Advertisement

Convergence of brain transcriptomic and neuroimaging patterns in schizophrenia, bipolar disorder, autism spectrum disorder, and major depressive disorder

  • Author Footnotes
    ∗ Equal contribution
    Dirk Jan Ardesch
    Correspondence
    Corresponding author: Dirk Jan Ardesch,
    Footnotes
    ∗ Equal contribution
    Affiliations
    Department of Complex Trait Genetics, Center for Neurogenomics and Cognitive Research, Vrije Universiteit Amsterdam, Amsterdam Neuroscience, 1081 HV Amsterdam, The Netherlands
    Search for articles by this author
  • Author Footnotes
    ∗ Equal contribution
    Ilan Libedinsky
    Footnotes
    ∗ Equal contribution
    Affiliations
    Department of Complex Trait Genetics, Center for Neurogenomics and Cognitive Research, Vrije Universiteit Amsterdam, Amsterdam Neuroscience, 1081 HV Amsterdam, The Netherlands
    Search for articles by this author
  • Lianne H. Scholtens
    Affiliations
    Department of Complex Trait Genetics, Center for Neurogenomics and Cognitive Research, Vrije Universiteit Amsterdam, Amsterdam Neuroscience, 1081 HV Amsterdam, The Netherlands
    Search for articles by this author
  • Yongbin Wei
    Affiliations
    Department of Complex Trait Genetics, Center for Neurogenomics and Cognitive Research, Vrije Universiteit Amsterdam, Amsterdam Neuroscience, 1081 HV Amsterdam, The Netherlands

    School of Artificial Intelligence, Beijing University of Posts and Telecommunications, Beijing, 100876, China
    Search for articles by this author
  • Martijn P. van den Heuvel
    Affiliations
    Department of Complex Trait Genetics, Center for Neurogenomics and Cognitive Research, Vrije Universiteit Amsterdam, Amsterdam Neuroscience, 1081 HV Amsterdam, The Netherlands

    Department of Child Psychiatry, Amsterdam UMC, Vrije Universiteit Amsterdam, Amsterdam Neuroscience, 1081 HV Amsterdam, The Netherlands
    Search for articles by this author
  • Author Footnotes
    ∗ Equal contribution
Open AccessPublished:January 03, 2023DOI:https://doi.org/10.1016/j.bpsc.2022.12.013

      Abstract

      Background

      Psychiatric conditions show overlap in their symptoms, genetics and their involvement in brain areas and circuits. Structural alterations in the brain have been found to run in parallel with expression profiles of risk genes at the level of the brain transcriptome, which may point towards a potential transdiagnostic vulnerability of the brain for disease processes.

      Methods

      We characterized the transcriptomic vulnerability of the cortex across four major psychiatric disorders based on collated data from psychiatric patients (N = 390) and matched controls (N = 293). We compared normative expression profiles of risk genes linked to schizophrenia, bipolar disorder, autism spectrum disorder, and major depressive disorder to examine cross-disorder overlap in spatial expression profiles across the cortex and its concordance with MRI-derived cross-disorder profile of structural brain alterations.

      Results

      We show high expression of psychiatric risk genes converging on multimodal cortical regions of the limbic, ventral attention, and default mode network vs primary somatosensory networks. Risk genes are found to be enriched among genes associated with the MRI cross-disorder profile, suggestive of a common link between brain anatomy and the transcriptome in psychiatric conditions. Characterization of this cross-disorder structural alteration map further shows enrichment for gene markers of astrocytes and microglia and supragranular cortical layers.

      Conclusions

      Our findings suggest that normative expression profiles of disorder risk genes confer a shared and spatially patterned vulnerability of the cortex across multiple psychiatric conditions. Transdiagnostic overlap in transcriptomic risk suggests a common pathway to brain dysfunction across psychiatric disorders.

      Keywords

      Introduction

      Psychiatric disorders are prevalent, multifactorial conditions accompanied by diverse changes in brain function (
      • Mueser K.T.
      • McGurk S.R.
      Schizophrenia.
      ,
      • Kupfer D.J.
      • Frank E.
      • Phillips M.L.
      Major depressive disorder: new clinical, neurobiological, and treatment perspectives.
      ,
      • Grande I.
      • Berk M.
      • Birmaher B.
      • Vieta E.
      Bipolar disorder.
      ,
      • Lord C.
      • Elsabbagh M.
      • Baird G.
      • Veenstra-Vanderweele J.
      Autism spectrum disorder.
      ). Psychiatric disorders are moderately to highly heritable (
      • Pardiñas A.F.
      • Holmans P.
      • Pocklington A.J.
      • Escott-Price V.
      • Ripke S.
      • Carrera N.
      • et al.
      Common schizophrenia alleles are enriched in mutation-intolerant genes and in regions under strong background selection.
      ,
      • Ruderfer D.M.
      • Ripke S.
      • McQuillin A.
      • Boocock J.
      • Stahl E.A.
      • Pavlides J.M.W.
      • et al.
      Genomic Dissection of Bipolar Disorder and Schizophrenia, Including 28 Subphenotypes.
      ,
      • Grove J.
      • Ripke S.
      • Als T.D.
      • Mattheisen M.
      • Walters R.K.
      • Won H.
      • et al.
      Identification of common genetic risk variants for autism spectrum disorder [no. 3].
      ,
      • Howard D.M.
      • Adams M.J.
      • Clarke T.-K.
      • Hafferty J.D.
      • Gibson J.
      • Shirali M.
      • et al.
      Genome-wide meta-analysis of depression identifies 102 independent variants and highlights the importance of the prefrontal brain regions [no. 3].
      ), suggesting an important genetic component in disease etiology. A major challenge towards understanding psychiatric conditions is therefore determining how genetic and molecular factors impact disorder risk in the brain (
      • Seidlitz J.
      • Nadig A.
      • Liu S.
      • Bethlehem R.A.I.
      • Vértes P.E.
      • Morgan S.E.
      • et al.
      Transcriptomic and cellular decoding of regional brain vulnerability to neurogenetic disorders [no. 1].
      ). Most genetic risk is found in regulatory regions of the genome, suggesting a key role for altered gene expression in the neurobiology of psychiatric disorders (
      • Hernandez L.M.
      • Kim M.
      • Hoftman G.D.
      • Haney J.R.
      • Torre-Ubieta L de la
      • Pasaniuc B.
      • Gandal M.J.
      Transcriptomic Insight Into the Polygenic Mechanisms Underlying Psychiatric Disorders.
      ,
      • Watanabe K.
      • Stringer S.
      • Frei O.
      • Umićević Mirkov M.
      • de Leeuw C.
      • Polderman T.J.C.
      • et al.
      A global overview of pleiotropy and genetic architecture in complex traits [no. 9].
      ).
      Large-scale collaborations have made important progress in identifying psychiatric disease signatures at the level of gene expression in the brain (
      • Akbarian S.
      • Liu C.
      • Knowles J.A.
      • Vaccarino F.M.
      • Farnham P.J.
      • Crawford G.E.
      • et al.
      The PsychENCODE project.
      ). Genes are known to display high variation in expression levels across cortical areas, and disruption of genes related to psychiatric conditions may impact different brain regions depending on the expression level of risk genes. Recent studies integrating transcriptome findings with neuroimaging-derived measures of disease-related brain alterations have reported structural differences to coincide with regional variation in normative (healthy) expression of risk genes across the cortex (
      • Seidlitz J.
      • Nadig A.
      • Liu S.
      • Bethlehem R.A.I.
      • Vértes P.E.
      • Morgan S.E.
      • et al.
      Transcriptomic and cellular decoding of regional brain vulnerability to neurogenetic disorders [no. 1].
      ). For example, normative expression profiles of risk genes are found to correlate with structural alterations - such as changes in connectivity, cortical thickness, and brain morphometry - in a range of psychiatric disorders including, among others, schizophrenia (
      • Romme I.A.C.
      • de Reus M.A.
      • Ophoff R.A.
      • Kahn R.S.
      • van den Heuvel M.P.
      Connectome Disconnectivity and Cortical Gene Expression in Patients With Schizophrenia.
      ,
      • Morgan S.E.
      • Seidlitz J.
      • Whitaker K.J.
      • Romero-Garcia R.
      • Clifton N.E.
      • Scarpazza C.
      • et al.
      Cortical patterning of abnormal morphometric similarity in psychosis is associated with brain expression of schizophrenia-related genes.
      ), autism spectrum disorder (
      • Romero-Garcia R.
      • Warrier V.
      • Bullmore E.T.
      • Baron-Cohen S.
      • Bethlehem R.A.I.
      Synaptic and transcriptionally downregulated genes are associated with cortical thickness differences in autism.
      ), and major depressive disorder (
      • Anderson K.M.
      • Collins M.A.
      • Kong R.
      • Fang K.
      • Li J.
      • He T.
      • et al.
      Convergent molecular, cellular, and cortical neuroimaging signatures of major depressive disorder.
      ,
      • Li J.
      • Seidlitz J.
      • Suckling J.
      • Fan F.
      • Ji G.-J.
      • Meng Y.
      • et al.
      Cortical structural differences in major depressive disorder correlate with cell type-specific transcriptional signatures [no. 1].
      ).
      Genetic, behavioral and neuroimaging studies have highlighted high overlap between psychiatric phenotypes: disorders share symptoms (
      • Keshavan M.S.
      • Morris D.W.
      • Sweeney J.A.
      • Pearlson G.
      • Thaker G.
      • Seidman L.J.
      • et al.
      A dimensional approach to the psychosis spectrum between bipolar disorder and schizophrenia: The Schizo-Bipolar Scale.
      ), exhibit comorbidity (
      • van Loo H.M.
      • Romeijn J.-W.
      Psychiatric comorbidity: fact or artifact?.
      ,
      • Hossain M.M.
      • Khan N.
      • Sultana A.
      • Ma P.
      • McKyer E.L.J.
      • Ahmed H.U.
      • Purohit N.
      Prevalence of comorbid psychiatric disorders among people with autism spectrum disorder: An umbrella review of systematic reviews and meta-analyses.
      ), show shared pathological changes to brain structure and circuits (
      • de Lange S.C.
      • Scholtens L.H.
      • van den Berg L.H.
      • Boks M.P.
      • Bozzali M.
      • Cahn W.
      • et al.
      Shared vulnerability for connectome alterations across psychiatric and neurological brain disorders [no. 9].
      ,
      • Opel N.
      • Goltermann J.
      • Hermesdorf M.
      • Berger K.
      • Baune B.T.
      • Dannlowski U.
      Cross-Disorder Analysis of Brain Structural Abnormalities in Six Major Psychiatric Disorders: A Secondary Analysis of Mega- and Meta-analytical Findings From the ENIGMA Consortium.
      ,
      • Patel Y.
      • Parker N.
      • Shin J.
      • Howard D.
      • French L.
      • Thomopoulos S.
      • et al.
      Virtual Histology of Cortical Thickness and Shared Neurobiology in 6 Psychiatric Disorders.
      ) and can show strong genetic correlation (
      • Watanabe K.
      • Stringer S.
      • Frei O.
      • Umićević Mirkov M.
      • de Leeuw C.
      • Polderman T.J.C.
      • et al.
      A global overview of pleiotropy and genetic architecture in complex traits [no. 9].
      ,
      • Anttila V.
      • Bulik-Sullivan B.
      • Finucane H.K.
      • Walters R.K.
      • Bras Duncan L.
      • et al.
      Analysis of shared heritability in common disorders of the brain.
      ,
      • Lee P.H.
      • Anttila V.
      • Won H.
      • Feng Y.-C.A.
      • Rosenthal J.
      • Zhu Z.
      • et al.
      Genomic Relationships, Novel Loci, and Pleiotropic Mechanisms across Eight Psychiatric Disorders.
      ). Particularly high degrees of overlap have been reported by transcriptomic studies of post-mortem brain tissue that investigated differential gene expression between patients and controls, with similar genes showing altered expression across multiple conditions (
      • Enwright III, J.F.
      • Lewis D.A.
      Similarities in Cortical Transcriptome Alterations Between Schizophrenia and Bipolar Disorder Are Related to the Presence of Psychosis.
      ,
      • Zeighami Y.
      • Bakken T.
      • Nickl-Jockschat T.
      • Miller J.
      • Evans A.C.
      • Lein E.
      • Hawrylycz M.
      Structural and Cellular Transcriptome Foundations of Human Brain Disease.
      ). Furthermore, the degree of shared transcriptional dysregulation is linked to the amount of polygenic overlap between conditions (
      • Gandal M.J.
      • Haney J.R.
      • Parikshak N.N.
      • Leppa V.
      • Ramaswami G.
      • Hartl C.
      • et al.
      Shared molecular neuropathology across major psychiatric disorders parallels polygenic overlap.
      ). This transdiagnostic overlap in transcriptome dysregulation - together with the existence of spatial variation in gene expression across the cortex - suggests a potentially shared disorder vulnerability at the level of the brain transcriptome.
      We here combined and integrated data from differential gene expression studies in psychiatric disorders (
      • Gandal M.J.
      • Haney J.R.
      • Parikshak N.N.
      • Leppa V.
      • Ramaswami G.
      • Hartl C.
      • et al.
      Shared molecular neuropathology across major psychiatric disorders parallels polygenic overlap.
      ) with regional gene expression data of neurotypical individuals from the Allen Human Brain Atlas (
      • Hawrylycz M.J.
      • Lein E.S.
      • Guillozet-Bongaarts A.L.
      • Shen E.H.
      • Ng L.
      • Miller J.A.
      • et al.
      An anatomically comprehensive atlas of the adult human brain transcriptome [no. 7416].
      )to assess the spatial expression profiles of risk genes across four major psychiatric conditions. We study to what extent normative expression of genes differentially expressed in psychiatric disorders converge to a shared spatial profile across the cortex - and whether these brain regions are particularly vulnerable to disease effects across multiple psychiatric conditions. We show that differentially expressed genes exhibit highly similar normative expression patterns and are particularly expressed in brain regions involved in multimodal cognitive networks. Risk genes show enrichment among genes associated with the spatial profile of disorder-shared structural alterations across the cortex, with this structural pattern found to be further enriched for gene markers of astrocytes, microglia, and cortical layers I, II, and V. Our findings indicate a convergent pattern of vulnerability to psychiatric disorders reflected in the brain transcriptome.

