CoChat
Sign Up
PreprintCited by 980Schizophrenia research and treatmentGenetic Associations and EpidemiologyGenomics and Rare Diseases

Exome sequencing identifies rare coding variants in 10 genes which confer substantial risk for schizophrenia

Open access
Singh, Tarjinder, Neale, Benjamin M., Daly, Mark J.·2020·Nature
DOI 10.1101/2020.09.18.20192815PMC PMC9805802

Preprint — published as 10.1038/s41586-022-04556-w · Paper page

Open this paper in CoChat — free accountOpen PDF

Abstract

By meta-analyzing the whole-exomes of 24,248 cases and 97,322 controls, we implicate ultra-rare coding variants (URVs) in ten genes as conferring substantial risk for schizophrenia (odds ratios 3 - 50, P < 2.14 × 10 -6 ), and 32 genes at a FDR < 5%. These genes have the greatest expression in central nervous system neurons and have diverse molecular functions that include the formation, structure, and function of the synapse. The associations of NMDA receptor subunit GRIN2A and AMPA receptor subunit GRIA3 provide support for the dysfunction of the glutamatergic system as a mechanistic hypothesis in the pathogenesis of schizophrenia. We find significant evidence for an overlap of rare variant risk between schizophrenia, autism spectrum disorders (ASD), and severe neurodevelopmental disorders (DD/ID), supporting a neurodevelopmental etiology for schizophrenia. We show that protein-truncating variants in GRIN2A, TRIO , and CACNA1G confer risk for schizophrenia whereas specific missense mutations in these genes confer risk for DD/ID. Nevertheless, few of the strongly associated schizophrenia genes appear to confer risk for DD/ID. We demonstrate that genes prioritized from common variant analyses of schizophrenia are enriched in rare variant risk, suggesting that common and rare genetic risk factors at least partially converge on the same underlying pathogenic biological processes. Even after excluding significantly associated genes, schizophrenia cases still carry a substantial excess of URVs, implying that more schizophrenia risk genes await discovery using this approach.

AI snapshot

Written from the paper’s full text.

TL;DR

Meta-analyzing exomes from 24,248 schizophrenia cases, 97,322 controls, and 3,402 trios identified 10 genes with ultra-rare coding variants at exome-wide significance and 32 genes at FDR < 5%. The results implicate synaptic and glutamatergic biology, show shared rare-variant risk with ASD/DD/ID, and suggest partial convergence with common-variant GWAS signals.

Key findings

  • Ten genes reached exome-wide significance (P < 2.14 × 10⁻⁶), with odds ratios for PTVs and Class I variants ranging from 3 to 50; 22 additional genes were significant at FDR < 5%. The 10 genes were SETD1A, CUL1, XPO7, TRIO, CACNA1G, SP4, GRIA3, GRIN2A, HERC1, and RB1CC1.
  • GRIN2A and GRIA3 support glutamatergic dysfunction: GRIN2A Class I OR = 24.1 (95% CI 5.36–221), Class II OR = 2.37 (95% CI 1.1–4.92); GRIA3 Class I OR = 20.1 (95% CI 4.28–188).
  • Schizophrenia rare-variant risk overlaps with neurodevelopmental disorders: enrichment in DD/ID genes (P = 1.5 × 10⁻¹⁴; ORClass I = 1.44, 95% CI 1.3–1.6) and ASD genes (P = 3.7 × 10⁻⁷; ORClass I = 1.45, 95% CI 1.23–1.72). PTVs in GRIN2A, TRIO, and CACNA1G confer schizophrenia risk, whereas specific missense mutations in these genes confer DD/ID risk.
  • Common and rare variant signals partially converge: genes prioritized by GWAS fine-mapping were enriched for ultra-rare variants (Pmeta = 1.1 × 10⁻³; ORClass I = 1.42, 95% CI 1.17–1.73), and the constrained-gene burden remained after excluding the 32 FDR < 5% genes (OR = 1.23, P = 4.3 × 10⁻²⁷).

