.00046). We found no genome-wide significant associations in the U.S. sample or in the combined US/Israeli AJ meta-analysis of 1505 cases and 2145 controls. The strongest AJ specific associations (P-values in 10(-6) -10(-7) range) were in the 22q 11.2 deletion region and included the genes TBX1, GLN1, and COMT. Supportive evidence (meta P < 1 × 10(-4) ) was also found for several previously identified genome-wide significant findings, including the HLA region, CNTN4, IMMP2L, and GRIN2A. The meta-analysis of the U.S. sample with the PGC2 results provided initial genome-wide significant evidence for six new loci. Among the novel potential susceptibility genes is PEPD, a gene involved in proline metabolism, which is associated with a Mendelian disorder characterized by developmental delay and cognitive deficits.