The study “Multivariate genome-wide analysis of education, socioeconomic status and brain phenome” by five Yale researchers was published November 21 in Nature Human Behaviour.
The first author is Frank R. Wendt, PhD, Postdoctoral Fellow in the Yale Department of Psychiatry. Co-authors are Gita A. Pathak, PhD; John H. Krystal, MD; Joel Gelernter, MD; and senior author Renato Polimanti, PhD, MSc.
Read more about the study below in text prepared by Wendt:
Why did we perform this study? - Education and socioeconomic variables are routinely implicated in studies of mental health and behavior as either risk factors (lower socioeconomic status or education level) of protective factors (higher socioeconomic status or education level). Interestingly, different education or socioeconomic status variables can have different effects. Recently, education and socioeconomic status have been subjected to large genetic studies which demonstrate that genetic liability to these traits correlate with psychopathology and mental health. We investigated how these genetic relationships influence what we can learn about mental health from large-scale genetic studies.
What did we do in this study? - With data from over 1 million participants we used statistics and bioinformatics to adjust large genetic studies of psychiatric disorders, personality traits, externalizing behaviors, social science outcomes, and brain imaging phenotypes for the genetic liability to education and socioeconomic status. We used these data to investigate the questions (1) which cell and tissue types are informative for each trait? (2) how do traits correlate after removing effects of education and socioeconomic status? (3) is there evidence of causality between traits after removing effects of education and socioeconomic status?
What did we discover? - The ability of common variation in the genome to explain trait variability (termed "heritability") generally decreased, suggesting that part of what we understand from genetic studies of mental health is due to education and socioeconomic status variables. Interestingly, the heritability of two traits (neuroticism and subjective well-being) increased and this increase was accompanied by a greater number of risk variants. Our most impactful finding stems from the use of methods to untangle correlation from causation. By modeling a latent factor connecting trait pairs, we uncovered putative causal relationships between brain volumes and mental health. Finally, we uncovered specific inhibitory and excitatory cell type information enriched in the genetic signal of mental health traits after adjusting for education and socioeconomic status. Importantly, these results could not be detected without adjusting for the effects of education and socioeconomic status.
How does this study impact mental health? - By systematically removing the genetic effects of education and socioeconomic status we reveal biology unique to psychiatric disorders, personality traits, externalizing behaviors, social science outcomes, and brain imaging phenotypes. Derived from these multi-trait analysis of genetic data, trait specific genetic discoveries will be essential for identifying and understanding potential treatment strategies for mental health or related psychopathologies.