Methods

  • Aggregated exome sequence data from 24,248 schizophrenia cases and 50,437 individuals without a known psychiatric diagnosis across 11 global collections, then added 46,885 gnomAD controls for a combined discovery set of 24,248 cases and 97,322 controls; also analyzed de novo mutations from 3,402 parent-proband trios.
  • Samples were jointly reprocessed and variant-called with a standardized BWA-Picard-GATK pipeline; analyses were restricted to coding exons and ultra-rare variants (MAC ≤ 5), including PTVs and damaging missense variants defined by MPC > 3 (Class I) and MPC 2–3 (Class II).
  • Gene-based burden tests used Fisher’s exact test with case-control permutations within strata, meta-analysis by weighted Z-score, and Poisson rate tests for de novo mutations; pathway, tissue, and cell-type enrichments were also tested, and results were jointly analyzed with the largest schizophrenia GWAS.

Limitations

  • The paper states that earlier sequencing studies had limited power for individual gene discovery and that even after excluding the 32 FDR < 5% genes, cases still carry a substantial excess of ultra-rare variants (OR = 1.23, P = 4.3 × 10⁻²⁷), implying that many schizophrenia risk genes remain undiscovered.
  • The paper notes that only summary-level variant counts were available for 46,885 external gnomAD controls, and it used permutation testing and strata to control for possible technical or methodological batch effects.
  • The paper states that confidence intervals for the top gene associations are wide (e.g., GRIN2A Class I 5.36–221; GRIA3 Class I 4.28–188), reflecting imprecise effect estimates from few variant carriers; no formal limitations section is included.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1101/2020.09.18.20192815

Cites

80 of 80 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 31 cited sources checked · 0 supported · 0 flagged

  • Findingnone given

    Combining these findings with other large-scale sequencing studies, we find shared and distinct genetic signals between schizophrenia and other neurodevelopmental disorders.

    “Combining these findings with other large-scale sequencing studies, we find shared and distinct genetic signals between schizophrenia and other neurodevelopmental disorders.”
    Quote in the paperNot a substantial claim
  • Findingown data

    Few of the strongly associated schizophrenia genes appear to confer risk for DD/ID.

    “Nevertheless, few of the strongly associated schizophrenia genes appear to confer risk for DD/ID.”
    • Also stated in: § Shared and distinct genetic risk with other neurodevelopmental disorders · plain
      “Nine of the ten schizophrenia genes showed limited de novo PTV signal in DD/ID, with a combined 8 de novo PTVs observed in these genes (Xexp = 4.98; PPois = 0.13; Figure 5B; Table S13).”
    Quote in the paperSubstantial claim
  • Findingown data

    PTV and damaging missense variants in NMDA receptor subunit GRIN2A confer substantial risk for schizophrenia (P = 7.37 x 10 -7 ; Class I OR 24.1, 95% CI 5.36 -221; Class II OR 2.37, 95% CI 1.1 -4.92).

    “Here, we find that PTV and damaging missense variants in NMDA receptor subunit GRIN2A confer substantial risk for schizophrenia (P = 7.37 x 10 -7 ; Class I [PTV and MPC > 3] OR 24.1, 95% CI 5.36 -221; Class II [MPC > 2] OR 2.37, 95% CI 1.1 -4.92).”
    Quote in the paperSubstantial claim
  • Findingown data

    There is significant evidence for an overlap of rare variant risk between schizophrenia, autism spectrum disorders (ASD), and severe neurodevelopmental disorders (DD/ID).

    “We find significant evidence for an overlap of rare variant risk between schizophrenia, autism spectrum disorders (ASD), and severe neurodevelopmental disorders (DD/ID), supporting a neurodevelopmental etiology for schizophrenia.”
    • Also stated in: § Shared and distinct genetic risk with other neurodevelopmental disorders · plain
      “We found a significant excess of URVs in schizophrenia cases compared to controls in the 299 DD/ID-associated genes (Pmeta = 1.5 x 10 -14 ; ORClass I = 1.44, 1.3 -1.6 95% CI), and in the 102 ASD-associated genes (Pmeta = 3.7 x 10 -7 ; ORClass I = 1.45, 1.23 -1.72 95% CI; Figure 5A; Table S12).”
    Quote in the paperSubstantial claim
  • Findingown data

    We observed significant enrichment of URVs in 33 gene sets (P < 2.9 x 10 -5 ) that recapitulated consistent and overlapping cellular compartments and biological processes, including postsynaptic density, chromatin modification, regulation of ion transmembrane transport, axon guidance, voltage-gated cation channel activity, and synaptic transmission.

