Eating Disorders Working Group of the Psychiatric Genomics Consortium
Prof. Cynthia Bulik, UNC Chapel Hill, North Carolina & Karolinska Institutet
The Eating Disorders Working Group of the Psychiatric Genomics Consortium is building on momentum and enthusiasm generated by our 2019 Nature Genetics paper that ignited interest in both psychiatric and metabolic origins of anorexia nervosa. Our next steps include not only increasing the sample size for anorexia nervosa, but developing parallel initiatives for bulimia nervosa and binge-eating disorder. The Eating Disorders Genetics Initiative (EDGI), funded by the National Institute of Mental Health in the US, New Zealand, Australia, and Denmark, with parallel funded efforts in the UK and Sweden will form the backbone of the next phase of sample and data collection. EDGI is being cloned in several countries around the world utilizing harmonized and translated phenotypic assessments and DNA collection procedures. A collection for avoidant/restrictive food intake disorder (ARFID) is also underway. A rich array of secondary analyses are beginning to explicate similarities and differences across the three eating disorders and their relation to other psychiatric and metabolic phenotypes. Prof. Bulik will present novel findings and projections for the PGC-ED in 2020-2021.
Accelerating innovative drug discovery: An introduction to the Psychiatry Consortium
Dr. Laura Ajram, Psychiatry Consortium
The Psychiatry Consortium is a strategic collaboration of 2 leading medical research charities and 6 pharmaceutical companies focusing on the challenge of identifying and validating novel drug targets to address the unmet therapeutic needs of the people living with mental health conditions. The Consortium is managed by the Medicines Discovery Catapult and supported by the Wellcome Trust via a grant which funds our academic outreach work. The Psychiatry Consortium acts as a syndicate, whereby the Partners collectively share the funding of, and therefore the risk associated with drug discovery. This is a new model of funding is unlike typical grant funding- here we work with the applicant, provide project management support and access to Industry and commercial know-how, in a collaborative effort between the PI, the Pharma Partners and our Partner CROs to robustly validate targets to Industry standards.
Network signature of complement component 4 variation in the human brain identifies convergent molecular risk for schizophrenia
Dr. Michael Gandal, University College Los Angeles
Genome-wide association studies have successfully identified hundreds of loci associated with schizophrenia (SCZ). Fine-mapping of the strongest signal found that risk is partially mediated by complex structural variation of the complement component 4 (C4) genes and resulting increased expression of C4A. Although C4A is believed to partake in synaptic pruning, its precise function in the human brain—and relation to other risk factors—remains unknown, due to lack of an appropriate experimental system and lack of evolutionary conservation. Here we perform a large-scale investigation of the systems-level architecture of C4A co-expression in the human frontal cortex and how this network is remodeled by C4A copy-number variation. We identify a putative transcriptomic signature of synaptic pruning as well as spatiotemporal and sex differences in C4A co-expression with largest effects observed in frontal cortical brain regions of middle-aged males. We find that negative, but not positive, co-expression partners of C4A exhibit substantial enrichment for SNP-heritability in SCZ. In line with this finding, there is limited evidence that the complement system is a core SCZ-relevant pathway in comparison to synaptic components. Overall, our results highlight human brain-specific function of C4A and strong and specific convergence of polygenic effects in SCZ pathophysiology.
Genetic Identification of Cell Types Underlying Brain Complex Traits Yields Insights Into the Etiology of Parkinson’s Disease
Dr. Julien Bryois, Karolinska Institutet
Genome-wide association studies (GWAS) have discovered hundreds of loci associated with complex brain disorders but it remains unclear in which cell types these loci are active. In this study, we integrated GWAS results with single-cell transcriptomic data from the entire mouse nervous system to systematically identify cell types underlying brain complex traits. We show that psychiatric disorders are predominantly associated with projecting excitatory and projecting inhibitory neurons. Neurological diseases were associated with different cell types, which is consistent with other lines of evidence. Notably, Parkinson’s disease was not only genetically associated with cholinergic and monoaminergic neurons (which include dopaminergic neurons) but also with enteric neurons and oligodendrocytes. Using post-mortem brain transcriptomic data, we confirmed alterations in these cells, even at the earliest stages of disease progression. Our study provides an important framework for understanding the cellular basis of complex brain maladies, and reveals an unexpected role of oligodendrocytes in Parkinson’s disease.