      Methods & Materials

      Study approach

      We analyzed cortical profiles of normative expression of differentially expressed genes (DEGs) across four major psychiatric disorders: schizophrenia (SCZ), bipolar disorder (BD), autism spectrum disorder (ASD), and major depressive disorder (MDD). These phenotypes were chosen as they have been extensively characterized using differential gene expression studies (
      • Iwamoto K.
      • Bundo M.
      • Kato T.
      Altered expression of mitochondria-related genes in postmortem brains of patients with bipolar disorder or schizophrenia, as revealed by large-scale DNA microarray analysis.
      ,
      • Garbett K.
      • Ebert P.J.
      • Mitchell A.
      • Lintas C.
      • Manzi B.
      • Mirnics K.
      • Persico A.M.
      Immune transcriptome alterations in the temporal cortex of subjects with autism.
      ,
      • Narayan S.
      • Tang B.
      • Head S.R.
      • Gilmartin T.J.
      • Sutcliffe J.G.
      • Dean B.
      • Thomas E.A.
      Molecular profiles of schizophrenia in the CNS at different stages of illness.
      ,
      • Maycox P.R.
      • Kelly F.
      • Taylor A.
      • Bates S.
      • Reid J.
      • Logendra R.
      • et al.
      Analysis of gene expression in two large schizophrenia cohorts identifies multiple changes associated with nerve terminal function [no. 12].
      ,
      • Voineagu I.
      • Wang X.
      • Johnston P.
      • Lowe J.K.
      • Tian Y.
      • Horvath S.
      • et al.
      Transcriptomic analysis of autistic brain reveals convergent molecular pathology.
      ,
      • Chow M.L.
      • Pramparo T.
      • Winn M.E.
      • Barnes C.C.
      • Li H.-R.
      • Weiss L.
      • et al.
      Age-Dependent Brain Gene Expression and Copy Number Anomalies in Autism Suggest Distinct Pathological Processes at Young Versus Mature Ages.
      ,
      • Chen C.
      • Cheng L.
      • Grennan K.
      • Pibiri F.
      • Zhang C.
      • Badner J.A.
      • et al.
      Two gene co-expression modules differentiate psychotics and controls.
      ,
      • Chang L.-C.
      • Jamain S.
      • Lin C.-W.
      • Rujescu D.
      • Tseng G.C.
      • Sibille E.
      A Conserved BDNF, Glutamate- and GABA-Enriched Gene Module Related to Human Depression Identified by Coexpression Meta-Analysis and DNA Variant Genome-Wide Association Studies.
      ,
      • Reinhart V.
      • Bove S.E.
      • Volfson D.
      • Lewis D.A.
      • Kleiman R.J.
      • Lanz T.A.
      Evaluation of TrkB and BDNF transcripts in prefrontal cortex, hippocampus, and striatum from subjects with schizophrenia, bipolar disorder, and major depressive disorder.
      ). DEG data on neurological disorders including Alzheimer’s disease (AD), Parkinson’s disease (PD), and Huntington’s disease (HD) were included as a non-psychiatric outgroup. We focused on differentially expressed genes as they are widely reported in psychiatric disorders and their expression can be directly measured in the post-mortem brain (
      • Hernandez L.M.
      • Kim M.
      • Hoftman G.D.
      • Haney J.R.
      • Torre-Ubieta L de la
      • Pasaniuc B.
      • Gandal M.J.
      Transcriptomic Insight Into the Polygenic Mechanisms Underlying Psychiatric Disorders.
      ). We analyzed the extent of transdiagnostic overlap in expression patterns among psychiatric disorders, benchmarked these patterns against neuroimaging-derived functional networks and spatial profiles of cross-disorder structural alterations, and explored enrichment among cell type and laminar markers in the brain (see Supplemental Methods). Data on normative and differential gene expression were obtained from public databases and previously published studies as described below.