    “We observed significant enrichment of URVs in 33 gene sets (P < 2.9 x 10 -5 ) that recapitulated consistent and overlapping cellular compartments and biological processes, including definitions of the postsynaptic density (human cortex biopsy post-synaptic density; P = 1.2 x 10 -12 ), chromatin modification (GO:0016568; P = 1.8 x 10 -12 ), regulation of ion transmembrane transport (GO:0034765; P =”
    Quote in the paperSubstantial claim
  • May combine two claims. 28
  • Classified as probably not a substantial claim. 14

  • May combine two claims.

    Findingown data

    De novo mutations are enriched for the 244 genes with P < 0.01 in our case-control analysis, with limited or no signal in the remaining genes in the genome.

    “Using Poisson rate tests based on expected mutation rate 25 , we found these de novo mutations are enriched for the 244 genes with P < 0.01 in our case-control analysis (Figure S16, Table S5), with limited or no signal in the remaining genes in the genome (Figure 1D).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [25] source

      Not checked in the free run.

  • May combine two claims.

    Findingown data

    Protein-truncating variants in GRIN2A, TRIO, and CACNA1G confer risk for schizophrenia whereas specific missense mutations in these genes confer risk for DD/ID.

    “We show that proteintruncating variants in GRIN2A, TRIO, and CACNA1G confer risk for schizophrenia whereas specific missense mutations in these genes confer risk for DD/ID.”
    • Also stated in: § Shared and distinct genetic risk with other neurodevelopmental disorders · plain
      “Notably, three of the ten risk genes for schizophrenia (TRIO, GRIN2A, and CACNA1G) were associated with risk of severe DD/IDs exclusively through de novo missense mutations that cluster within each gene (Figure 5B; Table S13), while the schizophrenia signal was largely driven by PTVs.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Genes with the highest specific expression in brain regions showed the strongest enrichment of risk URVs, most significantly in the human frontal cortex (P = 1.63 x 10 -8 ) and with limited signal in the other tissue types.

    “Following earlier reports studying heritability enrichment in GTEx tissues 5,30 , we found that genes with the highest specific expression in brain regions showed the strongest enrichment of risk URVs, most significantly in the human frontal cortex (P = 1.63 x 10 -8 ) and with limited signal in the other tissue types (Figure S19, Table S8, S9).”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [5] source

      Not checked in the free run.

    • cites [30] source

      Not checked in the free run.

  • May combine two claims.

    Findingown data

    Individual genes SETD1A, TRIO, and SP4 exhibited prenatal expression while GRIN2A and GRIA3 showed postnatal expression.

    “Indeed, individual genes like SETD1A, TRIO, and SP4 exhibited prenatal expression while GRIN2A and GRIA3 showed postnatal expression (Figure S24).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    De novo missense mutations in GRIN2A, TRIO, and CACNA1G clustered in specific domains and are associated with more severe neurodevelopmental, syndromic disorders with cognitive impairment.

    “De novo missense mutations in these genes clustered in specific domains and are associated with more severe neurodevelopmental, syndromic disorders with cognitive impairment”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Statistical fine-mapping prioritized the likely underlying protein-coding gene at 69 of the GWAS associations, and we found a case-control enrichment of URVs in these genes (Pmeta = 1.1 x 10 -3 ; ORClass I = 1.42, 1.17 -1.73 95% CI).

    “Statistical fine-mapping prioritized the likely underlying protein-coding gene at 69 of these associations (Table S11), and we found a case-control enrichment of URVs in these genes (Pmeta = 1.1 x 10 -3 ; ORClass I = 1.42, 1.17 -1.73 95% CI; Figure 4A, Table S12).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    While PTV mutations in certain genes are joint risk factors for schizophrenia and DD/ID, the majority of schizophrenia associations reported here appear to have little or no role in DD/ID.