Measurement of Trauma Exposure and Traumatic Stress: Moving Beyond Case Status
Dr. Erika Wolf, Boston University
Technological and statistical advances allow for greater precision and complexity in multi-omics research. However, these breakthroughs can only further the understanding of disorder pathophysiology and the development of novel therapeutic molecular targets if the phenotype assessment is also sophisticated. Complex behavioral phenotypes, such as trauma-related psychiatric disorders, have too often been reduced to binary diagnostic variables, which have inadequate reliability and statistical power, are not construct valid, and are far removed from the end product of biological variables. There are alternative approaches to better represent the phenotype side of the equation and a well-developed scientific literature on psychological construct measurement. These range from simple assessment of symptom severity, to more advanced endophenotypic modeling of symptom heterogeneity and overlap across psychiatric diagnoses (i.e., using latent variable analyses), to consideration of the moderating effects of trauma on the association between epi/genetic factors and psychiatric variables. Many of these analytic approaches also lend themselves to phenotype harmonization efforts and are particularly well suited to understanding cross-disorder epi/genetic risk. There are also pitfalls to avoid, such as misapplication of network and class models. Improved understanding of approaches to phenotype measurement will ultimately enhance the identification of the etiology, pathophysiology, and treatment of psychiatric disorders.
Analysis of signal from GWAS of Alcohol and Nicotine Behaviors Provides Targets for Laboratory Research
Dr. Alexander Hatoum, University of Colorado Boulder
GWAS of complex traits in humans point to relevant loci, but do not themselves reveal potential mechanisms. GWAS in animal models provide additional data, but translation of human behaviors to animal models and vice versa is not always simple. As larger GWAS of alcohol and nicotine behaviors and substance use disorders discover genome-wide significant loci, we are presented with the opportunity to prioritize genes for additional study. Utilizing bioinformatic approaches to analyze data from human and animal studies, we make GWAS-informed recommendations to laboratory researchers on the most promising genes. Finally, we speculate on the pharmacological utility of these genes and what types of laboratory paradigms will be needed to test for their applicability as drug targets. Preliminary results suggest that GWAS heritability estimates for alcohol and nicotine behaviors are enriched for conservation across mammalian species, with half of the heritability driven by conserved genetic variability. GWAS of nicotine behaviors are enriched for genes utilized in animal models, particularly CHRNA3, CHRNA4, CHRNA5, CHRNB2, CHRNB4 variants and CYPA7. Signals from alcohol behavior GWAS were not enriched in neuro-molecular associations with alcohol animal use behaviors, thus presenting caution in interpreting animal findings in the context of context of human GWAS data.
Depression Grand Challenge
Prof. Michelle Craske, University of California Los Angeles
UCLA Depression Grand Challenge is comprised of four components: identifying underlying causes of depression, discovery neuroscience, innovative treatment network, and hope and awareness. The DGC aims to elucidate the genetic and environmental causes of depression in its varied forms in large samples from the UCLA Health System (> 2M patients). The Discovery Neuroscience component addresses the neural mechanisms of genetic and environmental risk in studies ranging from the fruit fly to humans. The Hope and Awareness component aims to understand stigma around depression and ways to reduce stigma. The Innovative Treatment Network aims to develop novel and more effective treatments for depression and anxiety, informed by discovery neuroscience findings, and increase access to existing effective treatments. The STAND program (Screening and Treatment for Anxiety and Depression) is a scalable model of treatment delivery that is implemented for UCLA students and soon to be implemented in local community colleges where rates of depression and anxiety are especially high and mental health resource are limited. The current platform, results to date, and future directions of STAND will be described.
Zebrafish to study brain development and function
Prof. Robert Hindges, King's College London
One of the major challenges in neuroscience is investigate the development and function of the brain in vivo. This includes molecular and cellular mechanisms underlaying neural connectivity, synapse formation, function and plasticity, as well as behaviour. Zebrafish (Danio rerio) has become a prominent model in basic and translational neuroscience to study brain function and dysfunction. Its high genetic and physiological homology to mammals, rapid development, embryonic/larval optical transparency and genetic accessibility are some of the important advantages of zebrafish. Furthermore, the small size in combination with those features enable high-throughput in vivo genetic and small molecule screens, for example for drug discovery projects or the identification of novel candidate genes. Our research has focused on a number of genes linked with different neurodevelopmental disorders, including bipolar disorder, schizophrenia and autism spectrum disorder. Using genome-edited mutant lines we were able to show specific changes in the structural and functional properties of neural circuits, in particular during synaptic development. We therefore highlight the use of the zebrafish as an ideal model for translational neuroscience.
