Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 51 of the 52 references it cites were found in a registry or an index.

Ag-driven CD8+ T cell clonal expansion is a prominent feature of MASH in humans and mice

Open access

Abstract

Background and Aims: Chronic liver disease due to metabolic dysfunction–associated steatohepatitis (MASH) is a rapidly increasing global epidemic. MASH progression is a consequence of the complex interplay between inflammatory insults and dysregulated hepatic immune responses. T lymphocytes have been shown to accumulate in the liver during MASH, but the cause and consequence of T cell accumulation in the liver remain unclear. Our study aimed to define the phenotype and T cell receptor diversity of T cells from human cirrhotic livers and an animal model of MASH to begin resolving their function in disease. Approach and Results: In these studies, we evaluated differences in T cell phenotype in the context of liver disease. Accordingly, we isolated liver resident T cell populations from humans with cirrhosis and from mice with diet-induced MASH. Using both 5’ single-cell sequencing and flow cytometry, we defined the phenotype and T cell receptor repertoire of liver resident T cells during health and disease. Conclusions: MASH-induced human cirrhosis and diet-induced MASH in mice resulted in the accumulation of activated and clonally expanded T cells in the liver. The clonally expanded T cells in the liver expressed markers of chronic antigenic stimulation, including PD1 , TIGIT , and TOX . Overall, this study establishes for the first time that T cells undergo Ag-dependent clonal expansion and functional differentiation during the progression of MASH. These studies could lead to the identification of antigenic targets that drive T cell activation, clonal expansion, and recruitment to the liver during MASH.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1097/hep.0000000000000971