    “Broadly speaking, while PTV mutations in certain genes are joint risk factors for schizophrenia and DD/ID, the majority of schizophrenia associations reported here appear to have little or no role in DD/ID despite the enormous power of published DD/ID studies to date.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    We observe allelic series in GRIN2A, TRIO, and CACNA1G in which PTVs increase schizophrenia risk and de novo missense mutations confer strong DD/ID risk.

    “Of further interest, we observe allelic series in GRIN2A, TRIO, and CACNA1G in which PTVs increase schizophrenia risk and de novo missense mutations confer strong DD/ID risk.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    FAM120A and STAG1 resided in more complex GWAS-associated regions containing multiple genes but were prioritized among their neighbors as FDR < 5% in our sequencing study.

    “Furthermore, FAM120A and STAG1 resided in more complex GWAS-associated regions containing multiple genes but were prioritized among their neighbors as FDR < 5% in our sequencing study (Figure 4C, 4D).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Three of the ten risk genes for schizophrenia (TRIO, GRIN2A, and CACNA1G) were associated with risk of severe DD/IDs exclusively through de novo missense mutations that cluster within each gene, while the schizophrenia signal was largely driven by PTVs.

    “Notably, three of the ten risk genes for schizophrenia (TRIO, GRIN2A, and CACNA1G) were associated with risk of severe DD/IDs exclusively through de novo missense mutations that cluster within each gene (Figure 5B; Table S13), while the schizophrenia signal was largely driven by PTVs.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingnone given

    At GRIN2A, common regulatory variants modestly influence disease risk and PTV and predicted damaging missense variants increase risk more substantially.

    “A natural doseresponse curve occurs at this gene in which common regulatory variants modestly influence disease risk and PTV and predicted damaging missense variants increase risk more substantially.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    We found a significant excess of URVs in schizophrenia cases compared to controls in the 299 DD/ID-associated genes (Pmeta = 1.5 x 10 -14 ; ORClass I = 1.44, 1.3 -1.6 95% CI), and in the 102 ASD-associated genes (Pmeta = 3.7 x 10 -7 ; ORClass I = 1.45, 1.23 -1.72 95% CI).

    “We found a significant excess of URVs in schizophrenia cases compared to controls in the 299 DD/ID-associated genes (Pmeta = 1.5 x 10 -14 ; ORClass I = 1.44, 1.3 -1.6 95% CI), and in the 102 ASD-associated genes (Pmeta = 3.7 x 10 -7 ; ORClass I = 1.45, 1.23 -1.72 95% CI; Figure 5A; Table S12).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    SETD1A had a significant de novo PTV signal in DD/ID (Xobs = 8, Xexp = 0.41; P = 1.3 x 10 -8 ), supporting an earlier report that described SETD1A as a gene associated with both schizophrenia and broader neurodevelopmental disorders.

    “SETD1A had a significant de novo PTV signal in DD/ID (Xobs = 8, Xexp = 0.41; P = 1.3 x 10 -8 ), supporting an earlier report that described SETD1A as a gene associated with both schizophrenia and broader neurodevelopmental disorders 18 .”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [18] source

      Not checked in the free run.

  • May combine two claims.

    Findingown data

    Schizophrenia risk genes are involved in the postsynaptic density and broader synaptic function.

    “Genome-wide analyses recapitulated known biological processes and reaffirm that schizophrenia risk genes are involved in the postsynaptic density and broader synaptic function”
    Quote in the paperNot a substantial claim

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

    Findingown data

    ASD individuals have 0.0478 (95% CI 0.0387-0.0568) extra de novo PTV mutations.

    “while ASD individuals have 0.0478 (0.0387 -0.0568 95% CI) extra de novo PTV mutations”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: Schizophrenia risk URVs are concentrated in the 3,063 PTV-intolerant (pLI > 0.9) genes.

  • May combine two claims.

    Interpretationreasoning

    Increased scale following this approach will uncover additional risk genes and help complete the genetic architecture of schizophrenia.