      Gene expression data

      Differentially expressed genes. DEGs were defined based on differential gene expression (DGE) microarray studies of post-mortem cortical samples. DGE data across four major psychiatric disorders (SCZ, BD, ASD, and MDD) as well as one non-neural phenotype (inflammatory bowel disease, IBD) were included (
      • Gandal M.J.
      • Haney J.R.
      • Parikshak N.N.
      • Leppa V.
      • Ramaswami G.
      • Hartl C.
      • et al.
      Shared molecular neuropathology across major psychiatric disorders parallels polygenic overlap.
      ). This comprehensive study collated and reprocessed microarray data of cortical samples from published studies including subjects with SCZ (N = 159), BD (N = 94), ASD (N = 50), MDD (N = 87), IBD (N = 197), and matched controls (N = 293) (see Table 1 for cortical regions sampled). Quality control and normalization procedures are described in detail in the original paper (
      • Gandal M.J.
      • Haney J.R.
      • Parikshak N.N.
      • Leppa V.
      • Ramaswami G.
      • Hartl C.
      • et al.
      Shared molecular neuropathology across major psychiatric disorders parallels polygenic overlap.
      ). Gene-wise information about log2 fold change (log2FC) and associated P-values were extracted for each disorder. Of all genes that showed significant differential expression compared to healthy controls, on average 87% (range 65-95% depending on the disorder) had normative expression data available in the Allen Human Brain Atlas (AHBA) (
      • Hawrylycz M.J.
      • Lein E.S.
      • Guillozet-Bongaarts A.L.
      • Shen E.H.
      • Ng L.
      • Miller J.A.
      • et al.
      An anatomically comprehensive atlas of the adult human brain transcriptome [no. 7416].
      ) (Table 1).
      Table 1Differential gene expression data for the included psychiatric disorders. Data from meta-analysis of Gandal et al. (
      • Gandal M.J.
      • Haney J.R.
      • Parikshak N.N.
      • Leppa V.
      • Ramaswami G.
      • Hartl C.
      • et al.
      Shared molecular neuropathology across major psychiatric disorders parallels polygenic overlap.
      ), who included microarray experiments from a range of studies (see Original studies). a: Chen et al., (
      • Chen C.
      • Cheng L.
      • Grennan K.
      • Pibiri F.
      • Zhang C.
      • Badner J.A.
      • et al.
      Two gene co-expression modules differentiate psychotics and controls.
      ); b: Maycox et al., (
      • Maycox P.R.
      • Kelly F.
      • Taylor A.
      • Bates S.
      • Reid J.
      • Logendra R.
      • et al.
      Analysis of gene expression in two large schizophrenia cohorts identifies multiple changes associated with nerve terminal function [no. 12].
      ); c: Iwamoto et al., (
      • Iwamoto K.
      • Bundo M.
      • Kato T.
      Altered expression of mitochondria-related genes in postmortem brains of patients with bipolar disorder or schizophrenia, as revealed by large-scale DNA microarray analysis.
      ); d: Narayan et al., (
      • Narayan S.
      • Tang B.
      • Head S.R.
      • Gilmartin T.J.
      • Sutcliffe J.G.
      • Dean B.
      • Thomas E.A.
      Molecular profiles of schizophrenia in the CNS at different stages of illness.
      ); e: Reinhart et al., (
      • Reinhart V.
      • Bove S.E.
      • Volfson D.
      • Lewis D.A.
      • Kleiman R.J.
      • Lanz T.A.
      Evaluation of TrkB and BDNF transcripts in prefrontal cortex, hippocampus, and striatum from subjects with schizophrenia, bipolar disorder, and major depressive disorder.
      ); f: Voineagu et al., (
      • Voineagu I.
      • Wang X.
      • Johnston P.
      • Lowe J.K.
      • Tian Y.
      • Horvath S.
      • et al.
      Transcriptomic analysis of autistic brain reveals convergent molecular pathology.
      ); g: Garbett et al., (
      • Garbett K.
      • Ebert P.J.
      • Mitchell A.
      • Lintas C.
      • Manzi B.
      • Mirnics K.
      • Persico A.M.
      Immune transcriptome alterations in the temporal cortex of subjects with autism.
      ); h: Chow et al., (
      • Chow M.L.
      • Pramparo T.
      • Winn M.E.
      • Barnes C.C.
      • Li H.-R.
      • Weiss L.
      • et al.
      Age-Dependent Brain Gene Expression and Copy Number Anomalies in Autism Suggest Distinct Pathological Processes at Young Versus Mature Ages.
      ); i: Chang et al., (
      • Chang L.-C.
      • Jamain S.
      • Lin C.-W.
      • Rujescu D.
      • Tseng G.C.
      • Sibille E.
      A Conserved BDNF, Glutamate- and GABA-Enriched Gene Module Related to Human Depression Identified by Coexpression Meta-Analysis and DNA Variant Genome-Wide Association Studies.
      ).
      DisorderN significant genesN corresponding AHBA genes (%)Regions sampledOriginal studies
      Schizophrenia20431920 (94%)BA10, BA46, parietal cortexa, b, c, d, e
      Bipolar disorder518492 (95%)BA46, parietal cortexa, c, e
      Autism spectrum disorder24712284 (92%)BA9, BA41, BA42, BA46f, g, h
      Major depressive disorder328213 (65%)BA9, BA25, BA46e, i
      Disease specificity. DGE microarray data of three neurological disorders including AD, PD, and HD were further included as a non-psychiatric outgroup. Gene expression data for AD were included from region BA22 in 55 patients and 22 controls (
      • Horesh Y.
      • Katsel P.
      • Haroutunian V.
      • Domany E.
      Gene expression signature is shared by patients with Alzheimer’s disease and schizophrenia at the superior temporal gyrus.
      ), data for PD were included from cortical samples in 16 patients and 20 controls (
      • Riley B.E.
      • Gardai S.J.
      • Emig-Agius D.
      • Bessarabova M.
      • Ivliev A.E.
      • Schüle B.
      • et al.
      Systems-Based Analyses of Brain Regions Functionally Impacted in Parkinson’s Disease Reveals Underlying Causal Mechanisms.
      ), and data for HD were included from region BA4 and the caudate nucleus in 37 patients and 29 controls (
      • Hodges A.
      • Strand A.D.
      • Aragaki A.K.
      • Kuhn A.
      • Sengstag T.
      • Hughes G.
      • et al.
      Regional and cellular gene expression changes in human Huntington’s disease brain.
      ). We further assessed disease specificity by including a list of non-specific genes found to be differentially expressed across > 600 DGE studies of the brain and other tissues (
      • Crow M.
      • Lim N.
      • Ballouz S.
      • Pavlidis P.
      • Gillis J.
      Predictability of human differential gene expression.
      ), as disease-related DGE might reflect both disease-specific effects and general, non-specific differences in gene expression. This list of non-specific genes (referred to as a differential expression ‘prior’ or DEPrior) includes 19,172 genes and lists for each gene the probability of being differentially expressed in any particular microarray study, capturing non-specific hits in DGE studies (
      • Crow M.
      • Lim N.
      • Ballouz S.
      • Pavlidis P.
      • Gillis J.
      Predictability of human differential gene expression.
      ).
      Normative gene expression. Spatial gene expression patterns across the brain were examined by means of the normative gene expression data from the AHBA (
      • Hawrylycz M.J.
      • Lein E.S.
      • Guillozet-Bongaarts A.L.
      • Shen E.H.
      • Ng L.
      • Miller J.A.
      • et al.
      An anatomically comprehensive atlas of the adult human brain transcriptome [no. 7416].
      ), describing a rich collection of microarray studies across the brain from six postmortem brains of neurotypical donors (http://help.brain-map.org/display/humanbrain/Documentation). Data were preprocessed following procedures introduced previously (
      • Wei Y.
      • de Lange S.C.
      • Pijnenburg R.
      • Scholtens L.H.
      • Ardesch D.J.
      • Watanabe K.
      • et al.
      Statistical testing in transcriptomic-neuroimaging studies: A how-to and evaluation of methods assessing spatial and gene specificity.
      ), yielding a 57 x 20,949 (brain regions x genes) matrix (Supplemental Methods).
      Normative expression of disorder-related DEGs. For each DGE phenotype, genes were included that had available expression data in the AHBA and that were reported to be significantly differentially expressed in the original dataset. The top 100 most strongly differentially expressed genes (DEGs) were extracted based on absolute log2FC as the gene set of interest (for HD the t-statistic was used as log2FC was not available) (Figure 1A). The top 50 and top 200 most strongly differentially expressed genes were additionally taken as a verification analysis (see Supplemental Information). A spatial expression pattern across the cortex of each gene set of interest was obtained by averaging the normative expression data in the AHBA of the genes of interest (Figure 1B). These spatial expression patterns were then correlated with each other to assess overlap across disorders (Figure 1C), including null models of spatial and gene-specificity (see Statistical analysis).
      Figure thumbnail gr1
      Figure 1Study approach. A) Differentially expressed genes were obtained from literature, sorted based on differential expression, and the top 100 differentially expressed genes were extracted as gene set of interest for each phenotype. B) Normative gene expression profiles across the cortex were obtained from the Allen Human Brain Atlas (AHBA) and a mean expression pattern was calculated across the genes in each gene set of interest. C) Mean normative expression profiles were compared across disorders to assess spatial overlap.

      Neuroimaging data

      Resting-state functional networks. Resting-state functional networks were included from the Yeo-7 atlas, describing seven distinct functional resting-state networks including the visual, somatomotor, dorsal attention, ventral attention/salience, limbic, frontoparietal/central executive, and default mode network (
      • Yeo B.T.T.
      • Krienen F.M.
      • Sepulcre J.
      • Sabuncu M.R.
      • Lashkari D.
      • Hollinshead M.
      • et al.
      The organization of the human cerebral cortex estimated by intrinsic functional connectivity.
      ). The functional network atlas was mapped to the DK114 atlas by a majority vote procedure: for each cortical region of the DK114 atlas, the number of vertices belonging to every resting-state network was counted and the resting-state network with the highest number of vertices was assigned to that DK114 region (
      • Wei Y.
      • Lange S.C.
      • Scholtens L.H.
      • Watanabe K.
      • Ardesch D.J.
      • Jansen P.R.
      • et al.
      Genetic mapping and evolutionary analysis of human-expanded cognitive networks.
      ).
      Cross-disorder involvement map. Voxel-based morphometry (VBM) data were obtained from BrainMap (http://www.brainmap.org), an extensive database including data from neuroimaging experiments of a range of brain disorders. MRI studies assessing cortical volumetric changes were retrieved for SCZ, BD, ASD, and MDD, and a meta-analysis was performed on all experiments within each disorder using activation likelihood estimation (ALE). The cross-disorder involvement map was computed by averaging ALEs of all voxels within the cortical regions (for more information see Supplemental Methods).

      Statistical analysis

      Expression patterns were tested for gene-specificity by normalizing the expression level at each cortical region against a null distribution of expression levels derived from equal-sized sets of random genes. Normalized expression patterns were calculated by taking the average gene expression profile across the cortex of the gene set of interest, subtracting the average expression profile of 1,000 randomly selected set of genes of equal size, and dividing by the standard deviation of the 1,000 random gene sets, resulting in an expression z-score. All genes present in the AHBA were considered in the main analysis to include normative expression data of as many DEGs as possible; analyses using more specific sets of background genes (i.e., genes expressed in the brain and genes enriched in the brain (
      • Wei Y.
      • de Lange S.C.
      • Pijnenburg R.
      • Scholtens L.H.
      • Ardesch D.J.
      • Watanabe K.
      • et al.
      Statistical testing in transcriptomic-neuroimaging studies: A how-to and evaluation of methods assessing spatial and gene specificity.
      )) are additionally described in the Supplemental Information. Spatial autocorrelation is a prevalent feature of brain patterns, including gene expression patterns, and we therefore further evaluated transcriptomic associations using null-spin models that test for spatial specificity (
      • Alexander-Bloch A.F.
      • Shou H.
      • Liu S.
      • Satterthwaite T.D.
      • Glahn D.C.
      • Shinohara R.T.
      • et al.
      On testing for spatial correspondence between maps of human brain structure and function.
      ,
      • Markello R.D.
      • Misic B.
      Comparing spatial null models for brain maps.
      ). Gene set enrichment analysis (
      • Subramanian A.
      • Tamayo P.
      • Mootha V.K.
      • Mukherjee S.
      • Ebert B.L.
      • Gillette M.A.
      • et al.
      Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles.
      ) was performed using WebGestalt (
      • Zhang B.
      • Kirov S.
      • Snoddy J.
      WebGestalt: an integrated system for exploring gene sets in various biological contexts.
      ,
      • Liao Y.
      • Wang J.
      • Jaehnig E.J.
      • Shi Z.
      • Zhang B.
      WebGestalt 2019: gene set analysis toolkit with revamped UIs and APIs.
      ) (webgestalt.org) to test whether genes correlating with the cross-disorder involvement map were enriched for gene sets of interest, including the disorder risk genes. Normalized enrichment scores (NES) produced by WebGestalt were used as a measure of enrichment corrected for gene set size, with positive scores indicating enrichment and negative scores indicating depletion (
      • Liao Y.
      • Wang J.
      • Jaehnig E.J.
      • Shi Z.
      • Zhang B.
      WebGestalt 2019: gene set analysis toolkit with revamped UIs and APIs.
      ). Partial least squares (PLS) regression was used to obtain PLS weights for each gene in the AHBA as a measure of their association with the cross-disorder involvement map (
      • Krishnan A.
      • Williams L.J.
      • McIntosh A.R.
      • Abdi H.
      Partial Least Squares (PLS) methods for neuroimaging: A tutorial and review.
      ). Weights corresponding to the first PLS component (N = 20,949 genes) were used as a reference for the gene set enrichment analysis, testing whether genes with similar spatial expression pattern to the cross-disorder involvement map were enriched for the gene sets of interest. Multiple testing correction was carried out where appropriate using the Benjamini & Hochberg false discovery rate (FDR) procedure (
      • Benjamini Y.
      • Hochberg Y.
      Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing.
      ) at q = 0.05.