Cites

51 of 51 references verified0 citation errors · 0 not checkable

Statistics

Figures

Graphical abstract
Graphical abstract CC BY-NC-ND · PubMed Central
FIGURE 1 scRNA-sequencing of human T cells reveals transcriptional and inferred proteomic differences between healthy individuals and those with cirrhosis. (A) UMAP plot of 51,624 human T cells colored by scTriangulate clusters. (B) UMAP plot colored by disease status of T cell sample origin. (C) Dot plot of selected gene (features) expression grouped by cluster assignment. The larger dot size represents a larger percent of cells expressing each feature and the darker dot color represents the higher average expression level of cells expressing each feature. T cell subsets are labeled, as determined by select feature expression by cluster assignment. (D) UMAP plot colored by T cell subset assignment. (E) Bar graph displaying the proportion of cells within each cluster from each disease status. (F) Dendrogram of Euclidean distances between each sample as calculated from the expression levels of the top 2000 variable genes. (G) Heatmap of inferred transcription factor activity using DecoupleR inference of CollecTRI regulatory networks. Red represents higher inferred transcription factor activity, while purple represents lower inferred activity. The top and right dendrograms represent similarities between samples based on inferred activity or transcription factors, respectively. Abbreviations: EOMES, Eomesodermin; LEF1, Lymphoid enhancer-binding factor 1; TCF7, Transcription factor 7; UMAP, uniform manifold approximation and projection.
FIGURE 1 scRNA-sequencing of human T cells reveals transcriptional and inferred proteomic differences between healthy individuals and those with cirrhosis. (A) UMAP plot of 51,624 human T cells colored by scTriangulate clusters. (B) UMAP plot colored by disease status of T cell sample origin. (C) Dot plot of selected gene (features) expression grouped by cluster assignment. The larger dot size represents a larger percent of cells expressing each feature and the darker dot color represents the higher average expression level of cells expressing each feature. T cell subsets are labeled, as determined by select feature expression by cluster assignment. (D) UMAP plot colored by T cell subset assignment. (E) Bar graph displaying the proportion of cells within each cluster from each disease status. (F) Dendrogram of Euclidean distances between each sample as calculated from the expression levels of the top 2000 variable genes. (G) Heatmap of inferred transcription factor activity using DecoupleR inference of CollecTRI regulatory networks. Red represents higher inferred transcription factor activity, while purple represents lower inferred activity. The top and right dendrograms represent similarities between samples based on inferred activity or transcription factors, respectively. Abbreviations: EOMES, Eomesodermin; LEF1, Lymphoid enhancer-binding factor 1; TCF7, Transcription factor 7; UMAP, uniform manifold approximation and projection.CC BY-NC-ND · PubMed Central
FIGURE 2 MASH-induced cirrhosis results in the recruitment of clonally expanded, chronically stimulated T cells to the liver. (A) Inverse Simpson (left), Chao (middle), and ACE (right) indices measuring sample diversity. Boxplots range from the 25th to 75th percentile with a horizontal line at the mean. (B) Top 25 most clonally expanded T cell groups as clustered by GLIPH and ranked by the largest cumulative proportion of T cell population from each disease status. Table displays the percentage of T cells from disease conditions expressing each clonal group. Darker highlighted percentage indicate the most clonally expanded, CD8+ T cell clonal group unique to each disease condition. Lighter highlighted percentages indicate other top 25 CD8+ clones unique to one disease condition. (C) UMAP plot displaying the most clonally expanded CD8+ T cell clonal group unique to each disease condition as identified in B. (D) Gene expression dot plot of selected gene (Features) expression for the most clonally expanded CD8+ T cell clonal group unique to each disease condition. The larger dot size represents a larger percent of cells expressing each feature and the darker dot color represents a higher average expression level of cells expressing each feature. Abbreviations: ACE, abundance-based coverage estimator; GLIPH, grouping of lymphocyte interactions by paratope hotspots; MASH, metabolic dysfunction–associated steatohepatitis; UMAP, uniform manifold approximation and projection.