    “Finally, we demonstrate that increased scale following this approach will uncover additional risk genes and help complete the genetic architecture of schizophrenia.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    Analysis of URVs has had limited success in delivering individual gene discovery in schizophrenia because of power limitations, with only a single gene, SETD1A, identified as robustly associated.

    “However, the analysis of URVs has had limited success in delivering individual gene discovery in schizophrenia because of power limitations, with only a single gene, SETD1A, identified as robustly associated 13,18 .”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [13] source

      Not checked in the free run.

    • cites [18] source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Risk alleles are concentrated in genes with a near-complete depletion of protein-truncating variants in population studies, a result shared with other neurodevelopmental disorders and suggesting strong direct selection against such mutations.

    “Furthermore, these risk alleles are concentrated in genes with a near-complete depletion of protein-truncating variants in population studies, a result shared with other neurodevelopmental disorders 6,8 and suggesting strong direct selection against such mutations.”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [6] source

      Not checked in the free run.

    • cites [8] Singh, T. et al. The contribution of rare variants to risk of schizophrenia in individuals with and without intellectual disability. Nat. Genet. 1–10 (2017).

      Not checked in the free run.

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

    Premisecitation

    Common variants explain ~24% of the variance in disease liability.

    “As a class of variation, common variants explain ~24% of the variance in disease liability 11 .”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [11] source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S2 (checkable).

    Premisecitation

    The most natural grouping for rare variants is within a gene, based on predicted functional consequence or evidence for deleteriousness.

    “The most natural grouping for rare variants is within a gene, based on predicted functional consequence or evidence for deleteriousness 13,14 .”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [13] source

      Not checked in the free run.

    • cites [14] source

      Not checked in the free run.

    • Rests on an assumption: Most rare variants will have little or no functional consequence or impact on risk.

  • Probably not a substantial claim: may not pass S2 (checkable).

    Premisereasoning

    Analyzing rare coding variants offers a powerful complementary approach to identify genes in complex traits.

    “Analyzing rare coding variants offers a powerful complementary approach to identify genes in complex traits.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: Most rare variants will have little or no functional consequence or impact on risk.
    • Rests on an assumption: Causal variants in schizophrenia GWAS are predominantly noncoding.

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

    Premisecitation

    Schizophrenia has a substantial genetic component with contributions from across the allele frequency spectrum.

    “It is well-established that schizophrenia has a substantial genetic component with contributions from across the allele frequency spectrum [5][6][7][8] .”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [5][6][7][8]

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    NMDA receptor changes in composition during prenatal to postnatal neurodevelopment with GRIN2A predominantly expressed during late childhood and adolescence, recapitulating expected epidemiological observations on schizophrenia age-of-onset.

    “The NMDA receptor changes in composition during prenatal to postnatal neurodevelopment with GRIN2A predominantly expressed during late childhood and adolescence, recapitulating expected epidemiological observations on schizophrenia age-ofonset (Supplementary Information, Figure 3B) 27 .”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [27] source

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Premisecitation

    Genome-wide association studies have identified 270 common risk loci of individually small effect (median OR < 1.05).

    “This has been borne out by genome-wide association studies (GWAS) which have now, in a companion paper, identified 270 common (minor allele frequency [MAF] > 1%) risk loci of individually small effect (median odds ratio [OR] < 1.05) 10 .”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [10] Schizophrenia Working Group of the Psychiatric Genomics Consortium. Mapping genomic loci prioritises genes and implicates synaptic biology in schizophrenia. Submitted (2020).

      Not checked in the free run.

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

    Premisecitation

    Several rare recurrent copy number variants (CNVs) have been robustly associated with schizophrenia, as exemplified by dramatically higher rates of schizophrenia in 22q11.2 deletion carriers.

    “Several rare (MAF < 0.1%) recurrent copy number variants (CNVs) have also been robustly associated with schizophrenia, as exemplified by the dramatically higher rates of schizophrenia in 22q11.2deletion carriers 7,12 .”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [7] source

      Not checked in the free run.

    • cites [12] source

      Not checked in the free run.

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

    Premisecitation

    Earlier schizophrenia sequencing studies have established that ultra-rare and de novo mutations contribute to risk as a category, and have prioritized disease-relevant tissues and processes, specifically observing an enrichment in neuronal genes and synaptic processes.

