A polygenic p factor for major psychiatric disorders

Abstract

It has recently been proposed that a single dimension, called the p factor, can capture a person’s liability to mental disorder. Relevant to the p hypothesis, recent genetic research has found surprisingly high genetic correlations between pairs of psychiatric disorders. Here, for the first time, we compare genetic correlations from different methods and examine their support for a genetic p factor. We tested the hypothesis of a genetic p factor by applying principal component analysis to matrices of genetic correlations between major psychiatric disorders estimated by three methods—family study, genome-wide complex trait analysis, and linkage-disequilibrium score regression—and on a matrix of polygenic score correlations constructed for each individual in a UK-representative sample of 7 026 unrelated individuals. All disorders loaded positively on a first unrotated principal component, which accounted for 57, 43, 35, and 22% of the variance respectively for the four methods. Our results showed that all four methods provided strong support for a genetic p factor that represents the pinnacle of the hierarchical genetic architecture of psychopathology.

Editorial: The psychopathology p factor: will it revolutionise the science and practice of child and adolescent psychiatry?

“…A general genetic factor that influences eight major psychiatric disorders using full‐ and half‐sibling data from Swedish national registers has been reported (Pettersson et al., ). Complementary to these findings from family data, similar conclusions were reached with three other methods that employed measured genotypes (Selzam, Coleman, Caspi, Moffitt, & Plomin, ). We see the p factor being covaried for in genetic studies on specific psychopathology too.…”

The p factor: genetic analyses support a general dimension of psychopathology in childhood and adolescence

“…We used publicly available genome‐wide association summary statistics for eight major psychiatric traits: autism spectrum disorder (Grove et al., ), major depressive disorder (MDD; Wray et al., ), bipolar disorder (BIP), schizophrenia (SCZ; Pardiñas et al., ), attention deficit hyperactivity disorder (ADHD; Demontis et al., ), obsessive–compulsive disorder (OCD; International Obsessive Compulsive Disorder Foundation Genetics Collaborative (IOCDF‐GC) and OCD Collaborative Genetics Association Studies (OCGAS), ), anorexia nervosa (AN; Duncan et al., ) and posttraumatic stress disorder (PTSD; Duncan et al., ). For each psychiatric disorder, polygenic scores for each TEDS participant were created in LDpred (Vilhjálmsson et al., ), assuming a fraction of causal markers of 1 (analysis steps were similar to Selzam et al., ).…”

Shared risk alleles with discordant polygenic effects: Disentangling the genetic overlap between ASD and ADHD

“…However, they are unlikely to be limited to polygenic ASD and ADHD risk. The identification of discordant EA association profiles for ASD and MDD risk, across the same ASD risk alleles, and likewise the discordant polygenic EA association effects for ADHD and BD, across the same ADHD risk alleles, supports the widespread pleiotropy among neuropsychiatric disorders 30 . In particular, it suggests complex genetic interrelationships between ASD and MDD, and between ADHD and BD that may involve negative covariance.…”

“…This discordant association pattern is strongest for measures of years-of-schooling and college-completion 21,28 . Recent evidence for a polygenic p-factor, shared across major psychiatric disorders including ASD and ADHD 30 , suggests overarching genetic similarities between neurodevelopmental disorders. This may also involve shared polygenic variation among ASD, ADHD and EA, acting through complex pleiotropic, mediating or confounding mechanisms.…”