FIGURE 2 MASH-induced cirrhosis results in the recruitment of clonally expanded, chronically stimulated T cells to the liver. (A) Inverse Simpson (left), Chao (middle), and ACE (right) indices measuring sample diversity. Boxplots range from the 25th to 75th percentile with a horizontal line at the mean. (B) Top 25 most clonally expanded T cell groups as clustered by GLIPH and ranked by the largest cumulative proportion of T cell population from each disease status. Table displays the percentage of T cells from disease conditions expressing each clonal group. Darker highlighted percentage indicate the most clonally expanded, CD8+ T cell clonal group unique to each disease condition. Lighter highlighted percentages indicate other top 25 CD8+ clones unique to one disease condition. (C) UMAP plot displaying the most clonally expanded CD8+ T cell clonal group unique to each disease condition as identified in B. (D) Gene expression dot plot of selected gene (Features) expression for the most clonally expanded CD8+ T cell clonal group unique to each disease condition. The larger dot size represents a larger percent of cells expressing each feature and the darker dot color represents a higher average expression level of cells expressing each feature. Abbreviations: ACE, abundance-based coverage estimator; GLIPH, grouping of lymphocyte interactions by paratope hotspots; MASH, metabolic dysfunction–associated steatohepatitis; UMAP, uniform manifold approximation and projection.CC BY-NC-ND · PubMed Central
FIGURE 3 HFHC-feeding recapitulates MASH phenotype in murine models. (A) Mice were fed a chow or HFHC diet for 27 or 31 weeks or fed an HFHC diet for 26 weeks and then switched to a chow diet for 5 weeks (reversal [Rev]) and investigated for a MASH phenotype. (B) Representative steatosis is shown by H&E staining of livers. (C) Total liver injury scoring, as determined by liver cell injury, inflammation, reactive changes, and steatosis from H&E-stained livers. (D) Inflammation scoring, as determined from H&E-stained livers. (E) Representative fibrosis is shown by PSR staining of livers. (F) Quantification of PSR staining with polarized light. Shown are representative graphs of experiments repeated on 2–3 occasions. n = 4–5/group for quantification above. n.s.=not statistically significant, *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001. Abbreviations: AU, arbitrary units; CV, central vein; H&E, hematoxylin and eosin; HFHC, high-fat, high-cholesterol; MASH, metabolic dysfunction–associated steatohepatitis; PSR, picrosirius red; PT, portal triad.
FIGURE 3 HFHC-feeding recapitulates MASH phenotype in murine models. (A) Mice were fed a chow or HFHC diet for 27 or 31 weeks or fed an HFHC diet for 26 weeks and then switched to a chow diet for 5 weeks (reversal [Rev]) and investigated for a MASH phenotype. (B) Representative steatosis is shown by H&E staining of livers. (C) Total liver injury scoring, as determined by liver cell injury, inflammation, reactive changes, and steatosis from H&E-stained livers. (D) Inflammation scoring, as determined from H&E-stained livers. (E) Representative fibrosis is shown by PSR staining of livers. (F) Quantification of PSR staining with polarized light. Shown are representative graphs of experiments repeated on 2–3 occasions. n = 4–5/group for quantification above. n.s.=not statistically significant, *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001. Abbreviations: AU, arbitrary units; CV, central vein; H&E, hematoxylin and eosin; HFHC, high-fat, high-cholesterol; MASH, metabolic dysfunction–associated steatohepatitis; PSR, picrosirius red; PT, portal triad.CC BY-NC-ND · PubMed Central
FIGURE 4 HFHC-feeding leads to CD8+ T cell accumulation in the liver through flow cytometry. (A) Flow cytometry gating strategy to identify CD45+ cells. (B) Representative flow cytometry contour plot showing T cell gating and percentages of Chow, HFHC, or reversal mice. (C) Quantification of B. (D) Representative flow cytometry contour plot showing CD8+ and CD4+ T cell gating and percentages. (E) Quantification of D. (F) Representative contour plot showing CD44hiCXCR6+ of CD8+ T cells and percentages. (G) Quantification of F. (H) Quantification of CD44hiCXCR6+ CD4+ T cells. Shown are representative graphs of experiments repeated on 2–3 occasions. n = 4–5/group for quantification above. n.s.=not statistically significant, *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001. Abbreviation: HFHC, high-fat, high-cholesterol.
FIGURE 4 HFHC-feeding leads to CD8+ T cell accumulation in the liver through flow cytometry. (A) Flow cytometry gating strategy to identify CD45+ cells. (B) Representative flow cytometry contour plot showing T cell gating and percentages of Chow, HFHC, or reversal mice. (C) Quantification of B. (D) Representative flow cytometry contour plot showing CD8+ and CD4+ T cell gating and percentages. (E) Quantification of D. (F) Representative contour plot showing CD44hiCXCR6+ of CD8+ T cells and percentages. (G) Quantification of F. (H) Quantification of CD44hiCXCR6+ CD4+ T cells. Shown are representative graphs of experiments repeated on 2–3 occasions. n = 4–5/group for quantification above. n.s.=not statistically significant, *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001. Abbreviation: HFHC, high-fat, high-cholesterol.CC BY-NC-ND · PubMed Central
