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

The DOI matches its Crossref record; 68 of the 68 references it cites were found in a registry or an index.

Multi-modal quantification of pathway activity with MAYA

Open access
Landais, Yuna, Vallot, Céline · 2023 · Nature Communications

Abstract

Signaling pathways can be activated through various cascades of genes depending on cell identity and biological context. Single-cell atlases now provide the opportunity to inspect such complexity in health and disease. Yet, existing reference tools for pathway scoring resume activity of each pathway to one unique common metric across cell types. Here, we present MAYA, a computational method that enables the automatic detection and scoring of the diverse modes of activation of biological pathways across cell populations. MAYA improves the granularity of pathway analysis by detecting subgroups of genes within reference pathways, each characteristic of a cell population and how it activates a pathway. Using multiple single-cell datasets, we demonstrate the biological relevance of identified modes of activation, the robustness of MAYA to noisy pathway lists and batch effect. MAYA can also predict cell types starting from lists of reference markers in a cluster-free manner. Finally, we show that MAYA reveals common modes of pathway activation in tumor cells across patients, opening the perspective to discover shared therapeutic vulnerabilities.

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 · 10.1038/s41467-023-37410-2

Cites

68 of 68 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 MAYA overview. a MAYA takes as input a scRNA-Seq dataset and reference gene lists, and produces as output an activity matrix, with for each cell its activity score for each mode of every reference gene list. b Example of MAYA outputs: a heatmap to visualize the modes of activation of reference pathways, or a Uniform Manifold Approximation and Projection (UMAP) of the activity matrix to visualize cells according to any annotation (activity scores for different modes, predicted cell type or any user annotation).
Fig. 1 MAYA overview. a MAYA takes as input a scRNA-Seq dataset and reference gene lists, and produces as output an activity matrix, with for each cell its activity score for each mode of every reference gene list. b Example of MAYA outputs: a heatmap to visualize the modes of activation of reference pathways, or a Uniform Manifold Approximation and Projection (UMAP) of the activity matrix to visualize cells according to any annotation (activity scores for different modes, predicted cell type or any user annotation).CC BY · PubMed Central
Fig. 2 Activation modes of Hallmark pathways in kidney with MAYA. a Heatmap of activity matrix computed on kidney dataset with MSigDB Hallmark pathways, initial author annotation is indicated above heatmap. The two activation modes of Allograft Rejection are highlighted in bold and further described in the subsequent panels, and the four modes of activation of TNFA signaling via NFKB are further described in Supplementary Fig. 2. b Scatterplot of Mode 2 versus Mode 1 cell activity scores. Associated density histograms are indicated on the sides of the graph. c Heatmap of scaled gene expression for top 10 contributing genes for Mode1 (top) and Mode2 (bottom) of Allograft Rejection pathway, ordered by decreasing contribution for each. d UMAP representation of activity matrix of Hallmark pathways, cells are colored according to author annotation, or activity scores of modes 1 and 2 of Allograft rejection pathway. Specificity score of cell populations is displayed next to relevant clusters. e Heatmap of activity scores computed by Pagoda2, AUCell and MAYA for Allograft Rejection pathway, cells are grouped according to author annotation. f Barplot representation of the detection rate of modes 1 to 3 for the pathway Allograft Rejection when adding various numbers of random genes to the pathway gene list (n = 100 experiments each). Barplots are colored according to the cell population with the highest specificity score for the identified mode. g Jitter representation of specificity scores of modes 1 and 2 grouped by level of added noise, data points are colored according to author annotation. Specificity obtained for each mode with initial gene list is represented with a dashed line.