    “Earlier schizophrenia sequencing studies have established that ultra-rare and de novo mutations contribute to risk as a category, and have prioritized disease-relevant tissues and processes, specifically observing an enrichment in neuronal genes and synaptic processes 6,8,[16][17][18][19][20] .”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites 6,8,[16][17][18][19][20]

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Schizophrenia GWAS identified a common variant at GRIN2A (OR = 1.057, P = 1.57x 10 - 10 ), providing an allelic series in which different perturbation of gene function results in severity of disease risk.

    “Schizophrenia GWAS also identified a common variant at GRIN2A (OR = 1.057,P = 1.57x 10 - 10 ), providing an allelic series in which different perturbation of gene function results in severity of disease risk (Figure 3A) 5 .”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [5] source

      Not checked in the free run.

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

    Study design and conductown data

    This enrichment provides additional reassurance of the robustness of our case-control gene results.

    “The most striking enrichment was observed for the 52 genes with case-control P < 0.001 (Class I mutations: P = 2.1 x 10 -11 ; Rate ratio = 8.3, 95% CI = 4.9 -13), which provides additional reassurance of the robustness of our case-control gene results.”
    Quote in the paperNot a substantial claim

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

    Premisecitation

    De novo mutations in GRIN2B are associated with more severe disorders of neurodevelopment that manifest in childhood, including intellectual disability and autism.

    “De novo mutations in GRIN2B conversely are associated with more severe disorders of neurodevelopment that manifest in childhood, including intellectual disability and autism 41 .”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [41] source

      Not checked in the free run.

  • May combine two claims.

    Interpretationown data

    These results suggest there is at least partial convergence in the genes and biological processes implicated by common and ultra-rare genetic variation, and that ultra-rare coding variants can be leveraged to prioritize genes within GWAS loci.

    “Combined, these results suggest there is at least partial convergence in the genes and biological processes implicated by common and ultra-rare genetic variation, and that ultra-rare coding variants can be leveraged to prioritize genes within GWAS loci.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    These observations suggest schizophrenia and childhood onset neurodevelopmental disorders share some genes and biological processes, but that at least in some cases, the severity or the nature of the functional impairment differs between disorders.

    “These observations suggest schizophrenia and childhood onset neurodevelopmental disorders share some genes and biological processes, but that at least in some cases, the severity or the nature of the functional impairment differs between disorders.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    The genetic architecture of schizophrenia is composed of common variants of small effects, large CNVs with elevated frequencies driven by genomic instability, and URVs of large effect implicating individual genes.

    “Efforts in the past decade are beginning to generate a more comprehensive view of the genetic architecture for schizophrenia, composed of common variants of small effects, large CNVs with elevated frequencies driven by genomic instability, and now, URVs of large effect implicating individual genes (Figure 6A) 5,7 .”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [5] source

      Not checked in the free run.

    • cites [7] source

      Not checked in the free run.

  • May combine two claims.

    Interpretationown data

    In the ten schizophrenia-associated genes, 0.86% of patients are carriers and ~35% of variants are de novo.

    “suggesting that 0.86% of patients are carriers and ~35% of variants are de novo.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Schizophrenia as a trait is under strong selection.

    “Because schizophrenia as a trait is under strong selection [38][39][40]”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [38][39][40]

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

    Premisecitation

    We are still in the early stages of gene discovery.

    “Despite this progress, it is clear that we are still in the early stages of gene discovery 10 .”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [10] Schizophrenia Working Group of the Psychiatric Genomics Consortium. Mapping genomic loci prioritises genes and implicates synaptic biology in schizophrenia. Submitted (2020).

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

    Interpretationreasoning

    The vast majority of risk alleles, their direction and magnitude of effect, mode of action, and responsible genes are yet to be discovered.

    “The vast majority of risk alleles, their direction and magnitude of effect, mode of action, and responsible genes are yet to be discovered.”
    Quote in the paperNot a substantial claim

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

    Interpretationreasoning

    URVs of large effect are expected to be frequently de novo or of very recent origin and contribute to risk in only a fraction of diagnosed patients.