FIGURE 5 scRNA-sequencing shows T cells derived from HFHC diet-fed mouse models of MASH differentially cluster from control mouse T cells. (A) UMAP plot of 64,808 liver resident murine T cells colored by scTriangulate clusters. (B) UMAP plot colored by sample origin of murine T cells after 25 weeks on diet (chow and HFHC sample 5–8), 31 weeks on diet (chow and HFHC samples 1–4), or reversal diet (reversal samples 1-4). (C) Dot plot of selected gene (features) expression grouped by cluster assignment. Plots are separated based on predominate Cd8 (left) or Cd4 (right) expression. The larger dot size represents a larger percent of cells expressing each feature and the darker dot color represents a higher average expression level of cells expressing each feature. (D) UMAP plot colored by T-cell subset assignment. (E) UMAP plot colored by diet (chow/control, HFHC/MASH, or Reversal) status of T-cell sample origin. (F) Bar graphs showing the proportion of T cells within each cluster, separated by T cells derived from (left) Chow samples, (middle) HFHC samples, and (right) Reversal samples. Abbreviations: HFHC, high-fat, high-cholesterol; MASH, metabolic dysfunction–associated steatohepatitis; scRNA, scRNA-seq, single-cell mRNA sequencing; UMAP, uniform manifold approximation and projection.
FIGURE 5 scRNA-sequencing shows T cells derived from HFHC diet-fed mouse models of MASH differentially cluster from control mouse T cells. (A) UMAP plot of 64,808 liver resident murine T cells colored by scTriangulate clusters. (B) UMAP plot colored by sample origin of murine T cells after 25 weeks on diet (chow and HFHC sample 5–8), 31 weeks on diet (chow and HFHC samples 1–4), or reversal diet (reversal samples 1-4). (C) Dot plot of selected gene (features) expression grouped by cluster assignment. Plots are separated based on predominate Cd8 (left) or Cd4 (right) expression. The larger dot size represents a larger percent of cells expressing each feature and the darker dot color represents a higher average expression level of cells expressing each feature. (D) UMAP plot colored by T-cell subset assignment. (E) UMAP plot colored by diet (chow/control, HFHC/MASH, or Reversal) status of T-cell sample origin. (F) Bar graphs showing the proportion of T cells within each cluster, separated by T cells derived from (left) Chow samples, (middle) HFHC samples, and (right) Reversal samples. Abbreviations: HFHC, high-fat, high-cholesterol; MASH, metabolic dysfunction–associated steatohepatitis; scRNA, scRNA-seq, single-cell mRNA sequencing; UMAP, uniform manifold approximation and projection.CC BY-NC-ND · PubMed Central
FIGURE 6 HFHC-derived T cells express high levels of transcriptional markers of T cell exhaustion and inferred pathways associated with T cell activation. (A) Violin plots of selected transcriptional marker normalized gene expression: Ccr7 (Ccr7), Sell (L-Selectin/Cd62L), Cxcr6 (Cxcr6), Pdcd1 (Pd1), Tigit (Tigit), and Tox (Tox), as grouped by diet and length of diet (25=25 wk, 31=31 wk). (B-C) fGSEA enrichment plot showing the enrichment score (top) and the AUC for the TCR signaling pathway. Genes are arranged based on their rank by enrichment in (B) HFHC (left) or chow (right) T cells and (C) reversal (left) or chow (right) T cells. p = adjusted p-value. Abbreviations: fGSEA, fast gene set enrichment analysis; HFHC, high-fat, high-cholesterol; NES, normalized enrichment score; TCR, T cell receptor.
FIGURE 6 HFHC-derived T cells express high levels of transcriptional markers of T cell exhaustion and inferred pathways associated with T cell activation. (A) Violin plots of selected transcriptional marker normalized gene expression: Ccr7 (Ccr7), Sell (L-Selectin/Cd62L), Cxcr6 (Cxcr6), Pdcd1 (Pd1), Tigit (Tigit), and Tox (Tox), as grouped by diet and length of diet (25=25 wk, 31=31 wk). (B-C) fGSEA enrichment plot showing the enrichment score (top) and the AUC for the TCR signaling pathway. Genes are arranged based on their rank by enrichment in (B) HFHC (left) or chow (right) T cells and (C) reversal (left) or chow (right) T cells. p = adjusted p-value. Abbreviations: fGSEA, fast gene set enrichment analysis; HFHC, high-fat, high-cholesterol; NES, normalized enrichment score; TCR, T cell receptor.CC BY-NC-ND · PubMed Central
