CoChat
Sign Up
Cited by 1,247Autism Spectrum Disorder ResearchSchizophrenia research and treatmentRNA Research and Splicing

Transcriptome-wide isoform-level dysregulation in ASD, schizophrenia, and bipolar disorder

Open access
Gandal, Michael J., Zhang, Pan, Hadjimichael, Evi, Walker, Rebecca L., Chen, Chao, Liu, Shuang, Won, Hyejung, van Bakel, Harm +181 more·2018·Science
DOI 10.1126/science.aat8127PMC PMC6443102
Open this paper in CoChat — free account

Abstract

INTRODUCTION: Our understanding of the pathophysiology of psychiatric disorders, including autism spectrum disorder (ASD), schizophrenia (SCZ), and bipolar disorder (BD), lags behind other fields of medicine. The diagnosis and study of these disorders currently depend on behavioral, symptomatic characterization. Defining genetic contributions to disease risk allows for biological, mechanistic understanding but is challenged by genetic complexity, polygenicity, and the lack of a cohesive neurobiological model to interpret findings. RATIONALE: The transcriptome represents a quantitative phenotype that provides biological context for understanding the molecular pathways disrupted in major psychiatric disorders. RNA sequencing (RNA-seq) in a large cohort of cases and controls can advance our knowledge of the biology disrupted in each disorder and provide a foundational resource for integration with genomic and genetic data. RESULTS: Analysis across multiple levels of transcriptomic organization—gene expression, local splicing, transcript isoform expression, and coexpression networks for both protein-coding and noncoding genes—provides an in-depth view of ASD, SCZ, and BD molecular pathology. More than 25% of the transcriptome exhibits differential splicing or expression in at least one disorder, including hundreds of noncoding RNAs (ncRNAs), most of which have unexplored functions but collectively exhibit patterns of selective constraint. Changes at the isoform level, as opposed to the gene level, show the largest effect sizes and genetic enrichment and the greatest disease specificity. We identified coexpression modules associated with each disorder, many with enrichment for cell type–specific markers, and several modules significantly dysregulated across all three disorders. These enabled parsing of down-regulated neuronal and synaptic components into a variety of cell type– and disease-specific signals, including multiple excitatory neuron and distinct interneuron modules with differential patterns of disease association, as well as common and rare genetic risk variant enrichment. The glial-immune signal demonstrates shared disruption of the blood-brain barrier and up-regulation of NFkB-associated genes, as well as disease-specific alterations in microglial-, astrocyte-, and interferon-response modules. A coexpression module associated with psychiatric medication exposure in SCZ and BD was enriched for activity-dependent immediate early gene pathways. To identify causal drivers, we integrated polygenic risk scores and performed a transcriptome-wide association study and summary-data–based Mendelian randomization. Candidate risk genes—5 in ASD, 11 in BD, and 64 in SCZ, including shared genes between SCZ and BD—are supported by multiple methods. These analyses begin to define a mechanistic basis for the composite activity of genetic risk variants. CONCLUSION: Integration of RNA-seq and genetic data from ASD, SCZ, and BD provides a quantitative, genome-wide resource for mechanistic insight and therapeutic development at Resource.PsychENCODE.org. These data inform the molecular pathways and cell types involved, emphasizing the importance of splicing and isoform-level gene regulatory mechanisms in defining cell type and disease specificity, and, when integrated with genome-wide association studies, permit the discovery of candidate risk genes. The PsychENCODE cross-disorder transcriptomic resource.: Human brain RNA-seq was integrated with genotypes across individuals with ASD, SCZ, BD, and controls, identifying pervasive dysregulation, including protein-coding, noncoding, splicing, and isoform-level changes. Systems-level and integrative genomic analyses prioritize previously unknown neurogenetic mechanisms and provide insight into the molecular neuropathology of these disorders.

Verified 2026-10-06 · 1 reference not checked yet

Integrity signals

Registry

Crossref · 10.1126/science.aat8127

Cites

146 of 147 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 123 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    DE genes were enriched for known pathways and cell-type-specific markers from single-nucleus RNA-seq in human brain.

    “These DE genes exhibited substantial enrichment for known pathways and cell type specific markers derived from single nucleus RNA-Seq in human brain (Fig 1D-E) (21), consistent with previously observed patterns (1, 19).”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites [21] See supplemental methods.

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [1] Gandal MJ, Shared molecular neuropathology across major psychiatric disorders parallels polygenic overlap. Science. 359, 693– 697 (2018). 29439242 PMC5898828 source

      Not checked in the free run.

    • cites [19] Parikshak NN, Genome-wide changes in lncRNA, splicing, and regional gene expression patterns in autism. Nature. 540, 423– 427 (2016). 27919067 PMC7102905 source

      Not checked in the free run.

  • Findingown data

    Isoform-only DE genes overlapped most with excitatory neuron clusters.

    “exhibited greatest overlap with excitatory neuron clusters (OR’s> 4, Fisher’s exact test, FDR’s<10−10)”
    Quote in the paperNot a substantial claim
  • Findingown data

    Isoform-only DE genes were more often downregulated than upregulated in disease.

    “These isoform-only DE genes were more likely to be downregulated than upregulated in disease (one sample t-test, P<10−16)”
    Quote in the paperNot a substantial claim
  • Findingown data

    Isoform-level alterations in disease had larger effect sizes than gene-level changes, particularly for protein-coding biotypes.

    “In addition, isoform-level alterations in disease exhibited substantially larger effect sizes compared with gene-level changes (mean |log2FC| 0.25 vs 0.14, P<2×10−16, K-S test), particularly for protein coding biotypes (Fig 1A), consistent with recent work demonstrating the importance of splicing dysregulation in disease pathogenesis (25).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [25] Li YI, RNA splicing is a primary link between genetic variation and disease. Science. 352, 600– 604 (2016). 27126046 PMC5182069 source

      Not checked in the free run.

  • Findingown data

    Isoform-only DE genes were enriched for neuron projection development, mRNA metabolism, and synaptic pathways.

    “and showed significant enrichment for neuron projection development, mRNA metabolism, and synaptic pathways (FDR<3×10−3; Table S1).”
    Quote in the paperNot a substantial claim
  • May combine two claims. 85
  • Classified as probably not a substantial claim. 25

  • May combine two claims.

    Findingcitation

    Pervasive differential gene expression was observed in ASD, SCZ, and BD, with 1611, 4821, and 1119 genes at FDR<0.05, respectively.

    “In accordance with previous results (1), we observed pervasive differential gene expression (DGE) in ASD, SCZ, and BD (n=1611, 4821, and 1119 genes at FDR<0.05, respectively; Fig 1A; Table S1).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [1] Gandal MJ, Shared molecular neuropathology across major psychiatric disorders parallels polygenic overlap. Science. 359, 693– 697 (2018). 29439242 PMC5898828 source

      Not checked in the free run.

  • May combine two claims.

    Findingown data

    Widespread differential transcript expression was observed in ASD, SCZ, and BD, with 767, 3803, and 248 isoforms at FDR<0.05, respectively.

    “Expanding these analyses to the transcript isoform-level, we observe widespread differential transcript expression (DTE) across ASD, SCZ, and BD (n=767, 3803, and 248 isoforms at FDR<0.05, respectively; Table S1).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    There was substantial cross-disorder sharing of DE signal and a gradient of transcriptomic severity, with the largest changes in ASD compared with SCZ or BD.

    “There was substantial cross-disorder sharing of this DE signal and a gradient of transcriptomic severity with the largest changes in ASD compared with SCZ or BD (ASD vs SCZ, mean |log2FC| 0.26 vs 0.10, P<2×10−16, Kolmogorov-Smirnov (K-S) test; ASD vs BD, mean |log2FC| 0.26 vs 0.15, P<2×10−16, K-S test), as observed previously (1).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [1] Gandal MJ, Shared molecular neuropathology across major psychiatric disorders parallels polygenic overlap. Science. 359, 693– 697 (2018). 29439242 PMC5898828 source

      Not checked in the free run.