    “we expect that URVs of large effect to be frequently de novo or of very recent origin and contribute to risk in only a fraction of diagnosed patients.”
    Quote in the paperSubstantial claim
37 of 37

  • The paper relies on this without support.

    Most rare variants will have little or no functional consequence or impact on risk.

    “However, most rare variants will have little or no functional consequence or impact on risk, posing a significant challenge in identifying those that are truly causal and complicating required analyses in which rare variants are tested as a group rather than individually.”

  • The paper relies on this without support.

    Linkage disequilibrium can be readily avoided as a driver of spurious signals because these variants are each seen in very few individuals.

    “As these variants are each seen in very few individuals -and often de novo or very recent -linkage disequilibrium can be readily avoided as a driver of spurious signals meaning that genes frequently can be identified directly.”

  • The paper relies on this without support.

    Causal variants in schizophrenia GWAS are predominantly noncoding.

    “Because causal variants in schizophrenia GWAS are predominantly noncoding, challenges related to fine-mapping and interpretation of intergenic and intronic elements limit our ability to confidently identify underlying genes, infer the mechanism by which they influence disease risk, and determine the direction of effect.”

  • The paper relies on this without support.

    Schizophrenia risk URVs are concentrated in the 3,063 PTV-intolerant (pLI > 0.9) genes.

    “We restrict these analyses to the 3,063 PTV-intolerant (pLI > 0.9) genes in which schizophrenia risk URVs are concentrated.”

  • The paper relies on this without support.

    The NMDA receptor is composed of two GRIN2 units (GRIN2A and/or GRIN2B) along with two constitutive GRIN1 units.

    “the NMDA receptor is composed of two GRIN2 units (GRIN2A and/or GRIN2B) along with two constitutive GRIN1 units”

  • The paper relies on this without support.

    GRIN2A increases dramatically in expression later in childhood and adolescence, mimicking the age of onset of disease for schizophrenia.

    “GRIN2A increases dramatically in expression later in childhood and adolescence, mimicking the age of onset of disease for schizophrenia.”
6 of 6

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?

  • SP4 is a transcription factor highly expressed in brain, is regulated by NMDA transmission, and regulates NMDA receptor abundance.

    “SP4, a transcription factor highly expressed in brain and which is regulated by NMDA transmission, and also regulates NMDA receptor abundance37.”

    From Mapping genomic loci implicates genes and synaptic biology in schizophrenia (2022) · cites this paper as [37]

  • Of the 10 exome-wide significant genes identified by SCHEMA, two were prioritised candidates from fine-mapping: GRIN2A and SP4.

    “Of the 10 exome-wide significant genes identified by SCHEMA36, two were prioritised candidates from fine-mapping; GRIN2A encoding a glutamatergic NMDA receptor subunit, and SP4, a transcription factor highly expressed in brain and which is regulated by NMDA transmission, and also regulates NMDA receptor abundance37.”

    From Mapping genomic loci implicates genes and synaptic biology in schizophrenia (2022) · cites this paper as [37]

  • Significant SMR associations imply that schizophrenia associations colocalize with eQTLs.

    “Significant SMR associations imply colocalization of the schizophrenia associations with eQTL.”

    From Mapping genomic loci implicates genes and synaptic biology in schizophrenia (2022) · cites this paper as [37]

  • The authors used the HEIDI test to remove SMR associations likely caused by linkage disequilibrium between schizophrenia-associated variants and eQTLs.

    “We applied the HEIDI test37 to filter out SMR associations (PHEIDI < 0.01) due to linkage disequilibrium between SCZ-associated variants and eQTLs.”

    From Mapping genomic loci implicates genes and synaptic biology in schizophrenia (2022) · cites this paper as [37]

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.

  • GRIN2A Class II variants use a conflicting MPC cutoff

    Minor

    The GRIN2A result labels Class II as MPC > 2, which includes variants with MPC > 3. Elsewhere, the paper defines Class II specifically as MPC 2-3 and assigns MPC > 3 variants to Class I.