Genomic Characterization of Posttraumatic Stress Disorder and its Symptom Subdomains in the Million Veteran Program
Prof. Murray Stein, University of California San Diego
Genomewide association analyses in over 250,000 participants of European and African ancestry from the Million Veteran Program were conducted using electronic health record validated PTSD diagnosis and quantitative symptom phenotypes. We identified several genomewide significant loci in case-control analyses, and numerous such loci in quantitative symptom analyses. Genomic structural equation modeling indicated tight coherence of a PTSD symptom factor that partially shares genetic variance (through hyperarousal symptoms) with a distinct mood-anxiety-neuroticism factor. Partitioned heritability indicated enrichment in several cortical and subcortical regions, and transcriptome-wide analyses in these regions were used to identify potential drug repositioning candidates. These results validate the biological coherence of the PTSD syndrome, inform its relationship to comorbid anxiety and depressive disorders, and provide new considerations for treatment.
PTSD, Other Stress Related Disorders and Health Outcomes in the Swedish Registers
Prof. Unnur Valdimarsdóttir, University of Iceland
With unique identification numbers assigned at birth or immigration, the public health care systems (available to all), the nationwide-complete multigenerational-, sociodemographic-, inpatient-, outpatient-, and pharmaceutical registers, the Nordic countries offer unique opportunities for epidemiological studies of more than 27 million individuals with family-based designs. Using these resources in Sweden, we have conducted a series of studies using sibling-designs to explore the association between PTSD and other stress-related disorders and subsequent development of major (somatic) diseases. Accounting for familial factors and comorbidities, our findings suggest that stress-related disorders are associated with increased risks of several incident autoimmune diseases (Song et al., JAMA 2018), cardiovascular disease (Song et al., BMJ 2019), life-threatening infections (Song et al., BMJ 2019) and vascular neurodegenerative disease (Song et al., JAMA Neurology 2020). The findings of these studies (and others in progress) will be presented as well as opportunities for further studies and collaborations, e.g. on the genetics of stress-related disorders based on data in the electronic health registers and newly established cohorts in the Nordic countries.
Genome-wide association study of over 40,000 cases identifies novel loci associated with bipolar disorder
Dr. Niamh Mullins, Mount Sinai New York
The Bipolar Disorder Working Group of the Psychiatric Genomics Consortium present the PGC3 bipolar disorder (BD) genome-wide association study. Samples include 41,917 BD cases and 371,549 controls of European descent from 57 studies. Studies comprise 52 cohorts within the PGC, including 11 genotyped on the PsychChip array, along with biobanks iPSYCH, UK Biobank, deCODE genetics, the Nord-Trondelag Health Study and Estonian Biobank. Meta-analysis identifies 64 independent genomic loci associated with BD (P<5e-8), 34 of which are novel. Across the genome, association signal is significantly enriched in gene sets involved in synaptic functioning, neurogenesis and calcium signaling. Drug class enrichment analyses reveal enrichment in genes encoding the targets of known BD treatments, calcium channel blockers and drugs for functional gastrointestinal disorders. The integration of eQTL data from the PsychENCODE Consortium, provides strong support for 15 genes influencing BD through gene expression in the brain, including FURIN, HTR6, MCHR1. Polygenic risk scores explain approximately 5% of variance in European samples and up to 1.2% and 2.2% of variance in African American and East Asian samples respectively. While the major histocompatibility complex reaches genome-wide significance, imputation of C4 structural variants and gene expression do not support the involvement of C4 in BD.
Large-Scale Consortia efforts for PTSD: progress in addressing critical barriers
Dr. Caroline Nievergelt, University of California San Diego
Posttraumatic stress disorder (PTSD) is a commonly occurring mental health consequence of exposure to extreme, life threatening stress. PTSD is frequently associated with the occurrence of other mental disorders such as major depression and suicide. PGC-PTSD has the goal of conducting genome-wide association studies (GWAS) for symptoms and diagnosis of PTSD to identify robust variants and strong genetic risk scores (PRS) for PTSD.
