Testing structural models of psychopathology at the genomic level

Abstract

Genome‐wide association studies (GWAS) have revealed hundreds of genetic loci associated with the vulnerability to major psychiatric disorders, and post‐GWAS analyses have shown substantial genetic correlations among these disorders. This evidence supports the existence of a higher‐order structure of psychopathology at both the genetic and phenotypic levels. Despite recent efforts by collaborative consortia such as the Hierarchical Taxonomy of Psychopathology (HiTOP), this structure remains unclear. In this study, we tested multiple alternative structural models of psychopathology at the genomic level, using the genetic correlations among fourteen psychiatric disorders and related psychological traits estimated from GWAS summary statistics. The best‐fitting model included four correlated higher‐order factors – externalizing, internalizing, thought problems, and neurodevelopmental disorders – which showed distinct patterns of genetic correlations with external validity variables and accounted for substantial genetic variance in their constituent disorders. A bifactor model including a general factor of psychopathology as well as the four specific factors fit worse than the above model. Several model modifications were tested to explore the placement of some disorders – such as bipolar disorder, obsessive‐compulsive disorder, and eating disorders – within the broader psychopathology structure. The best‐fitting model indicated that eating disorders and obsessive‐compulsive disorder, on the one hand, and bipolar disorder and schizophrenia, on the other, load together on the same thought problems factor. These findings provide support for several of the HiTOP higher‐order dimensions and suggest a similar structure of psychopathology at the genomic and phenotypic levels.

General vs. Specific Vulnerabilities: Polygenic Risk Scores and Higher-Order Psychopathology Dimensions in the Adolescent Brain Cognitive Development (ABCD) Study

“…This suggests that GWASs for externalizing problems, which have more observable behaviors as target traits, might be better able to capture the narrow phenotypes of interest, resulting in more precise PRSs. This interpretation is consistent with prior finding that molecular genetic markers of antisocial behavior and substance use were only moderately associated with the general genetic factor (Waldman et al, 2020).…”

“…Second, GWASs for internalizing conditions were particularly saturated with general psychopathology variance, resulting in PRSs that captured little specificity. This is consistent with prior findings that genetic vulnerability to depression was the strongest marker of the general genetic factor (Selzam, Coleman, Caspi, Moffitt, & Plomin, 2018;Waldman, Poore, Luningham, & Yang, 2020). Third, genetic vulnerability for internalizing difficulties in adulthood manifests primarily as a general risk for psychopathology in childhood.…”

“…Overall, the pattern of results is in line with the evidence that, while some polygenetic vulnerability is broad and transdiagnostic, a significant proportion of polygenetic risk is specific to narrower psychiatric phenotypes, underscoring the added benefit of fine-grained modeling of psychopathology (Waszczuk et al, 2020). The hierarchical genetic architecture has previously been demonstrated in family and twin studies (Lahey et al, 2016;Pettersson, Larsson, & Lichtenstein, 2016;Waldman, Poore, van Hulle, Rathouz, & Lahey, 2016), and more recently using molecular genomic structural equation modeling (Grotzinger et al, 2019;Waldman et al, 2020). As differential genetic discovery is expected to emerge at each level of the phenotypic structure, future GWASs on redefined hierarchical phenotypes could help create more precise PRSs.…”

Unraveling the relationship of loneliness and isolation in schizophrenia: Polygenic dissection and causal inference

“…However, OCD and BIP are positively correlated with schizophrenia regardless of LNL-ISO based annotations, thus suggesting that these disorders are independently affected by the genetic predisposition to LNL-ISO. These results are in line with recent findings suggesting that schizophrenia, BIP, and OCD could belong to the same psychopathology factor at the genomic level 67 .…”

Validity and utility of Hierarchical Taxonomy of Psychopathology (HiTOP): II. Externalizing superspectrum

“…Newer multivariate methods such as genomic structural equa tion modeling (genomic SEM) 139,140 can be used to model the underlying factor structure of genetic correlations from a set of phenotypes of interest using GWAS summary statistics. These methods enable researchers to move beyond a single disorder or behavior in gene identification efforts, and instead focus on identifying genes contributing to the underlying latent factor(s).…”