  • May combine two claims.

    Findingown data

    Isoform-level analysis identified DE transcripts without gene-level DGE, including 811 in SCZ, 294 in ASD, and 60 in BD.

    “Furthermore, although isoform and gene-level changes exhibited similar pathway and cell type enrichments (e.g. Fig 1D-E), isoform-level analysis identified DE transcripts that did not show DGE (‘isoform-only DE’), including 811 in SCZ, 294 in ASD, and 60 in BD.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingcitation

    944 non-coding RNAs showed gene- or isoform-level DE in at least one disorder, including 693 in SCZ, 178 in ASD, and 174 in BD; 208, 60, and 52 were intergenic lincRNAs, respectively.

    “Based on Gencode annotations, we identify 944 ncRNAs exhibiting gene- or isoform-level DE in at least one disorder (herein referred to as ‘neuropsychiatric (NP) ncRNAs’ (21)), 693 of which were DE in SCZ, 178 in ASD, and 174 in BD, of which 208, 60, and 52 are annotated as intergenic long non-coding RNAs (lincRNAs), in each disorder, respectively.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [21] See supplemental methods.

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Findingown data

    DS genes overlapped significantly with DTE results for ASD and SCZ, but not BD, which likely remains underpowered.

    “DS genes overlapped significantly with DTE results for ASD and SCZ (Fig S8B), but not BD, which likely still remains underpowered.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Overlap among DS genes existed, but the majority of splicing changes were disorder specific.

    “and subsequently, overlap among DS genes (Fig 2A-B), although the majority of splicing changes still are disorder specific.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Of 4821 DE genes in SCZ, only 143 showed even nominal association between effect size magnitude and age; similar associations were seen in 29 of 1119 BD and 85 of 1611 ASD DE genes.

    “Of 4821 DE genes in SCZ, only 143 showed even nominal association between effect size magnitude and age. Similar associations were seen in 29 of 1119 DE genes in BD and 85 of 1611 DE genes in ASD.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    18 genes or isoforms had expression significantly associated with PRS at Bonferroni-corrected P<0.05, including 16 in ASD and 2 in SCZ, none in BD.

    “We identified 18 genes or isoforms whose expression was significantly associated with PRS ((21); Bonferroni-corrected P < 0.05), 16 in ASD, 2 in SCZ, with none in BD (Fig 3C; Table S4).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [21] See supplemental methods.

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Findingown data

    LINC00634 was downregulated in SCZ and had a genome-wide significant SCZ TWAS association.

    “and LINC00634, an unstudied brain enriched lincRNA downregulated in SCZ (log2FC −0.06, FDR 0.027) with a genome-wide significant SCZ TWAS association as described below.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Significant associations with PRS were observed for DGE and DTE signal in SCZ, with greater polygenic association at the isoform level.

    “Significant associations with PRS were observed for DGE and DTE signal in SCZ, with greater polygenic association at the isoform level in accordance with the larger transcript isoform effect sizes observed.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    515 DS intron clusters in 472 genes were identified across all disorders at FDR<0.1, and 117 (25%) contained one or more novel exons.

    “We identified 515 DS intron clusters in 472 genes across all disorders (FDR<0.1), 117 of which (25%) contained one or more novel exons (Table S3; Fig 2A).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The most common local splicing change was exon skipping (41-60%), followed by alternative 5' exon inclusion (11-21%) and alternative 3' splice site usage (5-18%).

    “The most commonly observed local splicing change was exon skipping (41–60%), followed by alternative 5’ exon inclusion (e.g. due to alternative promoter usage; 11–21%) and alternative 3’ splice site usage (5–18%) (Table S3; Fig S8A).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    CADPS, located within an ASD GWAS risk locus and supported by Hi-C as a putative target, manifests multiple isoform and splice alterations in ASD.

    “Another example is CADPS, which is located within an ASD GWAS risk locus and supported by Hi-C defined chromatin interactions as a putative target gene (38) and manifests multiple isoform and splice alterations in ASD (Fig S9; Tables S1 and S3).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [38] Grove J, Common risk variants identified in autism spectrum disorder. bioRxiv (2017), p. 224774. source

      Not checked in the free run.

  • May combine two claims.

    Findingcitation

    74 NPncRNAs (~8%) were under purifying selection in humans, with average exon-level CDTS below the 10th percentile.

    “We identify 74 NPncRNAs (~8%) under purifying selection in humans, with average exon-level context-dependent tolerance scores (CDTS) below the 10th percentile (21).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [21] See supplemental methods.

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Findingown data

    Over 200 NPncRNAs had broad non-specific expression across cell types, while 66 were expressed within a specific cell type class.

    “Over 200 NPncRNAs exhibited broad and non-specific expression patterns across cell types, whereas 66 were expressed within a specific cell type class (Table S2).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    LINC00996 was downregulated in SCZ and BD and restricted to microglia in brain.

    “Notable examples are: LINC00996, which is downregulated in SCZ (log2FC −0.71, FDR<5×10−11) and BD (log2FC −0.45, FDR=0.02) and restricted to microglia in brain (Fig S6)”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    GRIN1 was upregulated in SCZ and BD and showed increased skipping of exon 4 in ASD and SCZ, impacting its extracellular ligand-binding domain.

    “GRIN1 encodes the obligatory subunit of the NMDA-type glutamate ionotropic receptors, is upregulated in SCZ and BD and shows increased skipping of exon 4 in both ASD and SCZ that impacts its extracellular ligand-binding domain (Fig 2E-G).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    LINC00343 was expressed in excitatory neurons and downregulated in BD, with a trend in SCZ.

    “LINC00343, expressed in excitatory neurons, and downregulated in BD (log2FC −0.33, FDR=0.012) with a trend in SCZ (log2FC −0.15, FDR 0.065)”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingcitation

    48 DS genes (10%) encode RNA binding proteins or splicing factors, and at least six splicing factors also showed DTE.

    “Notably, 48 DS genes (10%; FDR=8.8×10−4) encode RNA binding proteins or splicing factors (34), with at least six splicing factors also showing DTE in ASD (MATR3), SCZ (QKI, RBM3, SRRM2, U2AF1) or both (SRSF11).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [34] Sebestyén E, Large-scale analysis of genome and transcriptome alterations in multiple tumors unveils novel cancer-relevant splicing networks. Genome Res. 26, 732– 744 (2016). 27197215 PMC4889968 source

      Not checked in the free run.

  • May combine two claims.

    Findingown data

    The second most abundant RBFOX1 isoform was in isoM17, which was downregulated in both ASD and SCZ.

    “The second most abundant RBFOX1 isoform is in another module, isoM17, which is downregulated in both ASD and SCZ (Fig 6B).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The two RBFOX1 modules showed differential association with common genetic risk, with isoM2 exhibiting GWAS enrichment across SCZ, BD, and MDD.

    “Moreover, the two modules show differential association with common genetic risk (Fig 6E), with isoM2 exhibiting GWAS enrichment across SCZ, BD, and MDD.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Of 107 conditionally significant SCZ genes, 62 were supported by SMR and 11 were concordantly DE in SCZ brain in the same direction as predicted by TWAS.

    “Of the 107 conditionally significant genes prioritized by TWAS, 62 were further supported by SMR (PSMR<0.05, PHEIDI>0.05) and 11 were also concordantly DE in SCZ brain in the same direction as predicted by TWAS.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    isoM13 showed the greatest overall significance of enrichment for SCZ and educational attainment GWAS, and was also enriched in BD GWAS to a lesser degree.

    “isoM13 showed the greatest overall significance of enrichment for SCZ and educational attainment GWAS, and was also enriched in BD GWAS to a lesser degree,”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Most genes identified in TWAS showed disease-specific effects, with only four genes overlapping between SCZ and BD TWAS and none overlapping with ASD.

    “Most genes identified in this analysis show disease-specific effects, as only four genes (MCHR1, VPS45, SNAP91, DCLK3) showed overlap between SCZ and BD TWAS, and none overlapped with ASD.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingcitation

    The 17q21.31 region showed multiple PRS associations but did not reach genome-wide significance in the largest GWAS.

