CoChat
Sign Up
Cited by 2,194Genetic Associations and EpidemiologyDiabetes and associated disordersGenetic Mapping and Diversity in Plants and Animals

Replication of Genome-Wide Association Signals in UK Samples Reveals Risk Loci for Type 2 Diabetes

Open access
Zeggini, Eleftheria, Weedon, Michael N., Lindgren, Cecilia M., Frayling, Timothy M., Elliott, Katherine S., Lango, Hana, Timpson, Nicholas J., Perry, John R. B. +18 more·2007·Science
DOI 10.1126/science.1142364PMC PMC3772310
Open this paper in CoChat — free account

Abstract

The molecular mechanisms involved in the development of type 2 diabetes are poorly understood. Starting from genome-wide genotype data for 1924 diabetic cases and 2938 population controls generated by the Wellcome Trust Case Control Consortium, we set out to detect replicated diabetes association signals through analysis of 3757 additional cases and 5346 controls and by integration of our findings with equivalent data from other international consortia. We detected diabetes susceptibility loci in and around the genes CDKAL1 , CDKN2A/CDKN2B , and IGF2BP2 and confirmed the recently described associations at HHEX/IDE and SLC30A8 . Our findings provide insight into the genetic architecture of type 2 diabetes, emphasizing the contribution of multiple variants of modest effect. The regions identified underscore the importance of pathways influencing pancreatic beta cell development and function in the etiology of type 2 diabetes.

Integrity signals

Registry

Crossref · 10.1126/science.1142364

Cites

24 of 24 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 41 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    Replication was obtained for SNPs within CDKAL1, including rs9465871 and rs10946398.

    “Replication was also obtained for SNPs within the CDKAL1 locus on chromosome 6, including rs9465871 and rs10946398.”
    Quote in the paperSubstantial claim
  • Findingown data

    rs10946398 showed more striking replication (OR 1.14 [1.07-1.22], p=8.4×10−5).

    “The replication signal at rs10946398 was more striking (OR 1.14 [1.07-1.22], p=8.4×10−5) (Tables 1, S2).”
    Quote in the paperNot a substantial claim
  • Findingown data

    rs8050136 in the FTO region was among the strongest association signals outside TCF7L2 in the original scan (OR 1.27 [1.16-1.37], p=2.0×10−8).

    “Rs8050136 (mapping to the FTO [fat mass associated with obesity] gene region on chr16) was among a cluster of SNPs generating the strongest evidence for association outside TCF7L2 in the original scan (risk allele OR 1.27 [1.16-1.37] p=2.0×10−8) (Figure S1).”
    Quote in the paperSubstantial claim
  • Findingown data

    In the first replication wave, genotyping 21 representative SNPs from 30 SNPs in 9 regions (excluding TCF7L2) with p<10^-5 gave replication (p<0.05) for three of the 9 regions.

    “Genotyping of 21 representative SNPs generated evidence of replication (p<0.05) for three of these 9 regions (Tables 1, S2).”
    Quote in the paperSubstantial claim
  • Findingown data

    Combining WTCCC scan data with the authors' replication studies and data from DGI and FUSION identified several additional T2D susceptibility variants.

    “Here, we describe how integration of data from the WTCCC scan and our own replication studies with similar information generated by the Diabetes Genetics Initiative [DGI] (6) and the Finland-United States Investigation of NIDDM Genetics [FUSION] (7) has identified several additional susceptibility variants for T2D.”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [6] Diabetes Genetics Initiative Science this issue

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

    • cites [7] Scott LJ Science this issue

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

    • Rests on an assumption: The DGI and FUSION genome-wide association data are comparable enough to the WTCCC and UK replication data to be combined in meta-analysis.
  • May combine two claims. 23
  • Classified as probably not a substantial claim. 5

  • May combine two claims.

    Findingown data

    CDKAL1 mRNA is expressed in human pancreatic islet and skeletal muscle.

    “We have detected expression of CDKAL1 mRNA in human pancreatic islet and skeletal muscle (Figure S2).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The study identified T2D susceptibility loci in and around CDKAL1, CDKN2A/CDKN2B and IGF2BP2, and confirmed associations at HHEX/IDE and SLC30A8.

    “We detected diabetes susceptibility loci in and around the genes CDKAL1, CDKN2A/CDKN2B and IGF2BP2 and confirmed the recently described associations at HHEX/IDE and SLC30A8.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    rs9465871 had the stronger WTCCC signal but its replication was modest (p=0.023).