FIGURE 7 HFHC feeding leads to the recruitment of clonally expanded CD8+ T cell with an exhausted phenotype. (A) Inverse Simpson (left), Chao (middle), and ACE (right) indices measuring sample diversity. Boxplots range from the 25th to 75th percentile with a horizontal line at the mean. (B) Bar plot showing the frequency of T cells with little to no clonal expansion (5 or less T cells within the clonal group), medium expansion (6–20 T cells within the clonal group), or large expansion (more than 20 T cells within the clonal group) within each diet. (C) UMAP plot colored by T cell expansion. Light grey=T cells with called TCR sequences shared with a total of 5 or less T cells. Clonally expanded T cells/T cells with called TCR sequences shared with more than 20 T cells are identified by diet. Blue=Chow, red=HFHC, yellow=reversal. (D) Bar graph showing the proportion of clonally expanded (clonal groups with > 20 T cells) T cells within each cluster, separated by expanded T cells derived from (top) Chow samples, (middle) HFHC samples, and (bottom) reversal samples. (E) Gene expression dot plot of selected gene (features) expression for clonally expanded (clonal groups with > 20 T cells) T cells, grouped by the diet origin of each T cell. The larger dot size represents a larger percent of cells expressing each feature and the darker dot color represents a higher average expression level of cells expressing each feature. Abbreviations: ACE, abundance-based coverage estimator; HFHC, high-fat, high-cholesterol; TCR, T-cell receptor; UMAP, uniform manifold approximation and projection.
FIGURE 7 HFHC feeding leads to the recruitment of clonally expanded CD8+ T cell with an exhausted phenotype. (A) Inverse Simpson (left), Chao (middle), and ACE (right) indices measuring sample diversity. Boxplots range from the 25th to 75th percentile with a horizontal line at the mean. (B) Bar plot showing the frequency of T cells with little to no clonal expansion (5 or less T cells within the clonal group), medium expansion (6–20 T cells within the clonal group), or large expansion (more than 20 T cells within the clonal group) within each diet. (C) UMAP plot colored by T cell expansion. Light grey=T cells with called TCR sequences shared with a total of 5 or less T cells. Clonally expanded T cells/T cells with called TCR sequences shared with more than 20 T cells are identified by diet. Blue=Chow, red=HFHC, yellow=reversal. (D) Bar graph showing the proportion of clonally expanded (clonal groups with > 20 T cells) T cells within each cluster, separated by expanded T cells derived from (top) Chow samples, (middle) HFHC samples, and (bottom) reversal samples. (E) Gene expression dot plot of selected gene (features) expression for clonally expanded (clonal groups with > 20 T cells) T cells, grouped by the diet origin of each T cell. The larger dot size represents a larger percent of cells expressing each feature and the darker dot color represents a higher average expression level of cells expressing each feature. Abbreviations: ACE, abundance-based coverage estimator; HFHC, high-fat, high-cholesterol; TCR, T-cell receptor; UMAP, uniform manifold approximation and projection.CC BY-NC-ND · PubMed Central
FIGURE 8 Clonally expanded HFHC-derived T cells increase expression of pathways associated with chronic T-cell stimulation and exhaustion. (A) Volcano plot of pathways significantly downregulated (blue), not significantly different (gray), or significantly upregulated (red) in clonally expanded HFHC T cells compared to clonally expanded Chow T cells as inferred by IPA. (B) Volcano plot of pathways significantly down-regulated (blue), not significantly different (gray), or significantly upregulated (yellow) in clonally expanded reversal T cells compared to clonally expanded Chow T cells as inferred by IPA. (C-D) fGSEA enrichment plot showing the enrichment score (top) and the AUC of genes relating to exhausted CD8+ T cells over memory CD8+ T cells. Genes are arranged based on their rank by enrichment in (C) expanded HFHC (left) or expanded Chow (right) T cells and (D) expanded reversal (left) or expanded Chow (right) T cells. p=adjusted p-value. Abbreviations: fGSEA, fast gene set enrichment analysis; HFHC, high-fat, high-cholesterol; IPA, ingenuity pathway analysis; NES, normalized enrichment score.
FIGURE 8 Clonally expanded HFHC-derived T cells increase expression of pathways associated with chronic T-cell stimulation and exhaustion. (A) Volcano plot of pathways significantly downregulated (blue), not significantly different (gray), or significantly upregulated (red) in clonally expanded HFHC T cells compared to clonally expanded Chow T cells as inferred by IPA. (B) Volcano plot of pathways significantly down-regulated (blue), not significantly different (gray), or significantly upregulated (yellow) in clonally expanded reversal T cells compared to clonally expanded Chow T cells as inferred by IPA. (C-D) fGSEA enrichment plot showing the enrichment score (top) and the AUC of genes relating to exhausted CD8+ T cells over memory CD8+ T cells. Genes are arranged based on their rank by enrichment in (C) expanded HFHC (left) or expanded Chow (right) T cells and (D) expanded reversal (left) or expanded Chow (right) T cells. p=adjusted p-value. Abbreviations: fGSEA, fast gene set enrichment analysis; HFHC, high-fat, high-cholesterol; IPA, ingenuity pathway analysis; NES, normalized enrichment score.CC BY-NC-ND · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