    “This includes the 17q21.31 region, which showed multiple PRS associations as described above, but did not reach genome-wide significance in the largest GWAS to date (38).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [38] Grove J, Common risk variants identified in autism spectrum disorder. bioRxiv (2017), p. 224774. source

      Not checked in the free run.

  • May combine two claims.

    Findingown data

    In SCZ, TWAS identified 193 genes, of which 107 remained significant after conditional analysis within multi-hit loci.

    “In SCZ, TWAS identifies 193 genes of which 107 remain significant following conditional analysis at each gene within multi-hit loci.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Module pair geneM1/isoM2 was downregulated only in ASD, contained the predominant brain-expressed RBFOX1 isoform, and included cation channels HCN1 and SCN8A.

    “The module pair geneM1/isoM2, downregulated only in ASD (Fig 6B), contains the predominant brain-expressed RBFOX1 isoform and includes several cation channels (e.g., HCN1, SCN8A).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    C4A expression was positively correlated with imputed C4A copy number and imputed C4-HERV elements, and slightly negatively associated with C4B copy number.

    “Concordantly, we find a strong positive correlation between C4A expression and genetically imputed C4A copy number (R=0.36, P=6×10−21) and imputed number of C4-HERV elements (R=0.35, P=4×10−20), but a slight negative association with C4B copy number (R=−0.087, P=0.03; ref (21)).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [21] See supplemental methods.

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Findingown data

    Five modules were shared across all three disorders (3 upregulated, 2 downregulated); 22 were shared by two disorders; 36 showed more specific dysregulation in ASD, SCZ, or BD.

    “Five modules are shared across all three disorders, 3 up and two downregulated; 22 modules are shared by 2 of the 3 disorders, and 36 demonstrate more specific patterns of dysregulation in either ASD, SCZ or BD (Fig 5; Table S5).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The third shared process, geneM26/isoM22, is downregulated and enriched for endothelial and pericyte genes, with hubs including blood-brain barrier markers ITIH5, SLC38A5, ABCB1, and GPR124.

    “The third biological process, geneM26/isoM22 (Fig 5C), is downregulated, and enriched for endothelial and pericyte genes, with hubs that represent markers of the blood-brain barrier, including ITIH5, SLC38A5, ABCB1, and GPR124, a critical regulator of brain-specific angiogenesis (53, 54).”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [53] Daneman R, The mouse blood-brain barrier transcriptome: a new resource for understanding the development and function of brain endothelial cells. PLoS One. 5, e13741 (2010). 21060791 PMC2966423 source

      Not checked in the free run.

    • cites [54] Obermeier B, Daneman R, Ransohoff RM, Development, maintenance and disruption of the blood-brain barrier. Nat. Med 19, 1584– 1596 (2013). 24309662 PMC4080800 source

      Not checked in the free run.

  • May combine two claims.

    Findingown data

    Neuronal modules were broadly dysregulated across multiple disorders, including isoM18 with multiple isoforms of ANK2 as hubs.

    “We do observe neuronal modules broadly dysregulated across multiple disorders, including a neuronal/synaptic module (isoM18) with multiple isoforms of the known ASD risk gene, ANK2, as hubs.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Modules showed enrichment for gene ontology pathways and markers of all major CNS cell types.

    “Consistent with expectations, modules showed enrichment for gene ontology pathways and we identified modules strongly and selectively enriched for markers of all major CNS cell types (Fig 5A-B; Fig S12)”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingcitation

    Altogether, 64 genes were consistently prioritized across multiple methods, including 10 ncRNAs.

    “Altogether, 64 genes were consistently prioritized across multiple methods, including 10 ncRNAs (Table S4; (21)).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [21] See supplemental methods.

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Findingown data

    Isoform-level networks captured greater detail and a larger proportion were associated with disease GWAS than gene-level networks (61% vs 41% nominal GWAS enrichment, P=0.07).

    “However, the isoform-level networks captured greater detail and a larger proportion were associated with disease GWAS than gene-level networks (61% vs 41% with nominal GWAS enrichment, P=0.07, χ2; Fig 5A).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingcitation

    Of seven TWAS-significant genes at 17q21.31, conditional analysis prioritized LRRC37A, supported by SMR and Hi-C interaction in fetal brain.

    “Of the seven TWAS-significant genes at this locus, conditional analysis prioritizes one – LRRC37A, which is further supported by SMR and Hi-C interaction in fetal brain (38).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [38] Grove J, Common risk variants identified in autism spectrum disorder. bioRxiv (2017), p. 224774. source

      Not checked in the free run.

  • May combine two claims.

    Findingown data

    A previous SCZ TWAS based primarily on non-neural tissue prioritized 157 genes, 37 of which overlapped with this study, a significant overlap.

    “A previous TWAS study in SCZ primarily based on non-neural tissue prioritized 157 genes, 37 of which are identified here, a significant overlap (OR 61, p<10−42, Fisher’s exact test).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Novel SCZ candidates included downregulated lysine methyltransferases SETD6 and SETD8; downregulated RERE; downregulated LINC00634; and SLC12A5.

    “These included a number of intriguing novel candidates for SCZ: two downregulated lysine methyltransferayjec9vses (SETD6, SETD8); RERE, a downregulated, mutationally intolerant nuclear receptor co-regulator of retinoic acid signaling associated with a rare neurodevelopmental genetic syndrome; LINC00634, a downregulated poorly annotated brain-enriched lincRNA; and SLC12A5, encoding a mitochondrial”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    geneM2/isoM13 is enriched for oligodendrocyte markers and neuron projection developmental pathways and is downregulated in ASD and SCZ, with a trend in BD.

    “A notable example is geneM2/isoM13, which is enriched for oligodendrocyte markers and neuron projection developmental pathways and is downregulated in ASD and SCZ, with a trend in BD (Fig 5C).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingcitation

    isoM17 showed greater enrichment for nuclear RBFOX1 targets, whereas isoM2 showed stronger overlap with cytoplasmic targets.

    “Here, we find that isoM17 shows greater enrichment for nuclear RBFOX1 targets (Fig 6C), whereas isoM2 shows stronger overlap with cytoplasmic targets (32).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [32] Lee J-A, Cytoplasmic Rbfox1 Regulates the Expression of Synaptic and Autism-Related Genes. Neuron. 89, 113– 128 (2016). 26687839 PMC4858412 source

      Not checked in the free run.

  • May combine two claims.

    Findingown data

    In BD, TWAS prioritized 17 genes across 14 distinct loci at Bonferroni-corrected P<0.05, none of which exhibited DE.

    “In BD, TWAS prioritizes 17 genes across 14 distinct loci (Bonferroni-corrected P<0.05; Fig 4; Table S4), none of which exhibited DE.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingcitation

    The interferon response module geneM32 contains critical components of the ISGF3 complex.

    “The interferon response module (geneM32) contains critical components of the IFN-stimulated gene factor 3 (ISGF3) complex”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [71] Rusinova I, Interferome v2.0: an updated database of annotated interferon-regulated genes. Nucleic Acids Res. 41, D1040– 6 (2013). 23203888 PMC3531205 source

      Not checked in the free run.

  • May combine two claims.

    Findingown data

    All 39 csuWGCNA modules were preserved in the signed networks with strong cell type and GWAS enrichments.

    “all preserved in the signed networks with strong cell type and GWAS enrichments”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    geneM23/isoM19 contained hubs GAD1, GAD2, multiple GABA transporters, interneuron markers RELN and VIP, and DLX1 and DLX6-AS1.

    “This module pair contained as hubs the two major GABA synthesizing enzymes (GAD1, GAD2), multiple GABA transporters (SLC6A1, SLC24A3), many other known interneuron markers (RELN, VIP), as well as DLX1 and the lncRNA DLX6-AS1, both critical known regulators of inhibitory neuron development (62).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [62] Bond AM, Balanced gene regulation by an embryonic brain ncRNA is critical for adult hippocampal GABA circuitry. Nat. Neurosci 12, 1020– 1027 (2009). 19620975 PMC3203213 source

      Not checked in the free run.