    “Although rs9465871 generated the stronger signal in the WTCCC scan, replication at this SNP was modest (p=0.023).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    A second chromosome 9 signal ~100kb 5′ (rs564398, OR 1.16 [1.07-1.27], p=3.2×10−4) was weakly supported in FUSION but not DGI, and replicated in UK RS samples (OR 1.12 [1.05-1.19], p=8.6×10−4).

    “A second signal from the WTCCC scan located ~100kb 5′ (rs564398, OR 1.16 [1.07-1.27], p=3.2×10−4) was weakly supported in the FUSION, but not the DGI scan (Tables 1, S3) and replicated in the UK RS samples (OR 1.12 [1.05-1.19], p=8.6×10−4) (Tables 1, S3).”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The DGI and FUSION genome-wide association data are comparable enough to the WTCCC and UK replication data to be combined in meta-analysis.

  • May combine two claims.

    Findingown data

    DGI and FUSION gave modest but consistent signals at HHEX/IDE; pooled evidence was strong (p~5.7×10−10) for T2D susceptibility.

    “Both DGI and FUSION studies showed modest, but consistent association signals generating strong combined evidence (p~5.7×10−10) for a role in T2D susceptibility (Tables 1, S3).”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The DGI and FUSION genome-wide association data are comparable enough to the WTCCC and UK replication data to be combined in meta-analysis.

  • May combine two claims.

    Findingown data

    Independent genotyping of rs13266634 replicated the SLC30A8 finding (OR 1.12 [1.05-1.18], p=7.0×10−5 in all UK data) and across all three studies (p~5.3×10−8).

    “However, we genotyped rs13266634 independently and obtained replication of the finding (risk allele OR 1.12 [1.05-1.18], p=7.0×10−5 in all UK data) and across all three studies (p~5.3×10−8, Tables 1, S4).”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The DGI and FUSION genome-wide association data are comparable enough to the WTCCC and UK replication data to be combined in meta-analysis.

  • May combine two claims.

    Findingown data

    None of the remaining first-wave regions replicated in UK samples, and none had strong support from DGI and FUSION.

    “Of the remaining regions selected in the first wave, none showed any evidence of replication in UK samples (Table S2), and for none was there strong support from the DGI and FUSION scans.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    rs9369425 near VEGFA showed association evidence in WTCCC (OR 1.16 [1.06-1.27], p=8.6×10−4), nominal replication in UK (1.08 [1.01-1.15], p=0.03), and DGI (1.17 [1.04-1.32], p=4.4×10−3).

    “Evidence for association in the WTCCC scan (OR 1.16 [1.06-1.27], p=8.6×10−4) is supported by nominal replication in UK samples (1.08 [1.01-1.15], p=0.03) and by DGI scan results (1.17 [1.04-1.32], p=4.4×10−3).”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: The DGI and FUSION genome-wide association data are comparable enough to the WTCCC and UK replication data to be combined in meta-analysis.

  • May combine two claims.

    Findingown data

    rs10811661 replicated in UK samples (OR 1.18 [1.08-1.28], p=1.7×10−4), DGI (p=2.2×10−5), and FUSION (rs2383208, p=9.7×10−3).

    “Replication was observed in UK samples (rs10811661:OR 1.18 [1.08-1.28], p=1.7×10−4), as well as DGI (p=2.2×10−5) and FUSION follow-up studies (rs2383208, p=9.7×10−3).”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The DGI and FUSION genome-wide association data are comparable enough to the WTCCC and UK replication data to be combined in meta-analysis.

  • May combine two claims.

    Findingown data

    Both CDKN2B and CDKN2A are highly expressed in pancreatic islets and pituitary.

    “Both CDKN2B and CDKN2A display high levels of expression in pancreatic islets and pituitary (Figure S2).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingcitation

    Four GWA studies have raised the count of genuine replicated T2D susceptibility signals from 3 to 9, adding HHEX/IDE, SLC30A8, CDKAL1, CDKN2A, IGF2BP2 and FTO.

    “As a result of the four GWA studies reported to date (5, 6, 7, 10), the number of genuine, replicated T2D susceptibility signals has climbed from 3 to 9 (adding HHEX/IDE, SLC30A8, CDKAL1, CDKN2A, IGF2BP2 and FTO).”
    Quote in the paperSubstantial claim0 of 4 sources checked
    • cites [5] Donnelly P the WTCCC, personal communication Data from the Wellcome Trust Case Control Consortium scan

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

    • cites [6] Diabetes Genetics Initiative Science this issue

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

    • cites [7] Scott LJ Science this issue

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

    • cites [10] Sladek R Nature 2007 445 881 885 17293876 source

      Not checked in the free run.

