Engin Keser, Wangjingyi Liao, Andrea G. Allegrini, Kaili Rimfeld, Thalia C. Eley, Robert Plomin, Margherita Malanchini
Abstract Background Evidence indicates substantial genetic overlap between psychiatric diagnoses. Accounting for these transdiagnostic effects can sharpen research on disorder‐specific genetic architecture and patterns of comorbidity. Methods We applied genomic structural equation modeling to genome‐wide association study summary statistics from 11 major psychiatric disorders to isolate genetic effects shared across disorders (the genomic p factor) from residual genetic effects associated with each disorder ( non ‐ p ). Using these non‐p summary statistics, we examined SNP heritability, genetic correlations among psychiatric disorders, and genetic correlations with external biobehavioural traits spanning socio‐demographic, anthropometric, health‐related, and psychological domains. Results After accounting for transdiagnostic effects, genetic associations between psychiatric disorders changed substantially, with many correlations attenuated and some showing marked shifts in magnitude and direction. Genetic correlations between psychiatric disorders and external biobehavioural traits showed greater specificity. Conclusion Removing transdiagnostic effects provides a more nuanced view of the genetic architecture underlying psychiatric disorders and sharpens inference about genetic relationships between disorders and other comorbid traits. This may inform future work on psychiatric classification, prediction, and treatment research.