      Results

      Psychiatric disorder genes show convergent cortical expression profiles

      We started by investigating the transdiagnostic overlap in differentially expressed genes (DEGs) across four major psychiatric disorders including schizophrenia (SCZ), bipolar disorder (BD), autism spectrum disorder (ASD), and major depressive disorder (MDD). The top 100 DEGs (Figure 2A, Table S1) showed high overlap among disorders: of the 331 unique genes included in the top 100 DEGs for all four disorders, 62 genes were shared among two or more disorders and 6 genes among three or more disorders (Figure 2B; removing these shared genes from the analysis showed consistent findings, see Supplemental Results). The degree of overlap was the highest between SCZ and BD (46 shared genes), followed by SCZ and ASD (17 shared genes), both significantly exceeding the expected number of shared genes between randomly selected sets of genes (both P < 0.001, 1,000 permutations). The degree of shared genes between SCZ-MDD, BD-ASD, BD-MDD, and ASD-MDD were not significant (P > 0.05).
      Figure thumbnail gr2
      Figure 2Normative expression profiles of differentially expressed genes (DEGs) in four major psychiatric disorders. A) Volcano plots of differential gene expression in schizophrenia (SCZ), bipolar disorder (BD), autism spectrum disorder (ASD), and major depressive disorder (MDD). Colored dots represent the top 100 most differentially expressed genes included in the gene set of interest for each disorder; the dotted line represents the significance threshold (P < 0.05, FDR-corrected). B) Venn diagram of the number of DEGs shared among disorders. C) Normative expression patterns of the top 100 DEGs for SCZ, BD, ASD, and MDD (expression normalized to mean expression profile of randomized gene sets, 1,000 permutations). D) Correlation between cortical expression profile of each disorder (horizontal axis) with each of the other disorders (colored dots). Null distributions of correlation coefficients of each disorder with 1,000 randomized gene sets are plotted in gray.
      Normative expression profiles of the top 100 DEGs of each disorder, reflecting to what extent these genes are expressed across the cortical mantle of the neurotypical brain, showed high expression in superior frontal regions, temporal regions (temporal pole, entorhinal cortex), and the anterior cingulate cortex compared with randomized sets of genes (Figure 2C, Table S2). Relatively low expression was observed in lateral occipital, superior parietal, and postcentral regions (Table S2). These expression patterns exhibited a high level of overlap across disorders, with the temporal pole and entorhinal cortex showing high expression in all four psychiatric disorders (two-tailed z-test, z-score > 2.3, P < 0.05, FDR-corrected), and the lateral occipital cortex and postcentral gyrus showing relatively low normative expression of risk genes across the four disorders (z-score < -2.9, P < 0.05, FDR-corrected).
      Spatial overlap in expression profiles may reflect specific contributions of shared underlying biological mechanisms or may reflect non-specific transcriptional gradients across the cortex. We benchmarked the spatial correlations between each disorder pair against different null models to assess the gene set spatial specificity of the observed cross-disorder overlap. Spatial correlations of SCZ-BD, SCZ-ASD, BD-ASD, and ASD-MDD exceeded null distributions of correlations between the individual disorders and randomized gene sets (P < 0.05, FDR-corrected, 1,000 permutations per disorder), indeed suggesting a specific spatial correlation in gene expression patterns of these disorders (Figure 2D). Correlations of MDD with SCZ and BD were not found to exceed this null condition (P > 0.05, FDR-corrected, 1,000 permutations per disorder). Further testing for spatial specificity of the expression profiles using the null-spin model showed the spatial correlations to exceed the spinned null distribution for each disorder pair (P < 0.001, FDR-corrected, 1,000 permutations).
      We further characterized whether the cortical expression profiles were potentially specific to the included psychiatric disorders or whether they reflected a more general vulnerability to brain disorders. The analysis was therefore extended with the normative expression patterns of the top 100 DEGs from three neurological disorders: Alzheimer’s disease (AD), Huntington’s disease (HD), and Parkinson’s disease (PD) (Figure 3A). Spatial correlations between normative expression patterns (Figure 3B) were found to be higher among psychiatric disorders (M = 0.68, SD = 0.17) than between psychiatric disorders and neurological diseases (M = 0.21, SD = 0.18), t(
      • Anderson K.M.
      • Collins M.A.
      • Kong R.
      • Fang K.
      • Li J.
      • He T.
      • et al.
      Convergent molecular, cellular, and cortical neuroimaging signatures of major depressive disorder.
      ) = 5.20, P = 8.78 × 10-5, suggesting a degree of specificity of the observed transcriptomic overlap for psychiatric disorders (Figure 3C).
      Figure thumbnail gr3
      Figure 3Comparison with normative expression profiles of neurological diseases. A) Normative expression profiles of the top 100 DEGs for Alzheimer’s disease (AD), Huntington’s disease (HD), and Parkinson’s disease (PD) (expression normalized to mean expression profile of randomized gene sets, 1,000 permutations). B) Spatial correlations across cortical regions of psychiatric and neurological conditions. C) Specificity of the overlap in normative expression in psychiatric conditions. Spatial correlations among psychiatric disorders (black circles) are higher than correlations between psychiatric and neurological diseases (gray diamonds).

      Risk genes display high expression in multimodal cognitive networks

      We continued by asking whether psychiatric disorder genes are highly expressed in cortical regions that support multimodal cognitive functions, which are thought to be particularly affected in psychiatric disorders (
      • Cole M.W.
      • Repovš G.
      • Anticevic A.
      The Frontoparietal Control System: A Central Role in Mental Health.
      ). With normative expression values parcellated into seven canonical resting-state functional networks (Figure 4A), disorder risk genes indeed showed higher expression in cortical regions that support multimodal cognitive networks compared with regions that are part of primary somatosensory networks (Figure 4B). Expression was significantly higher in the limbic network than in the primary networks for risk genes in SCZ, BD, ASD, and MDD (P < 0.05, FDR-corrected, see Table 2). The ventral attention network also displayed higher risk gene expression than the primary networks across the four disorders (P < 0.05, Table 2). The default mode network additionally showed higher expression than the visual and somatomotor networks in SCZ, BD, and ASD, but not MDD (Table 2, all two-tailed and FDR-corrected across functional networks and disorders).
      Figure thumbnail gr4
      Figure 4Expression of disorder risk genes along resting-state functional networks. A) Parcellation of 7 functional networks derived from resting-state functional MRI (
      • Yeo B.T.T.
      • Krienen F.M.
      • Sepulcre J.
      • Sabuncu M.R.
      • Lashkari D.
      • Hollinshead M.
      • et al.
      The organization of the human cerebral cortex estimated by intrinsic functional connectivity.
      ) onto the DK114 atlas. C) Normative expression of disorder risk genes for SCZ, BD, ASD, and MDD in cortical regions belonging to each of the 7 functional networks (expression normalized to randomized gene sets, 1,000 permutations). Asterisks denote networks that show significantly higher expression than the visual network, plus signs denote networks that show significantly higher expression than the somatomotor network (FDR-corrected across networks and disorders).
      Table 2Comparison of risk gene expression in resting-state functional networks that support multimodal processes (dorsal attention, ventral attention, limbic, frontoparietal, default mode, rows) vs networks involved in primary sensory processes (visual and somatomotor, columns). Statistics represent t-tests between expression levels in the corresponding cortical areas of the resting-state functional networks (two-tailed, FDR-corrected across networks and disorders).
      VisualSomatomotor
      SCZtdfPtdfP
      Dorsal attention5.2101.6 × 10-31.6120.16
      Ventral attention4.8131.6 × 10-32.6150.035
      Limbic10.7131.6 × 10-65.8151.8 × 10-4
      Frontoparietal4.992.8 × 10-31.8110.14
      Default mode6.0242.3 × 10-52.8260.019
      BDtdfPtdfP
      Dorsal attention4.5103.5 × 10-31.7120.15
      Ventral attention4.5132.2 × 10-32.7150.030
      Limbic10.8131.6 × 10-67.3152.3 × 10-5
      Frontoparietal4.395.6 × 10-31.9110.12
      Default mode6.0242.3 × 10-53.3266.8 × 10-3
      ASDtdfPtdfP
      Dorsal attention2.1100.096-1.0120.40
      Ventral attention7.1134.4 × 10-52.0150.094
      Limbic9.2135.9 × 10-62.1150.081
      Frontoparietal2.490.069-0.5110.68
      Default mode3.5245.6 × 10-3-0.2260.85
      MDDtdfPtdfP
      Dorsal attention0.9100.431.0120.40
      Ventral attention3.0130.0223.1150.016
      Limbic2.7130.0312.9150.021
      Frontoparietal0.890.460.9110.43
      Default mode1.5240.161.7260.14

      Normative expression reflects cross-disorder cortical involvement

      We next examined whether the cortical expression patterns of disorder DEGs may mirror a general vulnerability of the brain to psychiatric disorders. Cortical disorder involvement maps derived from voxel-based morphometry studies from the BrainMap database were combined across SCZ, BD, ASD, and MDD into an average psychiatric cross-disorder involvement map, reflecting the extent of shared involvement across psychiatric disorders (Figure 5A).
      Figure thumbnail gr5
      Figure 5Associations between gene expression profiles and cross-disorder brain alterations. A) Individual disorder involvement maps based on voxel-based morphometry (VBM) meta-analysis from the BrainMap database were combined into a single cross-disorder involvement map representing structural brain alterations in SCZ, BD, ASD, and MDD. Positive z-scores indicates higher regional disease involvement. B) Genes correlated with the cross-disorder involvement map show enrichment for genes differentially expressed in SCZ, BD, and ASD, but not for non-psychiatric phenotypes. C) Genes correlated with the cross-disorder involvement map show cell type enrichment for marker genes of astrocytes and microglia. D) Genes correlated with the cross-disorder involvement map show laminar enrichment for marker genes of cortical L1, L2, and L5, and depletion for marker genes of L3 and L4. Abbreviations: SCZ, schizophrenia; BD, bipolar disorder; ASD, autism spectrum disorder; MDD, major depressive disorder; PD, Parkinson’s disease; IBD, inflammatory bowel disease; DEPrior, differential expression prior; AD, Alzheimer’s disease; HD, Huntington’s disease; NES, normalized enrichment score; RG: radial glial cells, IPC: intermediate progenitor cells, n.s., not significant.
      Gene set enrichment analysis showed genes correlating with the cross-disorder involvement map (Table S3) to be enriched for the top DEGs for SCZ (normalized enrichment score (NES) = 1.98, P < 0.001), BD (NES = 1.77, P = 5.2 × 10-3), and ASD (NES = 1.61, P = 0.02), but not for MDD (NES = 0.67, P = 0.95, FDR-corrected) (Figure 5B). This suggested potential specificity for psychiatric disorders, with no enrichment found for DEGs of neurological diseases including AD, HD, and PD, nor for DEGs of the non-brain phenotype inflammatory bowel disease (IBD), nor for a set of genes commonly found to be differentially expressed in the literature (DEPrior) (all P > 0.4, FDR-corrected, Figure 5B). Further testing the gene-specificity of these results showed that the observed enrichment of SCZ, BD, and ASD genes in the cross-disorder involvement pattern also exceeded a null distribution of enrichment of equal-sized, randomized sets of genes (SCZ P < 0.001, BD P < 0.001, ASD P = 0.009, FDR-corrected, 1,000 permutations). For sensitivity analyses see Supplemental Results.
      We further characterized the neuroimaging-based cross-disorder involvement map by exploring the enrichment of correlated genes among genetic markers of brain cell types and cortical layers (see Supplemental Methods for details). Genes correlated with the cross-disorder map showed enrichment for gene markers of astrocytes (NES = 2.07, P < 0.001) and microglia (NES = 1.58, P = 9.2 × 10-3, FDR-corrected) (Figure 5C). We furthermore observed enrichment for markers of cortical layer 1, (NES = 2.18, P < 0.001), layer 2 (NES = 1.69, P < 0.001), and layer 5 (NES = 1.38, P = 0.035), and a depletion for markers of layer 3 (NES = -2.26, P < 0.001) and layer 4 (NES = -2.44, P < 0.001, FDR-corrected) in the cross-disorder map (Figure 5D).