  • May combine two claims.

    Findingown data

    The loci account for only a minor share of familiality; the sibling relative risk from all loci in UK samples is about 1.07.

    “However, these loci explain only a small proportion of the observed familiality (the sibling relative risk, λs, attributable to all loci in the UK samples is only ~1.07).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    rs8050136 influences T2D risk mainly via adiposity, and controlling for BMI removes the T2D link.

    “As we recently reported (9), this effect on T2D risk is mediated through a primary effect on adiposity, and adjustment for BMI abolishes the T2D association.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [9] Frayling TM Science Published online April 12 2007

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

  • May combine two claims.

    Study design and conductcitation

    The T2D-specific dataset was restricted to 393,453 autosomal SNPs with MAF above 1% in cases and controls and no extreme HWE departure.

    “We considered only the 393,453 autosomal SNPs with minor allele frequency (MAF) exceeding 1% in both cases and controls and no extreme departure from Hardy Weinberg equilibrium (HWE) (p<10−4 in cases or controls) (8).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [8] Materials and methods are available as supporting material on Science Online

      Not checked in the free run.

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

    Study design and conductcitation

    WTCCC candidate signals were further checked by inspecting cluster plots and by genotyping on another platform.

    “we first submitted putative signals from the WTCCC study to additional quality control including cluster plot visualization and validation genotyping on a second platform (8).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [8] Materials and methods are available as supporting material on Science Online

      Not checked in the free run.

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

    Premisecitation

    CDK5 is implicated in pancreatic beta cell function through p35/CDK5 complexes that downregulate insulin expression.

    “CDK5 has been implicated in the regulation of pancreatic beta cell function, through formation of p35/CDK5 complexes that downregulate insulin expression (11, 12).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [11] Ubeda M Rukstalis JM Habener JF J. Biol. Chem 2006 281 28858 28864 16887799 source

      Not checked in the free run.

    • cites [12] Wei FY Nat. Med 2005 11 1104 1108 16155576 source

      Not checked in the free run.

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

    Study design and conductcitation

    In the WTCCC study, 459,448 of 490,032 autosomal SNPs in 16,179 samples passed initial quality control.

    “analysis of 490,032 autosomal SNPs in 16,179 samples yielded 459,448 SNPs that passed initial quality control (5).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [5] Donnelly P the WTCCC, personal communication Data from the Wellcome Trust Case Control Consortium scan

      Not checked in the free run.

  • May combine two claims.

    Interpretationown data

    Three loci-CDKAL1, IGF2BP2 and CDKN2A-replicated widely across studies and are novel T2D susceptibility loci.

    “The other three loci (near CDKAL1, IGF2BP2 and CDKN2A), all showing extensive replication across the three studies represent novel T2D susceptibility loci.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    Finding variants with these effect sizes reliably requires very large samples; this study typed 13,965 individuals.

    “Robust identification of variants with such effect sizes is only feasible with large-scale sample sets (13,965 individuals were typed in the present study).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    In mice, excess Cdkn2b causes islet hypoplasia and diabetes.

    “Cdkn2b overexpression is also causally related to islet hypoplasia and diabetes in murine models (23).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [23] Moritani M Mol. Cell Endocrinol 2005 229 175 184 15607541 source

      Not checked in the free run.

  • May combine two claims.

    Interpretationreasoning

    Tight first-wave thresholds reduce false positives, yet many true risk variants would miss them.

    “The relatively strict thresholds imposed for SNP selection in the first wave (i.e. point-wise p<10−5) help to limit false discovery, but many genuine susceptibility variants will fail to reach them.”
    Quote in the paperSubstantial claim

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

    Premisecitation

    HHEX is a powerful biological candidate.

    “and is a powerful biological candidate (13, 14).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [13] Foley AC Mercola M Genes Dev 2005 19 387 396 15687261 PMC546516 source

      Not checked in the free run.

    • cites [14] Bort R Martinez-Barbera JP Beddington RS Zaret KS Development 2004 131 797 806 14736744 source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    The tumour suppressor CDKN2A produces p16INK4a, which blocks CDK4, a key controller of pancreatic beta cell replication.

    “CDKN2A is a known tumour suppressor and its product, p16INK4a, inhibits CDK4 (cyclin-dependent kinase 4), a powerful regulator of pancreatic beta cell replication (19-21).”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites [19] Rane SG Nat. Genet 1999 22 44 52 10319860 source

      Not checked in the free run.