  • May combine two claims.

    Findingown data

    geneM21/isoM30 was associated with genes downregulated by chronic high-dose haloperidol and genes upregulated by fluoxetine.

    “Indeed, geneM21/isoM30 was associated with genes downregulated by chronic high-doses (but not low-doses) of haloperidol, as well as genes upregulated by the antidepressant fluoxetine (Fig S11A).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The NFkB module pattern is clearly distinct from autism spectrum disorder, which shows a dynamic trajectory but remains upregulated throughout.

    “This pattern is clearly distinct from ASD, which shows a dynamic trajectory, but remains upregulated throughout (Fig 7F).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The paper provides a table of putative mRNA targets for brain-expressed lncRNAs, including 209 differentially expressed in autism spectrum disorder, 122 in bipolar disorder, and 241 in schizophrenia.

    “We provide a table of putative mRNA targets for these brain expressed lncRNAs, including 209 exhibiting DE in ASD, 122 in BD and 241 in SCZ (Table S6).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The gene module geneM7 is enriched for synaptic and metabolic processes and includes the splicing regulator NOVA2.

    “including geneM7, enriched for synaptic and metabolic processes with the splicing regulator NOVA2 (Fig 5C).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Increased resolution allowed identification of modules enriched in inhibitory interneuron markers, including geneM23/isoM19, which was downregulated in ASD and SCZ, with a trend in BD.

    “The increased resolution here allowed us to identify several modules enriched in inhibitory interneuron markers (Fig 5B), including geneM23/isoM19, which is downregulated in ASD and SCZ, with a trend toward downregulation observed in BD;”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingcitation

    Several neuronal modules differentiated BD and SCZ from ASD, including geneM21/isoM30, which captures activity-dependent neuronal gene regulation with hubs including early-response genes ARC, EGR1, NPAS4, NR4A1 and late-response genes BDNF, HOMER1.

    “Several neuronal modules that distinguish between the disorders differentiate BD and SCZ from ASD, including the module pair geneM21/isoM30 (Fig 5C), which captures known elements of activity-dependent neuronal gene regulation, whose hubs include classic early-response (ARC, EGR1, NPAS4, NR4A1) and late-response genes (BDNF, HOMER1) (40).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [40] West AE, Greenberg ME, Neuronal activity-regulated gene transcription in synapse development and cognitive function. Cold Spring Harb. Perspect. Biol 3 (2011), doi: 10.1101/cshperspect.a005744. PMC3098681 21555405 source

      Not checked in the free run.

  • May combine two claims.

    Findingown data

    A large proportion of disease-associated co-expression modules are enriched for cell type specific markers, as is overall disease DE signal.

    “A large proportion of disease-associated co-expression modules are enriched for cell type specific markers, as is overall disease DE signal”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    These switch isoforms show markedly different expression patterns and belong to different co-expression modules, geneM3/isoM1 and isoM18.

    “These switch isoforms show markedly different expression patterns, belonging to different co-expression modules, geneM3/isoM1 (Fig 7C) and isoM18”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Isoforms of NIPBL are down and upregulated in autism spectrum disorder and schizophrenia, respectively.

    “are down and upregulated in ASD and SCZ, respectively (Fig S17).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Isoforms of SMARCA2 are up and downregulated in autism spectrum disorder and schizophrenia, respectively.

    “are up and downregulated in ASD and SCZ, respectively (Fig S17).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    LINC00643 and LINC1166 are present in both the cell nucleus and cytoplasm.

    “present both in the cell nucleus and cytoplasm (Fig 8A; Fig S16)”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Genes with switch isoforms are enriched for regulatory targets of CHD8 and FMRP, highly mutationally constrained genes (pLI>0.99), syndromic autism spectrum disorder genes, and genes with de novo exonic mutations in autism spectrum disorder, schizophrenia and bipolar disorder.

    “Genes with switch isoforms are also enriched for the regulatory targets of two ASD risk genes, CHD8 and FMRP, as well as highly mutationally constrained genes (pLI>0.99), syndromic ASD genes, and in genes with de novo exonic mutations in ASD, SCZ and BD (Fig 8F; Table S7; (21)).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [21] See supplemental methods.

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Findingown data

    LINC01166 is significantly downregulated in bipolar disorder (log2FC −0.17, FDR<0.05) with trends in autism spectrum disorder and schizophrenia (FDR<0.1).

    “whereas LINC01166 is significantly downregulated in BD (log2FC −0.17, FDR<0.05) with trends in ASD and SCZ (FDR’s < 0.1).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    geneM3/isoM1 and isoM18 are enriched in astrocyte and neuronal cell types, respectively.

    “which are enriched in astrocyte and neuronal cell types, respectively (Fig 5A; Fig S12).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Most of these ncRNAs show significant CNS enrichment but have limited functional annotation.

    “most of these ncRNAs show significant CNS enrichment, but until now, have limited functional annotation.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    A module enriched for microglial-associated genes shows strong upregulation in autism spectrum disorder and significant downregulation in schizophrenia and bipolar disorder.

    “A module enriched for microglial-associated genes, for example, shows a clear distinction between disorders, with strong upregulation observed on ASD and significant downregulation in SCZ and BD.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Regulatory relationships between non-coding variants and their target genes vary across human cell types, tissues, developmental stages, and species.

    “Such regulatory relationships show substantial heterogeneity across human cell types, tissues, and developmental stages (10), and are often highly species-specific (11).”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [10] GTEx Consortium, Human genomics. The Genotype-Tissue Expression (GTEx) pilot analysis: multitissue gene regulation in humans. Science. 348, 648– 660 (2015). 25954001 PMC4547484 source

      Not checked in the free run.

    • cites [11] Reilly SK, Evolutionary genomics. Evolutionary changes in promoter and enhancer activity during human corticogenesis. Science. 347, 1155– 1159 (2015). 25745175 PMC4426903 source

      Not checked in the free run.

  • May combine two claims.

    Study design and conductown data

    The study combined RNA-seq data from eight studies through a uniform processing pipeline, yielding 2188 samples that passed QC, from 1695 unique subjects, including 279 technical replicates.

    “Data was generated across eight studies (2, 19, 20), uniformly processed, and combined through a consolidated genomic data processing pipeline ((21); Fig S1), yielding a total of 2188 samples passing quality control (QC) for this analysis, representing frontal and temporal cerebral cortex from 1695 unique subjects across the lifespan, including 279 technical replicates (Fig S2).”
    Quote in the paperSubstantial claim0 of 4 sources checked
    • cites [2] Fromer M, Gene expression elucidates functional impact of polygenic risk for schizophrenia. Nat. Neurosci 19, 1442– 1453 (2016). 27668389 PMC5083142 source

      Not checked in the free run.

    • cites [19] Parikshak NN, Genome-wide changes in lncRNA, splicing, and regional gene expression patterns in autism. Nature. 540, 423– 427 (2016). 27919067 PMC7102905 source

      Not checked in the free run.

    • cites [20] Kang HJ, Spatio-temporal transcriptome of the human brain. Nature. 478, 483– 489 (2011). 22031440 PMC3566780 source

      Not checked in the free run.

    • cites [21] See supplemental methods.

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Hundreds of genetic variants are robustly associated with ASD, SCZ, and BD, and thousands more likely contribute to their pathogenesis.

    “Studies have identified hundreds of causal genetic variants robustly associated with these disorders, and thousands more that likely contribute to their pathogenesis (4).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [4] Gandal MJ, Leppa V, Won H, Parikshak NN, Geschwind DH, The road to precision psychiatry: translating genetics into disease mechanisms. Nat. Neurosci 19, 1397– 1407 (2016). 27786179 PMC9012265 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    Quality control detected 16,541 protein-coding and 9,233 non-coding genes using Gencode v19 annotations.

