Detection and interpretation of psychiatric pleiotropy

Co-occurrence of psychiatric symptoms such as depression, substance use, and psychosis is a pervasive phenomenon. Spectrum models of psychopathology hypothesize that this co-occurrence is related to shared liability... [ view full abstract ]

Co-occurrence of psychiatric symptoms such as depression, substance use, and psychosis is a pervasive phenomenon. Spectrum models of psychopathology hypothesize that this co-occurrence is related to shared liability among different forms of psychopathology and a substantial part of this liability is attributable to genetic influence. Twin and genome-wide association studies (GWAS) have suggested the existence of a close genetic relationship among psychiatric disorders. However, to date, no systematic search of genetic variants shared across these disorders has been conducted. The present study aimed to identify, characterize, and interpret pleiotropic variants associated with multiple psychiatric and related behavioral phenotypes. We applied a multivariate Bayesian test developed by Pickrell (2016) to GWAS summary statistics of 16 psychiatric and psychological traits that have been associated with externalizing, internalizing, and thought disorder (N=16,731~1.2 million depending on the phenotype), as well as summary statistics of heights and BMI as control traits (N=575,442). We used these variants to identify traits that have common genetic causes and conducted a functional interpretation to give insights as to shared etiology of psychiatric disorders. Finally, evidence for the causal relationship between pairs of traits was evaluated based on patterns of correlation between genetic variants associated with each trait.

Reference: Pickrell, J. K., Berisa, T., Liu, J. Z., Ségurel, L., Tung, J. Y., & Hinds, D. A. (2016). Detection and interpretation of shared genetic influences on 42 human traits. Nature genetics, 48(7), 709.