    • cites [20] Mettus RV Rane SG Oncogene 2003 22 8413 8421 14627982 source

      Not checked in the free run.

    • cites [21] Marzo N Diabetologia 2004 47 686 694 15298346 source

      Not checked in the free run.

  • May combine two claims.

    Interpretationreasoning

    No signal for rs9369425 is apparent in FUSION, but this does not allow rejection of the association.

    “While no signal is apparent in the FUSION study, this does not allow us to reject this association.”
    Quote in the paperNot a substantial claim

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

    Study design and conductnone given

    Detecting an OR of 1.11 with 80% power at α=0.05 requires more than 3,000 case-control pairs.

    “For 80% power to detect an OR of 1.11 (α=0.05), over 3,000 case-control pairs are needed.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    The work offers insights into T2D pathogenesis, pointing to pancreatic beta cell development, regeneration and function as likely important pathways.

    “It also generates insights into T2D pathogenesis emphasizing the likely importance of pathways involved in pancreatic beta cell development, regeneration and function.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    The study validates the genome-wide indirect association mapping approach and demonstrates the value of aggressive data sharing.

    “Our study provides an important validation of the genome-wide indirect association mapping approach and a demonstration of the value of aggressive data sharing efforts.”
    Quote in the paperSubstantial claim
27 of 27

  • The paper relies on this without support.

    The DGI and FUSION genome-wide association data are comparable enough to the WTCCC and UK replication data to be combined in meta-analysis.

    “integration of data from the WTCCC scan and our own replication studies with similar information generated by the Diabetes Genetics Initiative [DGI] (6) and the Finland-United States Investigation of NIDDM Genetics [FUSION] (7)”
    “Across all studies, combined evidence for association at CDKAL1 is compelling (p~4.1×10−11).”

  • The paper relies on this without support.

    The replication sets RS1-RS3 are independent of the WTCCC samples and free from population stratification that would bias replication.

    “we attempted replication of selected signals in up to 3,757 additional cases and 5,346 controls (replication sets RS1-RS3).”
2 of 2

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?

  • SNPs in the second intron of the IGF2BP2 gene, particularly rs4402960 and rs1470579, are significantly associated with an increased risk of T2D across various populations.

    “These studies identified single-nucleotide polymorphisms (SNPs) in the second intron of the IGF2BP2 gene, particularly rs4402960 and rs1470579, as being significantly associated with an increased risk of T2D across various populations.”

    From IMPlications of IMP2 in RNA Biology and Disease (2025) · cites this paper as [71]

    The IGF2BP2 association passage (section labeled Abstract; page not provided) reports an increased T2D risk for rs4402960 and a strong combined association across studies. It does not identify rs1470579 or place the associated SNPs in the second intron; it describes a region spanning the promoter and first two exons. The copy therefore supports the association claim, but not those specific details. (LLM, conf=0.88)

    Experimental — accuracy not yet measured

  • Recent genome-wide association studies have detected SNPs significantly associated with breast cancer, colorectal cancer, type 2 diabetes, and heart disease, among other diseases.

    “This idea, combined with the development of efficient methods for surveying large numbers of SNPs, has led to the many recent genome-wide association (GWA) studies that have detected SNPs that are significantly associated with breast cancer79,80, colorectal cancer81,82, type 2 diabetes83–86 and heart disease87,88, among other diseases.”

    From Linkage disequilibrium--understanding the evolutionary past and mapping the medical future. (2008) · cites this paper as [84]

  • The association between IMP2 and T2D was first identified through genome-wide association studies (GWAS) in 2007, with three independent research groups demonstrating a strong link between genetic variants in the IGF2BP2 gene and T2D risk.

    “The association between IMP2 and T2D was first identified through genome-wide association studies (GWAS) in 2007, with three independent research groups demonstrating a strong link between genetic variants in the IGF2BP2 gene and T2D risk [69,70,71].”

    From IMPlications of IMP2 in RNA Biology and Disease (2025) · cites this paper as [71]

  • Genome-wide association studies link CDKN2A, CDKN2B, CDC123 and CDKAL1 to type 2 diabetes risk.

    “Strikingly, recent genome-wide association studies link the cyclin-dependent kinase inhibitors CDKN2A and CDKN2B (encoding p15Ink4b) and the cell cycle regulators CDC123 and CDKAL1 to the risk of type 2 diabetes [34–37].”