    “Extensive quality control steps were taken within and across individual studies resulting in the detection of 16,541 protein-coding and 9,233 non-coding genes with the Gencode v19 annotations ((21); Fig S3).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [21] See supplemental methods.

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Previous studies identified alterations in local splicing events in ASD and in smaller cohorts in SCZ and BD.

    “Previous studies have identified alterations in local splicing events in ASD (19, 30) and in smaller cohorts in SCZ (2, 24) and BD (31).”
    Quote in the paperNot a substantial claim0 of 5 sources checked
    • cites [19] Parikshak NN, Genome-wide changes in lncRNA, splicing, and regional gene expression patterns in autism. Nature. 540, 423– 427 (2016). 27919067 PMC7102905 source

      Not checked in the free run.

    • cites [30] Irimia M, A highly conserved program of neuronal microexons is misregulated in autistic brains. Cell. 159, 1511– 1523 (2014). 25525873 PMC4390143 source

      Not checked in the free run.

    • cites [2] Fromer M, Gene expression elucidates functional impact of polygenic risk for schizophrenia. Nat. Neurosci 19, 1442– 1453 (2016). 27668389 PMC5083142 source

      Not checked in the free run.

    • cites [24] Jaffe AE, Developmental and genetic regulation of the human cortex transcriptome illuminate schizophrenia pathogenesis. Nat. Neurosci 21, 1117– 1125 (2018). 30050107 PMC6438700 source

      Not checked in the free run.

    • cites [31] Akula N, RNA-sequencing of the brain transcriptome implicates dysregulation of neuroplasticity, circadian rhythms and GTPase binding in bipolar disorder. Mol. Psychiatry 19, 1179– 1185 (2014). 24393808 PMC5560442 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductcitation

    Surrogate variable correction was used to account for unmeasured confounders, a standard approach.

    “We used surrogate variable correction in our analyses to account for such unmeasured confounders (43), which is a standard approach (44).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [43] Leek JT, Storey JD, Capturing heterogeneity in gene expression studies by surrogate variable analysis. PLoS Genet. 3, 1724– 1735 (2007). 17907809 PMC1994707 source

      Not checked in the free run.

    • cites [44] GTEx Consortium, Genetic effects on gene expression across human tissues. Nature. 550, 204 (2017). 29022597 PMC5776756 source

      Not checked in the free run.

  • May combine two claims.

    Interpretationown data

    DS showed greatest overlap with the ncRNA signal, suggesting non-coding genes regulate local splicing events.

    “Notably, DS exhibited greatest overlap with the ncRNA signal, suggesting a role for non-coding genes in regulating local splicing events.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductown data

    Local differential splicing was assessed with LeafCutter using de novo aligned RNA-seq reads and the same covariates as DGE/DTE.

    “We used LeafCutter (29) to assess local differential splicing (DS) differences in ASD, SCZ and BD using de novo aligned RNA-seq reads, controlling for the same covariates as DGE/DTE (Fig S7).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [29] Li YI, Annotation-free quantification of RNA splicing using LeafCutter. Nat. Genet 50, 151– 158 (2018). 29229983 PMC5742080 source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Mouse experiments indicate RBFOX1 has distinct nuclear and cytoplasmic isoforms with differing functions.

    “Experiments in mouse indicate that RBFOX1 has distinct nuclear and cytoplasmic isoforms with differing functions, the nuclear isoform primarily regulating pre-mRNA alternative splicing, and the cytoplasmic isoform binding to the 3’ UTR to stabilize target transcripts involved in regulation of neuronal excitability (28, 32, 58, 60).”
    Quote in the paperSubstantial claim0 of 4 sources checked
    • cites [28] Vuong CK, Black DL, Zheng S, The neurogenetics of alternative splicing. Nat. Rev. Neurosci 17, 265– 281 (2016). 27094079 PMC4861142 source

      Not checked in the free run.

    • cites [32] Lee J-A, Cytoplasmic Rbfox1 Regulates the Expression of Synaptic and Autism-Related Genes. Neuron. 89, 113– 128 (2016). 26687839 PMC4858412 source

      Not checked in the free run.

    • cites [58] Vuong CK, Rbfox1 Regulates Synaptic Transmission through the Inhibitory Neuron-Specific vSNARE Vamp1. Neuron. 98, 127– 141.e7 (2018). 29621484 PMC5890944 source

      Not checked in the free run.

    • cites [60] Gehman LT, The splicing regulator Rbfox1 (A2BP1) controls neuronal excitation in the mammalian brain. Nat. Genet 43, 706– 711 (2011). 21623373 PMC3125461 source

      Not checked in the free run.

  • May combine two claims.

    Interpretationown data

    Expression of HCG17, HCG23, and HLA-C was also correlated with imputed C4A copy number, suggesting pleiotropic effects.

    “However, expression of all three was also significantly (P<0.05) correlated with imputed C4A copy number, suggesting pleiotropic effects.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Interpretationreasoning

    The two BD PRS-associated isoforms NCALD and SNF8 were non-significant in TWAS, perhaps due to lack of a nearby genome-wide significant locus or isoform-specific regulation, suggesting trans-acting factors.

    “The two isoforms with PRS associations in BD (NCALD, SNF8) were non-significant in TWAS, perhaps due to lack of a nearby genome-wide significant locus or isoform-specific regulation, suggesting those expression changes may be driven by trans-acting factors.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    Large sample size and isoform-level specificity enabled identification of synaptic modules containing unique isoforms with distinct disease associations and separation of excitatory and inhibitory signals.

    “However, the large sample size, coupled with the specificity of isoform-level qualifications, enabled us to identify synaptic modules containing unique isoforms with distinct disease associations and to separate signals from excitatory and inhibitory neurons (Fig 5B).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    The TWAS analysis provides a core set of strong candidate genes implicated by risk loci and a mechanistic basis for composite activity of disease risk variants.

    “Overall, this analysis provides a core set of strong candidate genes implicated by risk loci, and provides a mechanistic basis for the composite activity of disease risk variants.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Previous work identified downregulated neuronal modules in ASD and SCZ containing RBFOX1 as a hub.

    “Previous work has identified downregulated neuronal modules in ASD and SCZ containing RBFOX1 as a hub (1, 16).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [1] Gandal MJ, Shared molecular neuropathology across major psychiatric disorders parallels polygenic overlap. Science. 359, 693– 697 (2018). 29439242 PMC5898828 source

      Not checked in the free run.

    • cites [16] Voineagu I, Transcriptomic analysis of autistic brain reveals convergent molecular pathology. Nature. 474, 380– 384 (2011). 21614001 PMC3607626 source

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Premisecitation

    Prior work has identified differential activation of glial and neural-immune processes in brains of patients with psychiatric disorders.

    “Previous work has identified differential activation of glial and neural-immune processes in brain from patients with psychiatric disorders (16, 52, 57, 66–69)”
    Quote in the paperSubstantial claim0 of 7 sources checked
    • cites [16] Voineagu I, Transcriptomic analysis of autistic brain reveals convergent molecular pathology. Nature. 474, 380– 384 (2011). 21614001 PMC3607626 source

      Not checked in the free run.

    • cites [52] Chen C, Two gene co-expression modules differentiate psychotics and controls. Mol. Psychiatry 18, 1308– 1314 (2013). 23147385 PMC4018461 source

      Not checked in the free run.

    • cites [57] Ramaker RC, Post-mortem molecular profiling of three psychiatric disorders. Genome Med. 9, 72 (2017). 28754123 PMC5534072 source

      Not checked in the free run.

    • cites [66] Birnbaum R, Investigating the neuroimmunogenic architecture of schizophrenia. Mol. Psychiatry 23, 1251– 1260 (2018). 28485405 source

      Not checked in the free run.

    • cites [67] Fillman SG, Increased inflammatory markers identified in the dorsolateral prefrontal cortex of individuals with schizophrenia. Mol. Psychiatry 18, 206– 214 (2013). 22869038 source

      Not checked in the free run.