Fig. 2 Activation modes of Hallmark pathways in kidney with MAYA. a Heatmap of activity matrix computed on kidney dataset with MSigDB Hallmark pathways, initial author annotation is indicated above heatmap. The two activation modes of Allograft Rejection are highlighted in bold and further described in the subsequent panels, and the four modes of activation of TNFA signaling via NFKB are further described in Supplementary Fig. 2. b Scatterplot of Mode 2 versus Mode 1 cell activity scores. Associated density histograms are indicated on the sides of the graph. c Heatmap of scaled gene expression for top 10 contributing genes for Mode1 (top) and Mode2 (bottom) of Allograft Rejection pathway, ordered by decreasing contribution for each. d UMAP representation of activity matrix of Hallmark pathways, cells are colored according to author annotation, or activity scores of modes 1 and 2 of Allograft rejection pathway. Specificity score of cell populations is displayed next to relevant clusters. e Heatmap of activity scores computed by Pagoda2, AUCell and MAYA for Allograft Rejection pathway, cells are grouped according to author annotation. f Barplot representation of the detection rate of modes 1 to 3 for the pathway Allograft Rejection when adding various numbers of random genes to the pathway gene list (n = 100 experiments each). Barplots are colored according to the cell population with the highest specificity score for the identified mode. g Jitter representation of specificity scores of modes 1 and 2 grouped by level of added noise, data points are colored according to author annotation. Specificity obtained for each mode with initial gene list is represented with a dashed line.CC BY · PubMed Central
Fig. 3 KEGG and REACTOME activity in colon with MAYA. a Heatmap of scaled gene expression for top10 contributing genes for the three activation modes of KEGG Cell Adhesion Molecules pathway, ordered by decreasing contribution. b UMAP representation of activity matrix of KEGG pathways, cells are colored according to author annotation, or activity scores of the three modes of Cell Adhesion Molecules pathway. Specificity score of cell populations is displayed next to relevant clusters. c Heatmap of scaled gene expression for top10 contributing genes for the four activation modes of the Ion Channel Transport pathway, ordered by decreasing contribution. d UMAP representation of activity matrix of REACTOME pathways, cells are colored according to author annotation, or activity scores of the four modes of Ion Channel Transport pathway. Specificity score of cell populations is displayed next to relevant clusters.
Fig. 3 KEGG and REACTOME activity in colon with MAYA. a Heatmap of scaled gene expression for top10 contributing genes for the three activation modes of KEGG Cell Adhesion Molecules pathway, ordered by decreasing contribution. b UMAP representation of activity matrix of KEGG pathways, cells are colored according to author annotation, or activity scores of the three modes of Cell Adhesion Molecules pathway. Specificity score of cell populations is displayed next to relevant clusters. c Heatmap of scaled gene expression for top10 contributing genes for the four activation modes of the Ion Channel Transport pathway, ordered by decreasing contribution. d UMAP representation of activity matrix of REACTOME pathways, cells are colored according to author annotation, or activity scores of the four modes of Ion Channel Transport pathway. Specificity score of cell populations is displayed next to relevant clusters.CC BY · PubMed Central
Fig. 4 MAYA annotates cell type. a Gene-based UMAP representation of kidney dataset, cells are colored according to author annotation. b Heatmap representing for each author annotation (rows) the fraction of cells labeled with each MAYA annotation (columns) for the kidney dataset. c Overlaid jitter and boxplot representation of F1-scores for automatic annotation of the kidney dataset using Cell-ID, SCINA and MAYA, data points are colored according to author annotation; n = 5 cell types; Center line, median; box limits, upper and lower quartiles; whiskers, 1.5× interquartile range; Adjusted p-values from two-sided Wilcoxon test are symbolized with: *: <0.05, **: <0.01, ***: <0.001, ****: <0.0001 (pval = 0.011 for Cell-ID and for SCINA). d Gene-based UMAP representation of colon dataset, cells are colored according to author