    From Expansion of β-cell mass in response to pregnancy (2010) · cites this paper as [34]

  • p16Ink4a is involved in aging and has been linked with type 2 diabetes in genome-wide studies.

    “The product of the Ink4a/Arf (Cdkn2a) locus, p16Ink4a (a negative regulator of CDK4 D-type cyclins), is thought to be involved in aging, and in genome-wide studies it has linked with type 2 diabetes (35,36).”

    From Age-Dependent Decline in β-Cell Proliferation Restricts the Capacity of β-Cell Regeneration in Mice (2009) · cites this paper as [35]

Cites

Flagged first, then by citations

Research works printed but not linked: 1

  1. Wellcome Trust Case Control Consortium Nature in press.

Cited by

Sorted by citations

Related

Authors
Zeggini, E., Weedon, M. N., Lindgren, C. M., Frayling, T. M., Elliott, K. S., Lango, H., Timpson, N. J., Perry, J. R. B., Rayner, N. W., Freathy, R. M., Barrett, J. C., Shields, B., Morris, A. P., Ellard, S., Groves, C. J., Harries, L. W., Marchini, J. L., Owen, K. R., Knight, B., Cardon, L. R., Walker, M., Hitman, G. A., Morris, A. D., Doney, A. S. F., McCarthy, M. I., & Hattersley, A. T.
Affiliations
  • Centre for Human Genetics (GB)
  • Oxford Centre for Diabetes, Endocrinology and Metabolism, University of Oxford, Churchill Hospital, Oxford, OX3 7LJ, UK.
Corresponding author
  • Zeggini, Eleftheria
  • Weedon, Michael N.
  • Lindgren, Cecilia M.
  • Frayling, Timothy M.
  • McCarthy, Mark I.
  • Hattersley, Andrew T.
ORCID
Venue
Science
Publisher
American Association for the Advancement of Science (AAAS)
Published
2007-06
Volume
316
Issue
5829
Pages
1336-1341
ISSN
  • 0036-8075 (ISSN-L)
  • 1095-9203
DOI
10.1126/science.1142364
PMC
PMC3772310
OpenAlex
W1944558458
Semantic Scholar
3acd41eeafe82b68094fdf0a487d8781b01fb790
Cited by
2,194
References
24
Citations by year
2026: 13 · 2025: 32 · 2024: 36 · 2023: 28 · 2022: 29 · 2021: 51 · 2020: 56 · 2019: 52 · 2018: 66 · 2017: 78 · 2016: 80 · 2015: 108 · 2014: 121 · 2013: 121 · 2012: 159
FWCI
178.50
Type
journal article
Language
en
Open access
Locations
  • Science (publishedVersion)
  • PubMed (publishedVersion)
  • Bristol Research (University of Bristol) (publishedVersion)
  • Europe PMC (PubMed Central) (acceptedVersion)
  • PubMed Central (acceptedVersion)
  • Discovery Research Portal (University of Dundee) (submittedVersion)
  • Research Portal (King's College London) (submittedVersion)
  • Edinburgh Research Explorer (University of Edinburgh) (submittedVersion, other-oa)
  • Research Portal (Queen's University Belfast) (submittedVersion)
  • Explore Bristol Research (submittedVersion)
  • UEA Digital Repository (University of East Anglia) (submittedVersion)
  • Newcastle University ePrints (Newcastle Univesity)
Funders
  • Wellcome Trust (090532, 518153, 083948)
  • Leverhulme Trust
  • Diabetes UK
  • Research Councils UK
  • European Commission
  • Medical Research Council (G0000934)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Genetics › Genetic Associations and Epidemiology
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Genetics › Diabetes and associated disorders
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Genetics › Genetic Mapping and Diversity in Plants and Animals
Keywords
type 2 diabetes, genetic susceptibility loci, CDKAL1, CDKN2A/CDKN2B, IGF2BP2, SLC30A8, beta-cell function, genetic architecture, Wellcome Trust Case Control Consortium
MeSH terms
Adult, Aged, Chromosome Mapping, Diabetes Mellitus, Type 2, Female, United Kingdom, Humans, Introns, Male, Middle Aged, Transcription Factors, Meta-Analysis as Topic, Genome, Human, Case-Control Studies, Homeodomain Proteins, Insulin-Like Growth Factor Binding Proteins, Genes, p16, Genetic Predisposition to Disease, Oligonucleotide Array Sequence Analysis, Polymorphism, Single Nucleotide

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