    • cites [68] Pacifico R, Davis RL, Transcriptome sequencing implicates dorsal striatum-specific gene network, immune response and energy metabolism pathways in bipolar disorder. Mol. Psychiatry 22, 441– 449 (2017). 27350034 source

      Not checked in the free run.

    • cites [69] Rao JS, Harry GJ, Rapoport SI, Kim HW, Increased excitotoxicity and neuroinflammatory markers in postmortem frontal cortex from bipolar disorder patients. Mol. Psychiatry 15, 384– 392 (2010). 19488045 PMC2844920 source

      Not checked in the free run.

  • May combine two claims.

    Interpretationown data

    Although these modules were not significantly downregulated in ASD, subsampling indicates differences between disorders could be driven by sample size.

    “Although these modules were not significantly downregulated in ASD, subsampling indicates that the differences between disorders could be driven by sample size (Fig S14).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    DLX6-AS1 is a known development regulator of interneuron specification.

    “the lncRNA DLX6-AS1, a known development regulator of interneuron specification (62)”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [62] Bond AM, Balanced gene regulation by an embryonic brain ncRNA is critical for adult hippocampal GABA circuitry. Nat. Neurosci 12, 1020– 1027 (2009). 19620975 PMC3203213 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Prior studies found both microglia and astrocytes upregulated in autism spectrum disorder.

    “and both microglia and astrocytes in ASD (19, 70).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [19] Parikshak NN, Genome-wide changes in lncRNA, splicing, and regional gene expression patterns in autism. Nature. 540, 423– 427 (2016). 27919067 PMC7102905 source

      Not checked in the free run.

    • cites [70] Gupta S, Transcriptome analysis reveals dysregulation of innate immune response genes and neuronal activity-dependent genes in autism. Nat. Commun 5, 5748 (2014). 25494366 PMC4270294 source

      Not checked in the free run.

  • May combine two claims.

    Interpretationown data

    Two additional immune-related modules appear to represent broadly expressed signaling pathways: interferon response (geneM32) and NFkB (geneM5/isoM5).

    “Two additional immune-related modules appear to represent more broadly expressed signaling pathways: interferon response (geneM32) and NFkB (geneM5/isoM5).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    A nearly identical module was recently identified as a sex-specific transcriptional signature of major depression and stress susceptibility.

    “Remarkably, a nearly identical module was recently identified as a sex-specific transcriptional signature of major depression and stress susceptibility (64).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [64] Labonté B, Sex-specific transcriptional signatures in human depression. Nat. Med 23, 1102– 1111 (2017). 28825715 PMC5734943 source

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Premisecitation

    A subset of lncRNAs are thought to function by repressing mRNA targets.

    “As a subset of lncRNAs are thought to function by repressing mRNA targets (72)”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [72] Necsulea A, The evolution of lncRNA repertoires and expression patterns in tetrapods. Nature. 505, 635– 640 (2014). 24463510 source

      Not checked in the free run.

  • May combine two claims.

    Interpretationown data

    The NFkB module maximally diverges from controls during early adulthood, coincident with typical disease onset in schizophrenia and bipolar disorder around age 25.

    “maximally diverges from controls during early adulthood, coincident with typical disease onset in SCZ and BD (~25).”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Cross-sectional post-mortem expression trajectories across age reflect disease progression and timing relative to clinical onset.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Prior studies found astrocyte upregulation in schizophrenia and bipolar disorder.

    “including upregulation of astrocytes in SCZ and BD (1, 57)”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [1] Gandal MJ, Shared molecular neuropathology across major psychiatric disorders parallels polygenic overlap. Science. 359, 693– 697 (2018). 29439242 PMC5898828 source

      Not checked in the free run.

    • cites [57] Ramaker RC, Post-mortem molecular profiling of three psychiatric disorders. Genome Med. 9, 72 (2017). 28754123 PMC5534072 source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    C4A is the top GWAS-supported and strongly upregulated risk gene for schizophrenia.

    “the top GWAS-supported, and strongly upregulated, risk gene for SCZ (5).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [5] Sekar A, Schizophrenia risk from complex variation of complement component 4. Nature. 530, 177– 183 (2016). 26814963 PMC4752392 source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    LINC00473 is regulated by synaptic activity and downregulates immediate early gene expression.

    “LINC00473 is regulated by synaptic activity and downregulates immediate early gene expression (74)”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [74] Pruunsild P, Bengtson CP, Bading H, Networks of Cultured iPSC-Derived Neurons Reveal the Human Synaptic Activity-Regulated Adaptive Gene Program. Cell Rep. 18, 122– 135 (2017). 28052243 PMC5236011 source

      Not checked in the free run.

  • May combine two claims.

    Interpretationown data

    The IFN-response and microglial modules peak during early development, coincident with clinical onset.

    “peaking during early development, coincident with clinical onset.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Cross-sectional post-mortem expression trajectories across age reflect disease progression and timing relative to clinical onset.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Evidence supports hyperactive complement-mediated synaptic pruning in schizophrenia pathophysiology, presumably through microglia.

    “Evidence supports hyperactive complement-mediated synaptic pruning in SCZ pathophysiology, presumably through microglia (5)”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [5] Sekar A, Schizophrenia risk from complex variation of complement component 4. Nature. 530, 177– 183 (2016). 26814963 PMC4752392 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    These genes play critical roles in regulating synaptic plasticity and the balance of excitatory and inhibitory synapses.

    “These genes play critical roles in regulating synaptic plasticity and the balance of excitatory and inhibitory synapses (40).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [40] West AE, Greenberg ME, Neuronal activity-regulated gene transcription in synapse development and cognitive function. Cold Spring Harb. Perspect. Biol 3 (2011), doi: 10.1101/cshperspect.a005744. PMC3098681 21555405 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    NIPBL is associated with Cornelia de Lange Syndrome.

    “NIPBL, a gene associated with Cornelia de Lange Syndrome (80)”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [80] Krantz ID, Cornelia de Lange syndrome is caused by mutations in NIPBL, the human homolog of Drosophila melanogaster Nipped-B. Nat. Genet 36, 631– 635 (2004). 15146186 PMC4902017 source

      Not checked in the free run.

  • May combine two claims.

    Interpretationown data

    These data confirm the importance of microexon regulation in neuropsychiatric disorders beyond autism spectrum disorder and its potential role in distinguishing biological pathways differentially affected across conditions.

    “These data confirm the importance of microexon regulation in neuropsychiatric disorders beyond ASD, and their potential role in distinguishing among biological pathways differentially affected across conditions.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Brain alternative splicing patterns are more distinct than in other organ systems.

    “alternative splicing patterns are more distinct than in other organ systems (82).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [82] Melé M, Human genomics. The human transcriptome across tissues and individuals. Science. 348, 660– 665 (2015). 25954002 PMC4547472 source

      Not checked in the free run.

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductnone given

    The study attempted to mitigate factors contributing to gene expression changes through known and hidden covariate correction, assessment of age trajectories, and enrichment for causal genetic variation.

    “At each step of analysis, we have attempted to mitigate the contribution of these factors through known and hidden covariate correction, assessment of age trajectories, and via enrichment for causal genetic variation.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation

    Several studies have demonstrated that genes carrying microexons are preferentially expressed in brain and their splicing is dysregulated in autism spectrum disorder.

    “several studies have demonstrated that genes carrying microexons are preferentially expressed in brain and their splicing is dysregulated in ASD (30, 82, 83).”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites [30] Irimia M, A highly conserved program of neuronal microexons is misregulated in autistic brains. Cell. 159, 1511– 1523 (2014). 25525873 PMC4390143 source

      Not checked in the free run.

    • cites [82] Melé M, Human genomics. The human transcriptome across tissues and individuals. Science. 348, 660– 665 (2015). 25954002 PMC4547472 source

      Not checked in the free run.

    • cites [83] Quesnel-Vallières M, Misregulation of an Activity-Dependent Splicing Network as a Common Mechanism Underlying Autism Spectrum Disorders. Mol. Cell 64, 1023– 1034 (2016). 27984743 source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    SOX2-OT is a heavily spliced, evolutionarily conserved lncRNA with predominant brain expression and is a hub of these oligodendrocyte modules.