      Discussion

      We show that normative expression profiles of risk genes for four major psychiatric disorders (SCZ, BD, ASD, and MDD) converge on the same cortical regions. The observed spatial overlap suggests a shared vulnerability of the cortex to transcriptional dysregulation across psychiatric disorders. Expression of shared disorder risk genes is particularly high in cortical regions that support integrative brain networks such as the limbic, ventral attention, and default mode network (
      • Yeo B.T.T.
      • Krienen F.M.
      • Sepulcre J.
      • Sabuncu M.R.
      • Lashkari D.
      • Hollinshead M.
      • et al.
      The organization of the human cerebral cortex estimated by intrinsic functional connectivity.
      ). We find that psychiatric risk genes are enriched among genes whose expression profiles coincide with a MRI-based disorder involvement map, suggesting a spatial convergence of both transcriptomic risk and neuroimaging-derived cortical volumetric alterations across psychiatric disorders.
      The observation of a transdiagnostic spatial expression profile of psychiatric risk genes fits with the notion that psychiatric disorders share important disease mechanisms at the genetic level (
      • Anttila V.
      • Bulik-Sullivan B.
      • Finucane H.K.
      • Walters R.K.
      • Bras Duncan L.
      • et al.
      Analysis of shared heritability in common disorders of the brain.
      ,
      • Lee P.H.
      • Anttila V.
      • Won H.
      • Feng Y.-C.A.
      • Rosenthal J.
      • Zhu Z.
      • et al.
      Genomic Relationships, Novel Loci, and Pleiotropic Mechanisms across Eight Psychiatric Disorders.
      ).Much of the genetic risk resides in non-coding and regulatory loci, stressing the relevance of gene dysregulation in psychiatric disorders (
      • Watanabe K.
      • Stringer S.
      • Frei O.
      • Umićević Mirkov M.
      • de Leeuw C.
      • Polderman T.J.C.
      • et al.
      A global overview of pleiotropy and genetic architecture in complex traits [no. 9].
      ,
      • Maurano M.T.
      • Humbert R.
      • Rynes E.
      • Thurman R.E.
      • Haugen E.
      • Wang H.
      • et al.
      Systematic Localization of Common Disease-Associated Variation in Regulatory DNA.
      ).Our findings suggest that expression profiles of risk genes mirror disorder-associated alterations across the cortex. Brain regions with high baseline expression of risk genes might therefore be particularly vulnerable to transcriptomic dysregulation in psychiatric conditions.
      Our cross-disorder analysis suggests that disorder risk genes are particularly expressed in multimodal cortical regions that are part of integrative cognitive networks such as the default mode, limbic, and ventral attention networks (
      • Yeo B.T.T.
      • Krienen F.M.
      • Sepulcre J.
      • Sabuncu M.R.
      • Lashkari D.
      • Hollinshead M.
      • et al.
      The organization of the human cerebral cortex estimated by intrinsic functional connectivity.
      ). Pathological alterations in these regions may have profound effects on brain function due to the central roles of multimodal areas and networks in the brain (
      • Mesulam M.
      From sensation to cognition.
      ,
      • van den Heuvel M.P.
      • Sporns O.
      Network hubs in the human brain.
      ). Transcriptomic dysregulation may particularly impact integrative networks such as the default mode network due to their dependence on highly connected network hubs that show high potential for brain plasticity (
      • Benowitz L.I.
      • Routtenberg A.
      GAP-43: an intrinsic determinant of neuronal development and plasticity.
      ,
      • García-Cabezas M.Á.
      • Zikopoulos B.
      • Barbas H.
      The Structural Model: a theory linking connections, plasticity, pathology, development and evolution of the cerebral cortex.
      ) and fast metabolism (
      • Vaishnavi S.N.
      • Vlassenko A.G.
      • Rundle M.M.
      • Snyder A.Z.
      • Mintun M.A.
      • Raichle M.E.
      Regional aerobic glycolysis in the human brain.
      ,
      • Fulcher B.D.
      • Fornito A.
      A transcriptional signature of hub connectivity in the mouse connectome.
      ), which is thought to render them vulnerable to pathological disturbances (
      • Cole M.W.
      • Repovš G.
      • Anticevic A.
      The Frontoparietal Control System: A Central Role in Mental Health.
      ,
      • Crossley N.A.
      • Mechelli A.
      • Scott J.
      • Carletti F.
      • Fox P.T.
      • McGuire P.
      • Bullmore E.T.
      The hubs of the human connectome are generally implicated in the anatomy of brain disorders.
      ,
      • van den Heuvel M.P.
      • Sporns O.
      A cross-disorder connectome landscape of brain dysconnectivity.
      ). Indeed, the default mode network, and regions of the ventral attention (
      • Goodkind M.
      • Eickhoff S.B.
      • Oathes D.J.
      • Jiang Y.
      • Chang A.
      • Jones-Hagata L.B.
      • et al.
      Identification of a Common Neurobiological Substrate for Mental Illness.
      ) and limbic networks (
      • Park B.
      • Kebets V.
      • Lariviere S.
      • Hettwer M.D.
      • Paquola C.
      • van Rooij D.
      • et al.
      Multilevel Neural Gradients Reflect Transdiagnostic Effects of Major Psychiatric Conditions on Cortical Morphology.
      ) are commonly implicated in the pathology and genetics of psychiatric disorders (
      • Wei Y.
      • Lange S.C.
      • Scholtens L.H.
      • Watanabe K.
      • Ardesch D.J.
      • Jansen P.R.
      • et al.
      Genetic mapping and evolutionary analysis of human-expanded cognitive networks.
      ,
      • Meda S.A.
      • Ruaño G.
      • Windemuth A.
      • O’Neil K.
      • Berwise C.
      • Dunn S.M.
      • et al.
      Multivariate analysis reveals genetic associations of the resting default mode network in psychotic bipolar disorder and schizophrenia.
      ,
      • Padmanabhan A.
      • Lynch C.J.
      • Schaer M.
      • Menon V.
      The Default Mode Network in Autism.
      ,
      • Doucet G.E.
      • Janiri D.
      • Howard R.
      • O’Brien M.
      • Andrews-Hanna J.R.
      • Frangou S.
      Transdiagnostic and disease-specific abnormalities in the default-mode network hubs in psychiatric disorders: A meta-analysis of resting-state functional imaging studies.
      ).
      Enrichment of genetic markers of astrocytes and microglia among genes associated with the cross-disorder involvement map suggests that these cell types play an important role in the cortical vulnerability to psychiatric disorders. Indeed, astrocyte-related genes are broadly upregulated in SCZ, BD, and ASD, with microglia-related genes additionally upregulated in ASD (
      • Gandal M.J.
      • Haney J.R.
      • Parikshak N.N.
      • Leppa V.
      • Ramaswami G.
      • Hartl C.
      • et al.
      Shared molecular neuropathology across major psychiatric disorders parallels polygenic overlap.
      ,
      • Parikshak N.N.
      • Swarup V.
      • Belgard T.G.
      • Irimia M.
      • Ramaswami G.
      • Gandal M.J.
      • et al.
      Genome-wide changes in lncRNA, splicing, and regional gene expression patterns in autism.
      ). Astrocyte- and microglia-specific expression profiles have also previously been found to coincide with brain morphological alterations in depression (
      • Anderson K.M.
      • Collins M.A.
      • Kong R.
      • Fang K.
      • Li J.
      • He T.
      • et al.
      Convergent molecular, cellular, and cortical neuroimaging signatures of major depressive disorder.
      ,
      • Li J.
      • Seidlitz J.
      • Suckling J.
      • Fan F.
      • Ji G.-J.
      • Meng Y.
      • et al.
      Cortical structural differences in major depressive disorder correlate with cell type-specific transcriptional signatures [no. 1].
      ), and cortical thickness differences in a range of psychiatric phenotypes (
      • Patel Y.
      • Parker N.
      • Shin J.
      • Howard D.
      • French L.
      • Thomopoulos S.
      • et al.
      Virtual Histology of Cortical Thickness and Shared Neurobiology in 6 Psychiatric Disorders.
      ). Astrocytes and microglia are thought to be critical for, among others, regulating synaptic pruning in neurodevelopment and synaptic plasticity in the adult brain, suggesting a common role for these processes in the etiology of psychiatric phenotypes (
      • Salter M.W.
      • Stevens B.
      Microglia emerge as central players in brain disease.
      ,
      • Vainchtein I.D.
      • Molofsky A.V.
      Astrocytes and Microglia: In Sickness and in Health.
      ).
      Laminar enrichment of genes associated with the cross-disorder involvement map indicates a potential relevance of cortical layers I, II, and V in the shared neurobiology of psychiatric disorders. There is indeed evidence of abnormal laminar organization and thinning of supragranular layers in SCZ (
      • Fatemi S.H.
      • Folsom T.D.
      The Neurodevelopmental Hypothesis of Schizophrenia, Revisited.
      ,
      • Wagstyl K.
      • Ronan L.
      • Whitaker K.J.
      • Goodyer I.M.
      • Roberts N.
      • Crow T.J.
      • Fletcher P.C.
      Multiple markers of cortical morphology reveal evidence of supragranular thinning in schizophrenia.
      ),and reports of enriched heritability for SCZ in excitatory and inhibitory neurons of layer V (
      • Hook P.W.
      • McCallion A.S.
      Leveraging mouse chromatin data for heritability enrichment informs common disease architecture and reveals cortical layer contributions to schizophrenia.