1 of 1

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Related

Authors
Singh, T., Neale, B. M., & Daly, M. J.
Affiliations
Broad Institute (US)
Corresponding author
  • Singh, Tarjinder
  • Neale, Benjamin M.
  • Daly, Mark J.
ORCID
Venue
Nature
Publisher
openRxiv
Published
2020-09-18
Volume
604
Issue
7906
Pages
509-516
ISSN
0028-0836 (ISSN-L)
DOI
10.1101/2020.09.18.20192815
PMC
PMC9805802
OpenAlex
W4223433728
Semantic Scholar
0700d2bf8e5a7f57f262a574f1d6005795936054
Cited by
980
References
78
Citations by year
2026: 156 · 2025: 214 · 2024: 220 · 2023: 197 · 2022: 138 · 2021: 53 · 2020: 1 · 2019: 1
FWCI
125.63
Type
journal article
Language
en
Open access
Licence
cc-by
Locations
  • Nature (publishedVersion)
  • medRxiv (acceptedVersion, cc-by)
  • PubMed (publishedVersion)
  • Research Portal (King's College London) (publishedVersion, other-oa)
  • Research at the University of Copenhagen (University of Copenhagen) (publishedVersion, other-oa)
  • ORCA Online Research @Cardiff (Cardiff University) (acceptedVersion)
  • PubMed Central (acceptedVersion)
  • EUR Research Repository (Erasmus University Rotterdam) (submittedVersion, other-oa)
  • UCL Discovery (University College London) (submittedVersion)
  • eScholarship (California Digital Library) (submittedVersion)
  • Työväentutkimus Vuosikirja (submittedVersion)
  • Queen Mary Research Online (Queen Mary University of London) (submittedVersion)
  • STM:n Hallinnonalan avoin julkaisuarkisto (Julkari) (submittedVersion)
Funders
  • GlaxoSmithKline
  • Massachusetts General Hospital
  • Biogen
  • Dalio Foundation
  • Broad Institute (U54HG003067)
  • Stanley Family Foundation
  • National Institute for Health and Care Research
  • King's College London
  • Cardiff University (MR/L010305/1, G0800509)
  • Lundbeckfonden (R155-2014-1724, R204-2015-4129, R341-2020-222, R248‐2017‐2003, R24-2008-3243, R364-2021-85)
  • National Institutes of Health (5u01da024417-02, -0001, 5p30ns069329-02, AA 09367, DA 05147, DA024417, HG007022)
  • Institute of Psychiatry, Psychology and Neuroscience, King’s College London
  • Canadian Institutes of Health Research
  • Medical Research Council (MR/L010305/1, G0800509)
  • National Institute on Alcohol Abuse and Alcoholism (R01AA009367, R37AA009367)
  • National Institute on Aging (R01AG064955, R01AG035083, P50AG005681, P01AG003991)
  • National Heart, Lung, and Blood Institute (U01HL117626, R01HL117626)
  • National Institute of Mental Health (U01MH109539, R01MH085521, U01MH105641, U01MH105666, U01MH105653, R01MH116037, R01MH085548, R01MH104964, R01MH094145, MH105653, U01MH105578, R01MH124851, R01MH123451, DA05147)
  • National Institute on Drug Abuse (U01-DA024417, R01DA037904, R01DA024417, R01DA005147, R01DA036216, R37DA005147)
  • National Human Genome Research Institute (U54HG003067, R01HG007022)
  • National Institute of Neurological Disorders and Stroke (P30-NS069329)
  • Eunice Kennedy Shriver National Institute of Child Health and Human Development (P50HD105351)
Topics
  • Health Sciences › Medicine › Psychiatry and Mental health › Schizophrenia research and treatment
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Genetics › Genetic Associations and Epidemiology
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Genetics › Genomics and Rare Diseases
Keywords
schizophrenia, risk genes, GRIN2A, GRIA3, NMDA receptor, AMPA receptors, glutamatergic system, synapses, neurodevelopmental disorders, autism spectrum disorder, epilepsy, schizophrenia pathogenesis
MeSH terms
Humans, Mutation, Schizophrenia, Case-Control Studies, Receptors, N-Methyl-D-Aspartate, Genetic Predisposition to Disease, Exome, Neurodevelopmental Disorders

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