    “SOX2-OT is a heavily spliced, evolutionarily-conserved lncRNA exhibiting predominant brain expression and a hub of these oligodendrocyte modules, without previous mechanistic links to myelination (75, 76).”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [75] Amaral PP, Complex architecture and regulated expression of the Sox2ot locus during vertebrate development. RNA. 15, 2013– 2027 (2009). 19767420 PMC2764477 source

      Not checked in the free run.

    • cites [76] Mercer TR, Dinger ME, Sunkin SM, Mehler MF, Mattick JS, Specific expression of long noncoding RNAs in the mouse brain. Proc. Natl. Acad. Sci. U. S. A 105, 716– 721 (2008). 18184812 PMC2206602 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Interpretationreasoning

    The modest correlations likely vary by transcript class and some are likely due to low abundance genes such as ncRNAs.

    “This likely varies by transcript class, and some of the modest correlations are likely due to low abundance genes, such as ncRNAs”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Premisenone given

    Previous studies observed broad shared patterns of gene expression dysregulation in post-mortem brain, including downregulation of neuronal and synaptic signaling genes and upregulation of glial-immune or neuroinflammatory signals.

    “Several broad shared patterns of gene expression dysregulation have been observed in post mortem brain in previous studies, most prominently, a gradient of downregulation of neuronal and synaptic signaling genes, and upregulation of glial-immune or neuroinflammatory signals.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    SMARCA2 is a member of the BAF-complex strongly implicated in several neurodevelopmental disorders including autism spectrum disorder.

    “SMARCA2, a member of the BAF-complex strongly implicated in several neurodevelopmental disorders including ASD (79)”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [79] Koga M, Involvement of SMARCA2/BRM in the SWI/SNF chromatin-remodeling complex in schizophrenia. Hum. Mol. Genet 18, 2483– 2494 (2009). 19363039 source

      Not checked in the free run.

  • May combine two claims.

    Study design and conductown data

    The study performed a well-powered brain-relevant TWAS in schizophrenia, and to a lesser extent in bipolar disorder and autism spectrum disorder.

    “by performing a well-powered brain-relevant TWAS in SCZ, and to a lesser extent in BD and ASD”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    ANK2 is an autism spectrum disorder risk gene.

    “the ASD risk gene ANK2 (81)”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [81] Sanders SJ, Insights into Autism Spectrum Disorder Genomic Architecture and Biology from 71 Risk Loci. Neuron. 87, 1215– 1233 (2015). 26402605 PMC4624267 source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    QKI is an established regulator of oligodendrocyte-gene splicing and is located in this module.

    “an established regulator of oligodendrocyte-gene splicing also located in this module (77, 78).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [77] Aberg K, Saetre P, Jareborg N, Jazin E, Human QKI, a potential regulator of mRNA expression of human oligodendrocyte-related genes involved in schizophrenia. Proc. Natl. Acad. Sci. U. S. A 103, 7482– 7487 (2006). 16641098 PMC1464365 source

      Not checked in the free run.

    • cites [78] Barry G, The long non-coding RNA Gomafu is acutely regulated in response to neuronal activation and involved in schizophrenia-associated alternative splicing. Mol. Psychiatry 19, 486– 494 (2014). 23628989 source

      Not checked in the free run.

  • May combine two claims.

    Interpretationcitation

    These analyses predict critical roles for these non-coding genes in oligodendrocyte function and potentially in psychiatric conditions.

    “These analyses predict critical roles for these or these often overlooked non-coding genes in oligodendrocyte function (77, 78) and potentially in psychiatric conditions.”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [77] Aberg K, Saetre P, Jareborg N, Jazin E, Human QKI, a potential regulator of mRNA expression of human oligodendrocyte-related genes involved in schizophrenia. Proc. Natl. Acad. Sci. U. S. A 103, 7482– 7487 (2006). 16641098 PMC1464365 source

      Not checked in the free run.

    • cites [78] Barry G, The long non-coding RNA Gomafu is acutely regulated in response to neuronal activation and involved in schizophrenia-associated alternative splicing. Mol. Psychiatry 19, 486– 494 (2014). 23628989 source

      Not checked in the free run.

  • May combine two claims.

    Interpretationreasoning

    These interactions are likely cell type specific and suggest distinct functions of ANK2 isoforms in different neural cell types and diseases.

    “likely cell type specific interactions that suggest distinct functions of the isoforms of this gene in different neural cell types and diseases.”
    Quote in the paperSubstantial claim
102 of 102

  • The paper relies on this without support.

    Post-mortem brain transcriptomes from these cohorts are a valid proxy for the molecular pathology of ASD, SCZ, and BD.

    “We present results of the analysis of RNA-sequencing (RNA-Seq) data from the PsychENCODE Consortium (17), integrating genetic and genomic data from over 2000 well-curated, high-quality post-mortem brain samples from individuals with SCZ, BD, ASD, and controls (18).”

  • The paper relies on this without support.

    The cell-type marker sets used for enrichment and deconvolution accurately identify the corresponding brain cell types.

    “These DE genes exhibited substantial enrichment for known pathways and cell type specific markers derived from single nucleus RNA-Seq in human brain (Fig 1D-E) (21), consistent with previously observed patterns (1, 19).”

  • The paper relies on this without support.

    Cross-sectional post-mortem expression trajectories across age reflect disease progression and timing relative to clinical onset.

    “peaking during early development, coincident with clinical onset.”
    “maximally diverges from controls during early adulthood, coincident with typical disease onset in SCZ and BD (~25).”

  • The paper relies on this without support.

    Negative correlation between an lncRNA and an mRNA within a co-expression module indicates a regulatory repression relationship.

    “As a subset of lncRNAs are thought to function by repressing mRNA targets (72), we applied csuWGCNA (73) to identify potential regulatory relationships (21).”
    “which captured 7186 negatively correlated lncRNA-mRNA pairs within the same module (Fig S15).”

  • The paper relies on this without support.

    Enrichment of a co-expression module for GWAS signals supports a causal role for that module in disease.

    “This neuronal module was significantly enriched for both BD and SCZ GWAS signals, supporting a causal role for this module.”
5 of 5

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • DLX5/6-N-Haplotype includes enhancers controlling gene expression in the central nervous system and sequences associated with hyper- or hypo-socialization/vocalization pathologies such as Williams-Beuren syndrome or ASDs.

    “that includes enhancers controlling gene expression in the central nervous system (Fazel Darbandi, et al. 2016) and sequences associated with either hyper- or hypo-socialization/vocalization pathologies such as Williams-Buren syndrome (Poitras, et al. 2010) or ASDs (Gandal, et al. 2018).”

    From <i>DLX5/6</i> GABAergic Expression Affects Social Vocalization: Implications for Human Evolution (2021) · cites this paper as Gandal, et al. 2018

    In “Non-coding Modules and lncRNA Regulatory Relationships” (page unavailable), the paper identifies DLX6-AS1 as a hub of an interneuron module downregulated in ASD. That supports a connection to ASD, but the cited passage does not identify a DLX5/6-N haplotype or link its sequences to socialization or vocalization pathologies. The other conditions and enhancer claim have separate citations in the citing sentence. (LLM, conf=0.84)

    Experimental — accuracy not yet measured

  • Modules including DLX genes and other key genetic components of GABAergic neurons constitute the major hub for both ASD and schizophrenia.

    “the modules including these DLX genes and other key genetic components of GABAergic neurons constitute the major hub for both ASD and schizophrenia (Gandal, et al. 2018).”

    From <i>DLX5/6</i> GABAergic Expression Affects Social Vocalization: Implications for Human Evolution (2021) · cites this paper as Gandal, et al. 2018

  • TWAS identified 708 schizophrenia-associated genes whose imputed expression values correlate with schizophrenia at FDR < 0.05.