Our current data freeze has brought together 60 studies, including data from >32K cases and >174K controls. GWAS in subjects of European and African ancestries, as well as sex-stratified analyses, identified 6 sex- and ancestry specific loci and sex-specific heritabilities. The predictive value of the PRS showed significant increase in odds of PTSD, but variance explained is still small.
While these results are promising, we are currently addressing critical barriers to progress, including sample size, diversity in ancestry, heterogeneity within PTSD, as well as the extensive comorbidity of PTSD with other mental disorders. Our large data collection with deep phenotyping has unique potential to improve upon typical case-control association analyses. Such improvements include leveraging granularity in PTSD symptoms, symptom clusters, and trauma exposure, and optimizing analysis strategies to minimize the effects of heterogeneity across individuals and studies.
STOP COVID NYC: Capturing city-wide mental health impacts
Dr. Laura Huckins, Pamela Sklar Division of Psychiatric Genomics, Mount Sinai, New York
The COVID-19 pandemic is causing unprecedented havoc on global health, society and the economy. Healthcare providers are under severe strain. Although initial efforts have rightfully focused on understanding contagion patterns, symptom tracking and risk factors for worsened outcomes, the pandemic also poses an immediate and severe mental health burden. The COVID-19 pandemic is, in itself, a severe, chronic stressor. In addition, social isolation, lack of daily structure, constant physical and emotional stressors, and insecure access to food, support, and healthcare all contribute to predicted increases in substance and alcohol abuse, disordered eating, anxiety, depression, OCD, psychotic experiences, PTSD, self-harm and suicidal ideation/behaviour. There is a critical and urgent need to monitor worsening mental health factors throughout NYC, and to connect at-risk individuals with resources and support wherever possible. We will leverage our phone-based city-wide STOP COVID NYC cohort (~35,000 cases with <1 week of enrollment) to deploy weekly PTSD and mood/anxiety surveys. Our city-wide approach allows us to target specific groups of vulnerable individuals, using our established STOP COVID NYC app, and in collaboration with well-established telehealth counselling and therapy resources.
Pleiotropy and Cross-Disorder Psychiatric Genomics
Prof. Jordan Smoller, Harvard Medical School & Department of Epidemiology, Harvard School of Public Health, Boston
Genetic influences on psychiatric disorders transcend diagnostic boundaries, suggesting substantial pleiotropy of contributing loci. Recent results from the PGC’s Cross-Disorder Workgroup have expanded our understanding of shared genetic effects across psychiatric phenotypes. Genomewide meta-analyses of more than 725,000 cases and controls spanning 8 psychiatric disorders identified three groups of highly genetically-related disorders and detected 109 genomewide significant pleiotropic loci. These included 23 loci with pleiotropic effects on four or more disorders and 11 loci with antagonistic effects on multiple disorders. The pleiotropic loci were localized to genes that, on average, show heightened expression in the brain throughout the lifespan, beginning in the second prenatal trimester and play prominent roles in neurodevelopmental processes. This presentation will review these and related findings as well as their potential implications for psychiatric nosology, drug development, and risk prediction.
Genetic Architecture of 11 Major Psychiatric Disorders at Biobehavioral and Functional Genomic Levels
Andrew Grotzinger, The University of Texas at Austin
Using Genomic Structural Equation Modeling (Genomic SEM; Grotzinger et al., 2019) we formally model genetic risk sharing across 11 major psychiatric disorders to identify four broad classes of psychiatric liability that index risk for compulsive, psychotic, neurodevelopmental and internalizing disorders. We subsequently validate these factors by examining the extent to which their genetic correlations with external biobehavioral traits (e.g., circadian rhythms, cognition, socioeconomic outcomes) conforms to the identified factor structure. Finally, we use Stratified Genomic SEM, a novel method for estimating enrichment of factors in functional categories, to identify genetic enrichment of the psychiatric factors. These results indicate that previously reported results that disorders are enriched in conserved regions may reflect enrichment of pleiotropic variants that are broadly relevant for many disorders. We also find that the intersection of protein-truncating variant (PTV)-intolerant (PI) genes and genes expressed in excitatory and GABAergic neurons are particularly enriched for genetic risk sharing across bipolar disorder and schizophrenia. By interrogating the genetic factors of psychiatric risk at multiple levels of analysis we are able to elucidate the underlying processes that give rise to their manifestation, while also evaluating their conceptual utility.