      ). Other studies however show altered morphology and decreased spine density of pyramidal cells in layer III (
      • Pierri J.N.
      • Volk C.L.E.
      • Auh S.
      • Sampson A.
      • Lewis D.A.
      Decreased Somal Size of Deep Layer 3 Pyramidal Neurons in the Prefrontal Cortex of Subjects With Schizophrenia.
      ,
      • Kolluri N.
      • Sun Z.
      • Sampson A.R.
      • Lewis D.A.
      Lamina-Specific Reductions in Dendritic Spine Density in the Prefrontal Cortex of Subjects With Schizophrenia.
      ) which correlates with macroscale connectivity disruptions in SCZ (
      • van den Heuvel M.P.
      • Scholtens L.H.
      • Reus MA de
      • Kahn R.S.
      Associated Microscale Spine Density and Macroscale Connectivity Disruptions in Schizophrenia.
      ), but not in subgranular layers V and VI. We note that our findings of enrichment are based on genes whose expression patterns correlate with a neuroimaging-based cross-disorder involvement map, which is informative of shared aspects of brain alterations in psychiatric disorders but does not necessarily capture the specific disease mechanism of each disorder included in the map. Disorder-specific investigations linking brain alterations at the macroscale to transcriptomic and genetic disruptions at the microscale will therefore be crucial to support conclusive claims about the involvement and specificity of each layer in different psychiatric disorders (
      • Scholtens L.H.
      • van den Heuvel M.P.
      Multimodal Connectomics in Psychiatry: Bridging Scales From Micro to Macro.
      ).
      Methodological considerations need to be taken into account when interpreting the findings of this study. The AHBA is one of the most comprehensive and spatially detailed gene expression databases available today, but it is limited to postmortem microarray experiments of six donor brains (
      • Hawrylycz M.J.
      • Lein E.S.
      • Guillozet-Bongaarts A.L.
      • Shen E.H.
      • Ng L.
      • Miller J.A.
      • et al.
      An anatomically comprehensive atlas of the adult human brain transcriptome [no. 7416].
      ). The AHBA is not fully suited to disentangling potentially relevant contributions of temporal dynamics and individual variation to brain transcription patterns (
      • Spielman R.S.
      • Bastone L.A.
      • Burdick J.T.
      • Morley M.
      • Ewens W.J.
      • Cheung V.G.
      Common genetic variants account for differences in gene expression among ethnic groups.
      ,
      • Berchtold N.C.
      • Cribbs D.H.
      • Coleman P.D.
      • Rogers J.
      • Head E.
      • Kim R.
      • et al.
      Gene expression changes in the course of normal brain aging are sexually dimorphic.
      ,
      • Kumar A.
      • Gibbs J.R.
      • Beilina A.
      • Dillman A.
      • Kumaran R.
      • Trabzuni D.
      • et al.
      Age-associated changes in gene expression in human brain and isolated neurons.
      ). In addition, the differential gene expression data used here are typically derived from one or a few brain regions (
      • Gandal M.J.
      • Haney J.R.
      • Parikshak N.N.
      • Leppa V.
      • Ramaswami G.
      • Hartl C.
      • et al.
      Shared molecular neuropathology across major psychiatric disorders parallels polygenic overlap.
      ,
      • Iwamoto K.
      • Bundo M.
      • Kato T.
      Altered expression of mitochondria-related genes in postmortem brains of patients with bipolar disorder or schizophrenia, as revealed by large-scale DNA microarray analysis.
      ,
      • Maycox P.R.
      • Kelly F.
      • Taylor A.
      • Bates S.
      • Reid J.
      • Logendra R.
      • et al.
      Analysis of gene expression in two large schizophrenia cohorts identifies multiple changes associated with nerve terminal function [no. 12].
      ,
      • Horesh Y.
      • Katsel P.
      • Haroutunian V.
      • Domany E.
      Gene expression signature is shared by patients with Alzheimer’s disease and schizophrenia at the superior temporal gyrus.
      ,
      • Riley B.E.
      • Gardai S.J.
      • Emig-Agius D.
      • Bessarabova M.
      • Ivliev A.E.
      • Schüle B.
      • et al.
      Systems-Based Analyses of Brain Regions Functionally Impacted in Parkinson’s Disease Reveals Underlying Causal Mechanisms.
      ,
      • Hodges A.
      • Strand A.D.
      • Aragaki A.K.
      • Kuhn A.
      • Sengstag T.
      • Hughes G.
      • et al.
      Regional and cellular gene expression changes in human Huntington’s disease brain.
      ). Although this leaves the option open that the set of differentially expressed genes in each disorder might differ from region to region, a recent study in ASD that sampled differential expression across the entire cortex reported good concordance between regions (
      • Haney J.R.
      • Wamsley B.
      • Chen G.T.
      • Parhami S.
      • Emani P.S.
      • Chang N.
      • et al.
      Broad transcriptomic dysregulation across the cerebral cortex in ASD.
      ), which suggests sampling of a limited set of regions to be a reasonable approach. Disorder risk genes can be alternatively defined using genome-wide association studies (GWAS), which have also been successfully linked to neuroimaging findings (
      • Romme I.A.C.
      • de Reus M.A.
      • Ophoff R.A.
      • Kahn R.S.
      • van den Heuvel M.P.
      Connectome Disconnectivity and Cortical Gene Expression in Patients With Schizophrenia.
      ,
      • Ji Y.
      • Zhang X.
      • Wang Z.
      • Qin W.
      • Liu H.
      • Xue K.
      • et al.
      Genes associated with gray matter volume alterations in schizophrenia.
      ). We here opted for defining risk genes based on differential expression due to its more direct measurement in the brain and its potentially closer proximity to disorder phenotypes (
      • Hernandez L.M.
      • Kim M.
      • Hoftman G.D.
      • Haney J.R.
      • Torre-Ubieta L de la
      • Pasaniuc B.
      • Gandal M.J.
      Transcriptomic Insight Into the Polygenic Mechanisms Underlying Psychiatric Disorders.
      ).
      Genes correlating with the cross-disorder involvement map were not found to be particularly enriched for DEGs for MDD. Possibly, the cross-disorder involvement map was less reflective of MDD-related brain alterations. This could be the case due to fewer available MRI experiments on MDD compared to the other disorders when computing the cross-disorder involvement map; or because MDD usually presents less severe gray matter reductions than other psychiatric disorders (
      • Schmaal L.
      • Pozzi E.
      • Ho T C.
      • van Velzen L.S.
      • Veer I.M.
      • Opel N.
      • et al.
      ENIGMA MDD: seven years of global neuroimaging studies of major depression through worldwide data sharing [no. 1].
      ).
      An outstanding question is to what extent the observed similarity in spatial patterns of brain alterations across disorders arises from similar baseline expression levels of disorder risk genes, and to what extent psychiatric disorders target the same cortical regions leading to potential dysregulation of the most highly expressed genes in those regions, which are then observed as differentially expressed in disease. Functional follow-up studies are necessary to further disentangle the causal direction of these effects. Furthermore, it will be interesting to explore whether the observed transcriptomic overlap extends to larger sets of brain disorders, such as other developmental disorders, mood disorders, and personality disorders, which have also been reported to show shared anatomical alterations in the brain (
      • de Lange S.C.
      • Scholtens L.H.
      • van den Berg L.H.
      • Boks M.P.
      • Bozzali M.
      • Cahn W.
      • et al.
      Shared vulnerability for connectome alterations across psychiatric and neurological brain disorders [no. 9].
      ,
      • Goodkind M.
      • Eickhoff S.B.
      • Oathes D.J.
      • Jiang Y.
      • Chang A.
      • Jones-Hagata L.B.
      • et al.
      Identification of a Common Neurobiological Substrate for Mental Illness.
      ). Comparisons with a broader set of neurological diseases will further help establish the degree to which our findings are potentially specific to psychiatric conditions. Astrocytes and microglia also play important roles in neurological conditions such as multiple sclerosis (
      • Ludwin S.K.
      • Rao V.T.
      • Moore C.S.
      • Antel J.P.
      Astrocytes in multiple sclerosis.
      ,
      • Voet S.
      • Prinz M.
      • van Loo G.
      Microglia in Central Nervous System Inflammation and Multiple Sclerosis Pathology.
      ), which calls for more detailed research into the specific pathways and functions that these cells are involved in across different brain disorders. Our findings suggest evidence of spatial overlap between brain transcriptomic and neuroimaging patterns in four major psychiatric conditions. Including other biological resources such as histological or neurotransmitter maps would be of great interest to further refine the cellular and molecular characterization of the pathophysiological processes underlying psychiatric disorders.
      We report differentially expressed genes across four major psychiatric disorders to show high spatial overlap in normative expression profiles across the cortex. These expression profiles converge on cortical regions involved in multimodal cognitive processing and are associated with cross-disorder structural alterations of the brain. Our findings point towards a shared underlying vulnerability of the brain to psychiatric disorders at the level of the transcriptome.