    “TWAS identified 708 SCZ-associated genes (TWAS SCZ genes) whose imputed expression values are correlated with SCZ (FDR<0.05).”

    From A computational tool (H-MAGMA) for improved prediction of brain-disorder risk genes by incorporating brain chromatin interaction profiles (2020) · cites this paper as [40]

  • Differentially expressed genes in postmortem brains with psychiatric disorders at FDR < 0.05 were obtained from Gandal et al., 2018.

    “Differentially expressed genes (DEG) in postmortem brains with psychiatric disorders were obtained from Gandal et al., 201840 (FDR<0.05).”

    From A computational tool (H-MAGMA) for improved prediction of brain-disorder risk genes by incorporating brain chromatin interaction profiles (2020) · cites this paper as [40]

  • The authors identified noncoding regions of transcription and sets of differentially expressed and coexpressed genes.

    “Using these data, we identified noncoding regions of transcription and sets of differentially expressed and coexpressed genes (21, 22).”

    From Comprehensive functional genomic resource and integrative model for the human brain (2018) · cites this paper as [22]

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Disease-associated module categories total 63, not 61

    Minor

    The stated categories contain 5, 22, and 36 modules, which sum to 63 rather than the reported 61 disease-associated modules.

    Reported 61 · recomputed 63

  • Eight reported BD genes are followed by eleven names

    Minor

    The text says eight genes showed consistent association, but the accompanying list names eleven.

2 of 2

Cites

Flagged first, then by citations

Research works printed but not linked: 2

  1. A. E. Jaffe et al . Developmental And Genetic Regulation Of The Human Cortex Transcriptome In Schizophrenia. bioRxiv 124321 [Preprint]. 22 November 2017. .10.1101/124321 · DOI
  2. PEC_DAC_RNAseq.py doi:.10.7303/syn12026837.1 · DOI

Cited by

Sorted by citations

Related

Authors
Gandal, M. J., Zhang, P., Hadjimichael, E., Walker, R. L., Chen, C., Liu, S., Won, H., van Bakel, H., Varghese, M., Wang, Y., Shieh, A. W., Haney, J., Parhami, S., Belmont, J., Kim, M., Moran Losada, P., Khan, Z., Mleczko, J., Xia, Y., Dai, R., Wang, D., Yang, Y. T., Xu, M., Fish, K., Hof, P. R., Warrell, J., Fitzgerald, D., White, K., Jaffe, A. E., PsychENCODE Consortium, Peters, M. A., Gerstein, M., Liu, C., Iakoucheva, L. M., Pinto, D., Geschwind, D. H., Ashley-Koch, A. E., Crawford, G. E., Garrett, M. E., Song, L., Safi, A., Johnson, G. D., Wray, G. A., Reddy, T. E., Goes, F. S., Zandi, P., Bryois, J., Jaffe, A. E., Price, A. J., Ivanov, N. A., Collado-Torres, L., Hyde, T. M., Burke, E. E., Kleiman, J. E., Tao, R., Shin, J. H., Akbarian, S., Girdhar, K., Jiang, Y., Kundakovic, M., Brown, L., Kassim, B. S., Park, R. B., Wiseman, J. R., Zharovsky, E., Jacobov, R., Devillers, O., Flatow, E., Hoffman, G. E., Lipska, B. K., Lewis, D. A., Haroutunian, V., Hahn, C.-G., Charney, A. W., Dracheva, S., Kozlenkov, A., Belmont, J., DelValle, D., Francoeur, N., Hadjimichael, E., Pinto, D., van Bakel, H., Roussos, P., Fullard, J. F., Bendl, J., Hauberg, M. E., Mangravite, L. M., Peters, M. A., Chae, Y., Peng, J., Niu, M., Wang, X., Webster, M. J., Beach, T. G., Chen, C., Jiang, Y., Dai, R., Shieh, A. W., Liu, C., Grennan, K. S., Xia, Y., Vadukapuram, R., Wang, Y., Fitzgerald, D., Cheng, L., Brown, M., Brown, M., Brunetti, T., Goodman, T., Alsayed, M., Gandal, M. J., Geschwind, D. H., Won, H., Polioudakis, D., Wamsley, B., Yin, J., Hadzic, T., De La Torre Ubieta, L., Swarup, V., Sanders, S. J., State, M. W., Werling, D. M., An, J.-Y., Sheppard, B., Willsey, A. J., White, K. P., Ray, M., Giase, G., Kefi, A., Mattei, E., Purcaro, M., Weng, Z., Moore, J., Pratt, H., Huey, J., Borrman, T., Sullivan, P. F., Giusti-Rodriguez, P., Kim, Y., Sullivan, P., Szatkiewicz, J., Rhie, S. K., Armoskus, C., Camarena, A., Farnham, P. J., Spitsyna, V. N., Witt, H., Schreiner, S., Evgrafov, O. V., Knowles, J. A., Gerstein, M., Liu, S., Wang, D., Navarro, F. C. P., Warrell, J., Clarke, D., Emani, P. S., Gu, M., Shi, X., Xu, M., Yang, Y. T., Kitchen, R. R., Gürsoy, G., Zhang, J., Carlyle, B. C., Nairn, A. C., Li, M., Pochareddy, S., Sestan, N., Skarica, M., Li, Z., Sousa, A. M. M., Santpere, G., Choi, J., Zhu, Y., Gao, T., Miller, D. J., Cherskov, A., Yang, M., Amiri, A., Coppola, G., Mariani, J., Scuderi, S., Szekely, A., Vaccarino, F. M., Wu, F., Weissman, S., Roychowdhury, T., & Abyzov, A.
Affiliations
  • University of California San Diego (US)
  • University of California, Los Angeles (US)
  • Icahn School of Medicine at Mount Sinai (US)
  • Department of Psychiatry, University of California San Diego, 9500 Gilman Dr., La Jolla, CA 92093, USA.
  • Department of Human Genetics, University of Chicago, Chicago, IL 60637, USA.
  • Sage Bionetworks (US)
  • Yale University (US)
  • Stony Brook University (US)
  • SUNY Upstate Medical University (US)
Corresponding author
Iakoucheva, Lilia M.
ORCID
Venue
Science
Publisher
American Association for the Advancement of Science (AAAS)
Published
2018-12-14
Volume
362
Issue
6420
ISSN
  • 0036-8075 (ISSN-L)
  • 1095-9203
DOI
10.1126/science.aat8127
PMC
PMC6443102
OpenAlex
W2904884010
Semantic Scholar
8841331ef6fa66e8859f0a5e26e240c3f57a16e9
Cited by
1,247
References
147
Citations by year
2026: 126 · 2025: 145 · 2024: 168 · 2023: 166 · 2022: 176 · 2021: 201 · 2020: 162 · 2019: 92 · 2018: 4 · 2017: 1
FWCI
69.69
Type
journal article
Language
en
Open access
Locations
  • Science (publishedVersion)
  • PubMed (publishedVersion)
  • Europe PMC (PubMed Central) (acceptedVersion)
  • PubMed Central (acceptedVersion)
  • eScholarship (California Digital Library) (submittedVersion)
Funders
  • National Institute of Mental Health
  • NIH Office of the Director (U01MH103340, R01MH110920)
  • Simons Foundation
Topics
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Autism Spectrum Disorder Research
  • Health Sciences › Medicine › Psychiatry and Mental health › Schizophrenia research and treatment
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › RNA Research and Splicing
Keywords
autism spectrum disorder, schizophrenia, bipolar disorder, transcriptome-wide association study, differential splicing, gene co-expression modules, non-coding RNA, genetic risk variants, polygenic risk score, Mendelian randomization, cell type-specific gene expression, blood-brain barrier, microglia, astrocytes, interferon response, immediate early genes, candidate risk genes
MeSH terms
Autism Spectrum Disorder, Bipolar Disorder, Brain, Humans, RNA Splicing, Schizophrenia, Sequence Analysis, RNA, Genetic Predisposition to Disease, Protein Isoforms, Transcriptome

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-06.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-06.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review