annotation. e Heatmap representing for each author annotation (rows) the fraction of cells labeled with each MAYA annotation (columns) for the colon dataset. f Overlaid jitter and boxplot representation of F1-scores for automatic annotation of the colon dataset using Cell-ID, SCINA and MAYA, data points are colored according to author annotation; n = 10 cell types; Center line, median; box limits, upper and lower quartiles; whiskers, 1.5× interquartile range; Adjusted p-values from two-sided Wilcoxon test are not significant. g UMAP representation of the larynx dataset, either gene-based or based on activity matrix of PanglaoDB cell-type markers lists, cells are colored according to cell type or to patient. h Overlaid jitter and boxplot representation of Shannon Diversity Index (SDI), for clusters derived from gene-based dimensionality reduction, Harmony dimensionality reduction, CCA dimensionality reduction and MAYA activity matrix of the larynx dataset; n = 12, 11, 7, and 7 independent clusters respectively; Center line, median; box limits, upper and lower quartiles; whiskers, 1.5× interquartile range; Adjusted p-values from two-sided Wilcoxon test are symbolized with: *: <0.05, **: <0.01, ***: <0.001, ****: <0.0001. Adjusted p-value is 0.026 for Gene-based relative to MAYA and non-significant for other comparisons.
Fig. 4 MAYA annotates cell type. a Gene-based UMAP representation of kidney dataset, cells are colored according to author annotation. b Heatmap representing for each author annotation (rows) the fraction of cells labeled with each MAYA annotation (columns) for the kidney dataset. c Overlaid jitter and boxplot representation of F1-scores for automatic annotation of the kidney dataset using Cell-ID, SCINA and MAYA, data points are colored according to author annotation; n = 5 cell types; Center line, median; box limits, upper and lower quartiles; whiskers, 1.5× interquartile range; Adjusted p-values from two-sided Wilcoxon test are symbolized with: *: <0.05, **: <0.01, ***: <0.001, ****: <0.0001 (pval = 0.011 for Cell-ID and for SCINA). d Gene-based UMAP representation of colon dataset, cells are colored according to author annotation. e Heatmap representing for each author annotation (rows) the fraction of cells labeled with each MAYA annotation (columns) for the colon dataset. f Overlaid jitter and boxplot representation of F1-scores for automatic annotation of the colon dataset using Cell-ID, SCINA and MAYA, data points are colored according to author annotation; n = 10 cell types; Center line, median; box limits, upper and lower quartiles; whiskers, 1.5× interquartile range; Adjusted p-values from two-sided Wilcoxon test are not significant. g UMAP representation of the larynx dataset, either gene-based or based on activity matrix of PanglaoDB cell-type markers lists, cells are colored according to cell type or to patient. h Overlaid jitter and boxplot representation of Shannon Diversity Index (SDI), for clusters derived from gene-based dimensionality reduction, Harmony dimensionality reduction, CCA dimensionality reduction and MAYA activity matrix of the larynx dataset; n = 12, 11, 7, and 7 independent clusters respectively; Center line, median; box limits, upper and lower quartiles; whiskers, 1.5× interquartile range; Adjusted p-values from two-sided Wilcoxon test are symbolized with: *: <0.05, **: <0.01, ***: <0.001, ****: <0.0001. Adjusted p-value is 0.026 for Gene-based relative to MAYA and non-significant for other comparisons.CC BY · PubMed Central
Fig. 5 MAYA detects pathway activation in tumors across patients. a UMAP representation of activity matrix of Hallmark pathways, cells are colored according to author annotation and patient. Clusters derived from activity matrix are displayed next to relevant groups of cells. b Overlaid jitter and boxplot representation of Shannon Diversity Index (SDI), for clusters derived from gene-based dimensionality reduction and MAYA activity matrix of the ovary dataset. Clusters corresponding to tumor cells are colored in pink; n = 11 and 4 independent clusters respectively; Center line, median; box limits, upper and lower quartiles; whiskers, 1.5× interquartile range. c Barplot representation of specificity scores of the top5 specific modes for the four most prevalent populations in the dataset. d Heatmap of activity scores of the three modes of the Hallmark