      Acknowledgements

      M.P.v.d.H. was supported by VIDI Grant 452-16-015 and Aard- en levenswetenschappen (ALW) Grant ALWOP.179 from the Netherlands Organization for Scientific Research, and Consolidator Grant 101001062 on cross-disorder connectomics from the European Research Council (ERC).
      Disclosures
      The authors report no biomedical financial interests or potential conflicts of interest.

      Supplementary Material

      References

        • Mueser K.T.
        • McGurk S.R.
        Schizophrenia.
        The Lancet. 2004; 363: 2063-2072
        • Kupfer D.J.
        • Frank E.
        • Phillips M.L.
        Major depressive disorder: new clinical, neurobiological, and treatment perspectives.
        The Lancet. 2012; 379: 1045-1055
        • Grande I.
        • Berk M.
        • Birmaher B.
        • Vieta E.
        Bipolar disorder.
        The Lancet. 2016; 387: 1561-1572
        • Lord C.
        • Elsabbagh M.
        • Baird G.
        • Veenstra-Vanderweele J.
        Autism spectrum disorder.
        The Lancet. 2018; 392: 508-520
        • Pardiñas A.F.
        • Holmans P.
        • Pocklington A.J.
        • Escott-Price V.
        • Ripke S.
        • Carrera N.
        • et al.
        Common schizophrenia alleles are enriched in mutation-intolerant genes and in regions under strong background selection.
        Nature Genetics. 2018; 50: 381-389
        • Ruderfer D.M.
        • Ripke S.
        • McQuillin A.
        • Boocock J.
        • Stahl E.A.
        • Pavlides J.M.W.
        • et al.
        Genomic Dissection of Bipolar Disorder and Schizophrenia, Including 28 Subphenotypes.
        Cell. 2018; 173 (e16): 1705-1715
        • Grove J.
        • Ripke S.
        • Als T.D.
        • Mattheisen M.
        • Walters R.K.
        • Won H.
        • et al.
        Identification of common genetic risk variants for autism spectrum disorder [no. 3].
        Nat Genet. 2019; 51: 431-444
        • Howard D.M.
        • Adams M.J.
        • Clarke T.-K.
        • Hafferty J.D.
        • Gibson J.
        • Shirali M.
        • et al.
        Genome-wide meta-analysis of depression identifies 102 independent variants and highlights the importance of the prefrontal brain regions [no. 3].
        Nat Neurosci. 2019; 22: 343-352
        • Seidlitz J.
        • Nadig A.
        • Liu S.
        • Bethlehem R.A.I.
        • Vértes P.E.
        • Morgan S.E.
        • et al.
        Transcriptomic and cellular decoding of regional brain vulnerability to neurogenetic disorders [no. 1].
        Nature Communications. 2020; 11: 3358
        • Hernandez L.M.
        • Kim M.
        • Hoftman G.D.
        • Haney J.R.
        • Torre-Ubieta L de la
        • Pasaniuc B.
        • Gandal M.J.
        Transcriptomic Insight Into the Polygenic Mechanisms Underlying Psychiatric Disorders.
        Biological Psychiatry. 2021; 89: 54-64
        • Watanabe K.
        • Stringer S.
        • Frei O.
        • Umićević Mirkov M.
        • de Leeuw C.
        • Polderman T.J.C.
        • et al.
        A global overview of pleiotropy and genetic architecture in complex traits [no. 9].
        Nature Genetics. 2019; 51: 1339-1348
        • Akbarian S.
        • Liu C.
        • Knowles J.A.
        • Vaccarino F.M.
        • Farnham P.J.
        • Crawford G.E.
        • et al.
        The PsychENCODE project.
        Nat Neurosci. 2015; 18: 1707-1712
        • Romme I.A.C.
        • de Reus M.A.
        • Ophoff R.A.
        • Kahn R.S.
        • van den Heuvel M.P.
        Connectome Disconnectivity and Cortical Gene Expression in Patients With Schizophrenia.
        Biological Psychiatry. 2017; 81: 495-502
        • Morgan S.E.
        • Seidlitz J.
        • Whitaker K.J.
        • Romero-Garcia R.
        • Clifton N.E.
        • Scarpazza C.
        • et al.
        Cortical patterning of abnormal morphometric similarity in psychosis is associated with brain expression of schizophrenia-related genes.
        PNAS. 2019; 116: 9604-9609
        • Romero-Garcia R.
        • Warrier V.
        • Bullmore E.T.
        • Baron-Cohen S.
        • Bethlehem R.A.I.
        Synaptic and transcriptionally downregulated genes are associated with cortical thickness differences in autism.
        Mol Psychiatry. 2019; 24: 1053-1064
        • Anderson K.M.
        • Collins M.A.
        • Kong R.
        • Fang K.
        • Li J.
        • He T.
        • et al.
        Convergent molecular, cellular, and cortical neuroimaging signatures of major depressive disorder.
        Proc Natl Acad Sci USA. 2020; 117: 25138-25149
        • Li J.
        • Seidlitz J.
        • Suckling J.
        • Fan F.
        • Ji G.-J.
        • Meng Y.
        • et al.
        Cortical structural differences in major depressive disorder correlate with cell type-specific transcriptional signatures [no. 1].
        Nature Communications. 2021; 12: 1647
        • Keshavan M.S.
        • Morris D.W.
        • Sweeney J.A.
        • Pearlson G.
        • Thaker G.
        • Seidman L.J.
        • et al.
        A dimensional approach to the psychosis spectrum between bipolar disorder and schizophrenia: The Schizo-Bipolar Scale.
        Schizophr Res. 2011; 133: 250-254
        • van Loo H.M.
        • Romeijn J.-W.
        Psychiatric comorbidity: fact or artifact?.
        Theor Med Bioeth. 2015; 36: 41-60
        • Hossain M.M.
        • Khan N.
        • Sultana A.
        • Ma P.
        • McKyer E.L.J.
        • Ahmed H.U.
        • Purohit N.
        Prevalence of comorbid psychiatric disorders among people with autism spectrum disorder: An umbrella review of systematic reviews and meta-analyses.
        Psychiatry Research. 2020; 287112922
        • de Lange S.C.
        • Scholtens L.H.
        • van den Berg L.H.
        • Boks M.P.
        • Bozzali M.
        • Cahn W.
        • et al.
        Shared vulnerability for connectome alterations across psychiatric and neurological brain disorders [no. 9].
        Nature Human Behaviour. 2019; 3: 988-998
        • Opel N.
        • Goltermann J.
        • Hermesdorf M.
        • Berger K.
        • Baune B.T.
        • Dannlowski U.
        Cross-Disorder Analysis of Brain Structural Abnormalities in Six Major Psychiatric Disorders: A Secondary Analysis of Mega- and Meta-analytical Findings From the ENIGMA Consortium.
        Biological Psychiatry. 2020; 88: 678-686
        • Patel Y.
        • Parker N.
        • Shin J.
        • Howard D.
        • French L.
        • Thomopoulos S.
        • et al.
        Virtual Histology of Cortical Thickness and Shared Neurobiology in 6 Psychiatric Disorders.
        JAMA Psychiatry. 2021; 78: 47-63
        • Anttila V.
        • Bulik-Sullivan B.
        • Finucane H.K.
        • Walters R.K.
        • Bras Duncan L.
        • et al.
        Analysis of shared heritability in common disorders of the brain.
        Science. 2018; 360eaap8757
        • Lee P.H.
        • Anttila V.
        • Won H.
        • Feng Y.-C.A.
        • Rosenthal J.
        • Zhu Z.
        • et al.
        Genomic Relationships, Novel Loci, and Pleiotropic Mechanisms across Eight Psychiatric Disorders.
        Cell. 2019; 179 (e11): 1469-1482
        • Enwright III, J.F.
        • Lewis D.A.
        Similarities in Cortical Transcriptome Alterations Between Schizophrenia and Bipolar Disorder Are Related to the Presence of Psychosis.
        Schizophrenia Bulletin. 2021; https://doi.org/10.1093/schbul/sbaa195
        • Zeighami Y.
        • Bakken T.
        • Nickl-Jockschat T.
        • Miller J.
        • Evans A.C.
        • Lein E.
        • Hawrylycz M.
        Structural and Cellular Transcriptome Foundations of Human Brain Disease.
        Neuroscience. 2021; https://doi.org/10.1101/2021.05.12.443911
        • Gandal M.J.
        • Haney J.R.
        • Parikshak N.N.
        • Leppa V.
        • Ramaswami G.
        • Hartl C.
        • et al.
        Shared molecular neuropathology across major psychiatric disorders parallels polygenic overlap.
        Science. 2018; 359: 693-697
        • Hawrylycz M.J.
        • Lein E.S.
        • Guillozet-Bongaarts A.L.
        • Shen E.H.
        • Ng L.
        • Miller J.A.
        • et al.
        An anatomically comprehensive atlas of the adult human brain transcriptome [no. 7416].
        Nature. 2012; 489: 391-399
        • Iwamoto K.
        • Bundo M.
        • Kato T.
        Altered expression of mitochondria-related genes in postmortem brains of patients with bipolar disorder or schizophrenia, as revealed by large-scale DNA microarray analysis.
        Human Molecular Genetics. 2005; 14: 241-253
        • Garbett K.
        • Ebert P.J.
        • Mitchell A.
        • Lintas C.
        • Manzi B.
        • Mirnics K.
        • Persico A.M.
        Immune transcriptome alterations in the temporal cortex of subjects with autism.
        Neurobiology of Disease. 2008; 30: 303-311
        • Narayan S.
        • Tang B.
        • Head S.R.
        • Gilmartin T.J.
        • Sutcliffe J.G.
        • Dean B.
        • Thomas E.A.
        Molecular profiles of schizophrenia in the CNS at different stages of illness.
        Brain Research. 2008; 1239: 235-248
        • Maycox P.R.
        • Kelly F.
        • Taylor A.
        • Bates S.
        • Reid J.
        • Logendra R.
        • et al.
        Analysis of gene expression in two large schizophrenia cohorts identifies multiple changes associated with nerve terminal function [no. 12].
        Molecular Psychiatry. 2009; 14: 1083-1094
        • Voineagu I.
        • Wang X.
        • Johnston P.
        • Lowe J.K.
        • Tian Y.
        • Horvath S.
        • et al.
        Transcriptomic analysis of autistic brain reveals convergent molecular pathology.
        Nature. 2011; 474: 380-384
        • Chow M.L.
        • Pramparo T.
        • Winn M.E.
        • Barnes C.C.
        • Li H.-R.
        • Weiss L.
        • et al.
        Age-Dependent Brain Gene Expression and Copy Number Anomalies in Autism Suggest Distinct Pathological Processes at Young Versus Mature Ages.
        PLOS Genetics. 2012; 8e1002592
        • Chen C.
        • Cheng L.
        • Grennan K.
        • Pibiri F.
        • Zhang C.
        • Badner J.A.
        • et al.
        Two gene co-expression modules differentiate psychotics and controls.
        Mol Psychiatry. 2013; 18: 1308-1314
        • Chang L.-C.
        • Jamain S.
        • Lin C.-W.
        • Rujescu D.
        • Tseng G.C.
        • Sibille E.
        A Conserved BDNF, Glutamate- and GABA-Enriched Gene Module Related to Human Depression Identified by Coexpression Meta-Analysis and DNA Variant Genome-Wide Association Studies.
        PLOS ONE. 2014; 9e90980
        • Reinhart V.
        • Bove S.E.
        • Volfson D.
        • Lewis D.A.
        • Kleiman R.J.
        • Lanz T.A.
        Evaluation of TrkB and BDNF transcripts in prefrontal cortex, hippocampus, and striatum from subjects with schizophrenia, bipolar disorder, and major depressive disorder.
        Neurobiology of Disease. 2015; 77: 220-227
        • Horesh Y.
        • Katsel P.
        • Haroutunian V.
        • Domany E.
        Gene expression signature is shared by patients with Alzheimer’s disease and schizophrenia at the superior temporal gyrus.