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?

Self-check

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

The self-check needs the paper’s full text.

Citations

Cites

Show all 52 references

Cited by

See all 35

Metadata

Authors
Burtis, A. E. C., DeNicola, D. M. C., Ferguson, M. E., Santos, R. G., Pinilla, C., Kriss, M. S., Orlicky, D. J., Tamburini, B. A. J., Gillen, A. E., & Burchill, M. A.
Affiliations
  • University of Colorado Anschutz (US)
  • Division of Gastroenterology and Hepatology, Department of Medicine, Aurora, Colorado, USA
  • Nova Southeastern University (US)
  • University of Minnesota (US)
Corresponding author
Burchill, Matthew A.
Venue
Hepatology
Published
2024-07-24
Volume
81
Issue
2
Pages
591-608
ISSN
  • 0270-9139 (ISSN-L)
  • 1527-3350
OpenAlex
W4400945319
Cited by
30
References
52
Citations by year
2026: 9 · 2025: 18 · 2024: 3
FWCI
9.15
Type
journal article
Language
en
Open access
Locations
  • Hepatology (publishedVersion, cc-by-nc-nd)
  • PubMed Central (publishedVersion, cc-by-nc-nd)
  • PubMed (publishedVersion)
Funders
  • U.S. Department of Veterans Affairs (IK2 BX004952)
  • National Institutes of Health
  • National Institute of Allergy and Infectious Diseases (R01AI155474)
  • National Institute of Diabetes and Digestive and Kidney Diseases (R01DK125595)
Topics
  • Health Sciences › Medicine › Hepatology › Liver Disease Diagnosis and Treatment
  • Health Sciences › Medicine › Hepatology › Liver physiology and pathology
  • Health Sciences › Medicine › Hepatology › Liver Diseases and Immunity
Keywords
nonalcoholic steatohepatitis, MASH, liver cirrhosis, CD8+ T cells, clonal expansion, T cell receptor repertoire, chronic antigenic stimulation, PD-1, TIGIT, TOX, T cell activation, diet-induced liver disease, hepatic immune response
MeSH terms
Animals, Disease Models, Animal, Fatty Liver, Female, Humans, Liver, Liver Cirrhosis, Male, Mice, Inbred C57BL, Receptors, Antigen, T-Cell, CD8-Positive T-Lymphocytes, Mice

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.

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

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Ag-driven CD8+ T cell clonal expansion is a prominent feature of MASH in humans and mice. https://app.cochat.ai/papers/10.1097/hep.0000000000000971
The paper
Burtis, A. E. C., DeNicola, D. M. C., Ferguson, M. E., Santos, R. G., Pinilla, C., Kriss, M. S., Orlicky, D. J., Tamburini, B. A. J., Gillen, A. E., & Burchill, M. A. (2025). Ag-driven CD8+ T cell clonal expansion is a prominent feature of MASH in humans and mice. Hepatology. https://doi.org/10.1097/hep.0000000000000971

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.