Epithelial Mesenchymal Transition (EMT) pathway, initial author annotation is indicated above heatmap. Heatmap of scaled gene expression for top10 contributing genes for the three modes of EMT, ordered by decreasing contribution. e UMAP representation of activity matrix of Hallmark pathways, cells are colored according to activity scores of the three EMT modes. Specificity score of cell populations is displayed next to relevant clusters. Violin plots of activity scores for corresponding modes, grouped by author annotation (adjusted p-values from two-sided Wilcoxon test are symbolized with: *: <0.05, **: <0.01, ***: <0.001, ****: <0.0001). Heatmap of the activity scores for the modes of (f) Hallmark Estrogen Response Early pathway, (g) Hallmark Coagulation pathway and (h) KEGG Wnt pathway, cells are grouped according to author annotation. Heatmap of scaled gene expression for top10 contributing genes for corresponding modes, ordered by decreasing contribution.
Fig. 5 MAYA detects pathway activation in tumors across patients. a UMAP representation of activity matrix of Hallmark pathways, cells are colored according to author annotation and patient. Clusters derived from activity matrix are displayed next to relevant groups of cells. b Overlaid jitter and boxplot representation of Shannon Diversity Index (SDI), for clusters derived from gene-based dimensionality reduction and MAYA activity matrix of the ovary dataset. Clusters corresponding to tumor cells are colored in pink; n = 11 and 4 independent clusters respectively; Center line, median; box limits, upper and lower quartiles; whiskers, 1.5× interquartile range. c Barplot representation of specificity scores of the top5 specific modes for the four most prevalent populations in the dataset. d Heatmap of activity scores of the three modes of the Hallmark Epithelial Mesenchymal Transition (EMT) pathway, initial author annotation is indicated above heatmap. Heatmap of scaled gene expression for top10 contributing genes for the three modes of EMT, ordered by decreasing contribution. e UMAP representation of activity matrix of Hallmark pathways, cells are colored according to activity scores of the three EMT modes. Specificity score of cell populations is displayed next to relevant clusters. Violin plots of activity scores for corresponding modes, grouped by author annotation (adjusted p-values from two-sided Wilcoxon test are symbolized with: *: <0.05, **: <0.01, ***: <0.001, ****: <0.0001). Heatmap of the activity scores for the modes of (f) Hallmark Estrogen Response Early pathway, (g) Hallmark Coagulation pathway and (h) KEGG Wnt pathway, cells are grouped according to author annotation. Heatmap of scaled gene expression for top10 contributing genes for corresponding modes, ordered by decreasing contribution.CC BY · 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

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 68 references

Cited by

See all 9

Metadata

Authors
Landais, Y., & Vallot, C.
Affiliations
Centre National de la Recherche Scientifique (FR)
Corresponding author
Vallot, Céline
Venue
Nature Communications
Published
2023-03-25
Volume
14
Issue
1
Pages
1668-1668
ISSN
2041-1723 (ISSN-L)
OpenAlex
W4360938434
Cited by
9
References
68
Citations by year
2026: 1 · 2024: 5 · 2023: 3
FWCI
0.80
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature Communications (publishedVersion, cc-by)
  • bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, cc-by-nc-nd)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • HAL (submittedVersion, other-oa)
  • RePEc: Research Papers in Economics (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Single-cell and spatial transcriptomics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Bioinformatics and Genomic Networks
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Gene Regulatory Network Analysis
Keywords
signaling pathways, single-cell transcriptomics, cell type annotation, tumor cell populations, therapeutic vulnerabilities

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: Multi-modal quantification of pathway activity with MAYA. https://app.cochat.ai/papers/10.1038/s41467-023-37410-2
The paper
Landais, Y., & Vallot, C. (2023). Multi-modal quantification of pathway activity with MAYA. Nature Communications. https://doi.org/10.1038/s41467-023-37410-2

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.