        European Journal of Neurology. 2011; 18: 410-424
        • Riley B.E.
        • Gardai S.J.
        • Emig-Agius D.
        • Bessarabova M.
        • Ivliev A.E.
        • Schüle B.
        • et al.
        Systems-Based Analyses of Brain Regions Functionally Impacted in Parkinson’s Disease Reveals Underlying Causal Mechanisms.
        PLOS ONE. 2014; 9e102909
        • Hodges A.
        • Strand A.D.
        • Aragaki A.K.
        • Kuhn A.
        • Sengstag T.
        • Hughes G.
        • et al.
        Regional and cellular gene expression changes in human Huntington’s disease brain.
        Human Molecular Genetics. 2006; 15: 965-977
        • Crow M.
        • Lim N.
        • Ballouz S.
        • Pavlidis P.
        • Gillis J.
        Predictability of human differential gene expression.
        PNAS. 2019; 116: 6491-6500
        • Wei Y.
        • de Lange S.C.
        • Pijnenburg R.
        • Scholtens L.H.
        • Ardesch D.J.
        • Watanabe K.
        • et al.
        Statistical testing in transcriptomic-neuroimaging studies: A how-to and evaluation of methods assessing spatial and gene specificity.
        Human Brain Mapping. 2022; 43: 885-901
        • Yeo B.T.T.
        • Krienen F.M.
        • Sepulcre J.
        • Sabuncu M.R.
        • Lashkari D.
        • Hollinshead M.
        • et al.
        The organization of the human cerebral cortex estimated by intrinsic functional connectivity.
        Journal of Neurophysiology. 2011; 106: 1125-1165
        • Wei Y.
        • Lange S.C.
        • Scholtens L.H.
        • Watanabe K.
        • Ardesch D.J.
        • Jansen P.R.
        • et al.
        Genetic mapping and evolutionary analysis of human-expanded cognitive networks.
        Nature Communications. 2019; 10: 1-11
        • Alexander-Bloch A.F.
        • Shou H.
        • Liu S.
        • Satterthwaite T.D.
        • Glahn D.C.
        • Shinohara R.T.
        • et al.
        On testing for spatial correspondence between maps of human brain structure and function.
        NeuroImage. 2018; 178: 540-551
        • Markello R.D.
        • Misic B.
        Comparing spatial null models for brain maps.
        NeuroImage. 2021; 236118052
        • Subramanian A.
        • Tamayo P.
        • Mootha V.K.
        • Mukherjee S.
        • Ebert B.L.
        • Gillette M.A.
        • et al.
        Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles.
        Proceedings of the National Academy of Sciences. 2005; 102: 15545-15550
        • Zhang B.
        • Kirov S.
        • Snoddy J.
        WebGestalt: an integrated system for exploring gene sets in various biological contexts.
        Nucleic Acids Research. 2005; 33: W741-W748
        • Liao Y.
        • Wang J.
        • Jaehnig E.J.
        • Shi Z.
        • Zhang B.
        WebGestalt 2019: gene set analysis toolkit with revamped UIs and APIs.
        Nucleic Acids Research. 2019; 47: W199-W205
        • Krishnan A.
        • Williams L.J.
        • McIntosh A.R.
        • Abdi H.
        Partial Least Squares (PLS) methods for neuroimaging: A tutorial and review.
        NeuroImage. 2011; 56: 455-475
        • Benjamini Y.
        • Hochberg Y.
        Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing.
        Journal of the Royal Statistical Society: Series B (Methodological). 1995; 57: 289-300
        • Cole M.W.
        • Repovš G.
        • Anticevic A.
        The Frontoparietal Control System: A Central Role in Mental Health.
        Neuroscientist. 2014; 20: 652-664
        • Maurano M.T.
        • Humbert R.
        • Rynes E.
        • Thurman R.E.
        • Haugen E.
        • Wang H.
        • et al.
        Systematic Localization of Common Disease-Associated Variation in Regulatory DNA.
        Science. 2012; 337: 1190-1195
        • Mesulam M.
        From sensation to cognition.
        Brain. 1998; 121: 1013-1052
        • van den Heuvel M.P.
        • Sporns O.
        Network hubs in the human brain.
        Trends in Cognitive Sciences. 2013; 17: 683-696
        • Benowitz L.I.
        • Routtenberg A.
        GAP-43: an intrinsic determinant of neuronal development and plasticity.
        Trends in Neurosciences. 1997; 20: 84-91
        • García-Cabezas M.Á.
        • Zikopoulos B.
        • Barbas H.
        The Structural Model: a theory linking connections, plasticity, pathology, development and evolution of the cerebral cortex.
        Brain Structure and Function. 2019; 224: 985-1008
        • Vaishnavi S.N.
        • Vlassenko A.G.
        • Rundle M.M.
        • Snyder A.Z.
        • Mintun M.A.
        • Raichle M.E.
        Regional aerobic glycolysis in the human brain.
        Proceedings of the National Academy of Sciences. 2010; 107: 17757-17762
        • Fulcher B.D.
        • Fornito A.
        A transcriptional signature of hub connectivity in the mouse connectome.
        Proceedings of the National Academy of Sciences. 2016; 113: 1435-1440
        • Crossley N.A.
        • Mechelli A.
        • Scott J.
        • Carletti F.
        • Fox P.T.
        • McGuire P.
        • Bullmore E.T.
        The hubs of the human connectome are generally implicated in the anatomy of brain disorders.
        Brain. 2014; 137: 2382-2395
        • van den Heuvel M.P.
        • Sporns O.
        A cross-disorder connectome landscape of brain dysconnectivity.
        Nature Reviews Neuroscience. 2019; 20: 435-446
        • Goodkind M.
        • Eickhoff S.B.
        • Oathes D.J.
        • Jiang Y.
        • Chang A.
        • Jones-Hagata L.B.
        • et al.
        Identification of a Common Neurobiological Substrate for Mental Illness.
        JAMA Psychiatry. 2015; 72: 305-315
        • Park B.
        • Kebets V.
        • Lariviere S.
        • Hettwer M.D.
        • Paquola C.
        • van Rooij D.
        • et al.
        Multilevel Neural Gradients Reflect Transdiagnostic Effects of Major Psychiatric Conditions on Cortical Morphology.
        Neuroscience. 2021; https://doi.org/10.1101/2021.10.29.466434
        • Meda S.A.
        • Ruaño G.
        • Windemuth A.
        • O’Neil K.
        • Berwise C.
        • Dunn S.M.
        • et al.
        Multivariate analysis reveals genetic associations of the resting default mode network in psychotic bipolar disorder and schizophrenia.
        PNAS. 2014; 111: E2066-E2075
        • Padmanabhan A.
        • Lynch C.J.
        • Schaer M.
        • Menon V.
        The Default Mode Network in Autism.
        Biological Psychiatry: Cognitive Neuroscience and Neuroimaging. 2017; 2: 476-486
        • Doucet G.E.
        • Janiri D.
        • Howard R.
        • O’Brien M.
        • Andrews-Hanna J.R.
        • Frangou S.
        Transdiagnostic and disease-specific abnormalities in the default-mode network hubs in psychiatric disorders: A meta-analysis of resting-state functional imaging studies.
        European Psychiatry. 2020; 63https://doi.org/10.1192/j.eurpsy.2020.57
        • Parikshak N.N.
        • Swarup V.
        • Belgard T.G.
        • Irimia M.
        • Ramaswami G.
        • Gandal M.J.
        • et al.
        Genome-wide changes in lncRNA, splicing, and regional gene expression patterns in autism.
        Nature. 2016; 540: 423-427
        • Salter M.W.
        • Stevens B.
        Microglia emerge as central players in brain disease.
        Nat Med. 2017; 23: 1018-1027
        • Vainchtein I.D.
        • Molofsky A.V.
        Astrocytes and Microglia: In Sickness and in Health.
        Trends in Neurosciences. 2020; 43: 144-154
        • Fatemi S.H.
        • Folsom T.D.
        The Neurodevelopmental Hypothesis of Schizophrenia, Revisited.
        Schizophrenia Bulletin. 2009; 35: 528-548
        • Wagstyl K.
        • Ronan L.
        • Whitaker K.J.
        • Goodyer I.M.
        • Roberts N.
        • Crow T.J.
        • Fletcher P.C.
        Multiple markers of cortical morphology reveal evidence of supragranular thinning in schizophrenia.
        Transl Psychiatry. 2016; 6 (e780–e780)
        • Hook P.W.
        • McCallion A.S.
        Leveraging mouse chromatin data for heritability enrichment informs common disease architecture and reveals cortical layer contributions to schizophrenia.
        Genome Res. 2020; 30: 528-539
        • Pierri J.N.
        • Volk C.L.E.
        • Auh S.
        • Sampson A.
        • Lewis D.A.
        Decreased Somal Size of Deep Layer 3 Pyramidal Neurons in the Prefrontal Cortex of Subjects With Schizophrenia.
        Archives of General Psychiatry. 2001; 58: 466-473
        • Kolluri N.
        • Sun Z.
        • Sampson A.R.
        • Lewis D.A.
        Lamina-Specific Reductions in Dendritic Spine Density in the Prefrontal Cortex of Subjects With Schizophrenia.
        AJP. 2005; 162: 1200-1202
        • van den Heuvel M.P.
        • Scholtens L.H.
        • Reus MA de
        • Kahn R.S.
        Associated Microscale Spine Density and Macroscale Connectivity Disruptions in Schizophrenia.
        Biological Psychiatry. 2016; 80: 293-301
        • Scholtens L.H.
        • van den Heuvel M.P.
        Multimodal Connectomics in Psychiatry: Bridging Scales From Micro to Macro.
        Biological Psychiatry: Cognitive Neuroscience and Neuroimaging. 2018; 3: 767-776
        • Spielman R.S.
        • Bastone L.A.
        • Burdick J.T.
        • Morley M.
        • Ewens W.J.
        • Cheung V.G.
        Common genetic variants account for differences in gene expression among ethnic groups.
        Nat Genet. 2007; 39: 226-231
        • Berchtold N.C.
        • Cribbs D.H.
        • Coleman P.D.
        • Rogers J.
        • Head E.
        • Kim R.
        • et al.
        Gene expression changes in the course of normal brain aging are sexually dimorphic.
        PNAS. 2008; 105: 15605-15610
        • Kumar A.
        • Gibbs J.R.
        • Beilina A.
        • Dillman A.
        • Kumaran R.
        • Trabzuni D.
        • et al.
        Age-associated changes in gene expression in human brain and isolated neurons.
        Neurobiology of Aging. 2013; 34: 1199-1209
        • Haney J.R.
        • Wamsley B.
        • Chen G.T.
        • Parhami S.
        • Emani P.S.
        • Chang N.
        • et al.
        Broad transcriptomic dysregulation across the cerebral cortex in ASD.
        bioRxiv 2020. 2020; (12.17.423129)
        • Ji Y.
        • Zhang X.
        • Wang Z.
        • Qin W.
        • Liu H.
        • Xue K.
        • et al.
        Genes associated with gray matter volume alterations in schizophrenia.
        NeuroImage. 2021; 225117526
        • Schmaal L.
        • Pozzi E.
        • Ho T C.
        • van Velzen L.S.
        • Veer I.M.
        • Opel N.
        • et al.
        ENIGMA MDD: seven years of global neuroimaging studies of major depression through worldwide data sharing [no. 1].
        Transl Psychiatry. 2020; 10: 1-19
        • Ludwin S.K.
        • Rao V.T.
        • Moore C.S.
        • Antel J.P.
        Astrocytes in multiple sclerosis.
        Mult Scler. 2016; 22: 1114-1124
        • Voet S.
        • Prinz M.
        • van Loo G.
        Microglia in Central Nervous System Inflammation and Multiple Sclerosis Pathology.
        Trends Mol Med. 2019; 25: 112-123