CoChat

Single-cell entropy for accurate estimation of differentiation potency from a cell’s transcriptome

Open access
Teschendorff, Andrew E., Enver, Tariq · 2017 · Nature Communications

Published version of a preprint

The DOI matches its Crossref record; 57 of the 57 references it cites were found in a registry or an index.

Abstract

The ability to quantify differentiation potential of single cells is a task of critical importance. Here we demonstrate, using over 7,000 single-cell RNA-Seq profiles, that differentiation potency of a single cell can be approximated by computing the signalling promiscuity, or entropy, of a cell’s transcriptome in the context of an interaction network, without the need for feature selection. We show that signalling entropy provides a more accurate and robust potency estimate than other entropy-based measures, driven in part by a subtle positive correlation between the transcriptome and connectome. Signalling entropy identifies known cell subpopulations of varying potency and drug resistant cancer stem-cell phenotypes, including those derived from circulating tumour cells. It further reveals that expression heterogeneity within single-cell populations is regulated. In summary, signalling entropy allows in silico estimation of the differentiation potency and plasticity of single cells and bulk samples, providing a means to identify normal and cancer stem-cell phenotypes.

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/ncomms15599

Cites

57 of 57 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 The single-cell entropy (SCENT) algorithm. (a) Signalling entropy of single cells as a proxy to their differentiation potential in Waddington's landscape. Depicted on the left is a population of cells with cells occupying either a pluripotent (magenta), a progenitor (cyan) or a differentiated state (green). The potency state of each cell is determined by a complex function of the transcriptomic profile of the cell. For a given interaction between proteins i and k in the network, signalling in a given cell occurs with a probability pik∼xixk, defining a stochastic matrix P=(pik). In a pluripotent state, there is high demand for phenotypic plasticity, and so promiscuous signalling proteins (that is, those of high connectivity) are highly expressed (red coloured node) with all major differentiation pathways kept at a similar basal activity level (grey edges). The probability of signalling between protein i and k, pik, is therefore 1/ki where ki is the connectivity of protein i in the network. Thus the local signalling entropy around node i is maximal. In a differentiated state, commitment to a specific lineage (activation of a specific signalling pathway shown by red coloured node) means that most pij∼0, except when j=k, so that pik∼1. Thus, local signalling entropy around node i is close to zero. (b) Estimation of signalling entropy. An overall measure of signalling promiscuity of the cell is given mathematically by the signalling entropy rate (SR), which is a weighted average of local signaling entropies Si over all the genes/proteins in the network, with weights specified by π (the steady-state probability satisfying πP=π). It is proposed that SR provides a proxy to the elevation in Waddington's landscape, quantifying differentiation potential of cells (i.e the number of accessible cell-fates within a given lineage). (c) Quantification of intercellular heterogeneity and reconstruction of lineage trajectories. Estimation of signalling entropy at the single-cell level across a population of cells, allows the distribution of potency states in the population to be determined through Bayes mixture modelling which infers the optimal number of potency states. From this, the heterogeneity of potency states in a cell population is computed using Shannon's Index. To infer lineage trajectories, SCENT uses a clustering algorithm over dimensionally reduced scRNA-Seq profiles to infer co-expression clusters of cells. Dual assignment of cells to a potency state and co-expression cluster allows the identification of landmarks as bi-clusters in potency-coexpression space. Finally, partial correlations between the expression profiles of the landmarks are used to infer a lineage trajectory network diagram linking cell clusters according to expression similarity, with their height or elevation determined by their potency (signalling entropy).
Figure 1 The single-cell entropy (SCENT) algorithm. (a) Signalling entropy of single cells as a proxy to their differentiation potential in Waddington's landscape. Depicted on the left is a population of cells with cells occupying either a pluripotent (magenta), a progenitor (cyan) or a differentiated state (green). The potency state of each cell is determined by a complex function of the transcriptomic profile of the cell. For a given interaction between proteins i and k in the network, signalling in a given cell occurs with a probability pik∼xixk, defining a stochastic matrix P=(pik). In a pluripotent state, there is high demand for phenotypic plasticity, and so promiscuous signalling proteins (that is, those of high connectivity) are highly expressed (red coloured node) with all major differentiation pathways kept at a similar basal activity level (grey edges). The probability of signalling between protein i and k, pik, is therefore 1/ki where ki is the connectivity of protein i in the network. Thus the local signalling entropy around node i is maximal. In a differentiated state, commitment to a specific lineage (activation of a specific signalling pathway shown by red coloured node) means that most pij∼0, except when j=k, so that pik∼1. Thus, local signalling entropy around node i is close to zero. (b) Estimation of signalling entropy. An overall measure of signalling promiscuity of the cell is given mathematically by the signalling entropy rate (SR), which is a weighted average of local signaling entropies Si over all the genes/proteins in the network, with weights specified by π (the steady-state probability satisfying πP=π). It is proposed that SR provides a proxy to the elevation in Waddington's landscape, quantifying differentiation potential of cells (i.e the number of accessible cell-fates within a given lineage). (c) Quantification of intercellular heterogeneity and reconstruction of lineage trajectories. Estimation of signalling entropy at the single-cell level across a population of cells, allows the distribution of potency states in the population to be determined through Bayes mixture modelling which infers the optimal number of potency states. From this, the heterogeneity of potency states in a cell population is computed using Shannon's Index. To infer lineage trajectories, SCENT uses a clustering algorithm over dimensionally reduced scRNA-Seq profiles to infer co-expression clusters of cells. Dual assignment of cells to a potency state and co-expression cluster allows the identification of landmarks as bi-clusters in potency-coexpression space. Finally, partial correlations between the expression profiles of the landmarks are used to infer a lineage trajectory network diagram linking cell clusters according to expression similarity, with their height or elevation determined by their potency (signalling entropy).CC BY · PubMed Central
Figure 2 Signalling entropy correlates with differentiation potency of single cells. (a) Violin plots of the signalling entropy (SR) against cell-type (DEP, definite endoderm progenitors; EC, endothelial cells (mesoderm progenitor derivatives); HFF, human foreskin fibroblasts; hESC, human embryonic stem cells; NPC, neural progenitor cells; TB, trophoblast cells). Number of single cells in each class is indicated. Total number is 1,018. Wilcoxon rank-sum test P values between each cell-type (ranked in decreasing order of SR) are given. Diamond shaped data points correspond to the matched bulk samples. (b) Scatterplot of the signalling entropy (SR, y-axis) against an independent mRNA expression based pluripotency score (TPSC, x-axis) for all 1,018 single cells. Cell-type is indicated by colour. Spearman Correlation Coefficient (SCC) and associated P value are given. (c) Violin plot comparing the signalling entropy (SR) between the hESCs and all other (non-pluripotent) cells. P value is from a Wilcoxon rank-sum test. Inlet figure is the associated ROC curve, which includes the AUC value. (d) Violin plot of signalling entropy (SR) values for non-malignant single cells found in the microenvironment of melanomas. Number of single cells of each cell-type are given (B, B-cells; CAF, cancer-associated fibroblasts; EndC, endothelial cells; MacPH, macrophages; NK, natural killer cells; T, T-cells). Wilcoxon rank-sum test P values between EndC and MacPH, and between MacPH and all lymphocytes are given. (e) Signalling entropy (SR) as a function of differentiation stage within the mesoderm lineage. Differentiation stages include hESCs (pluripotent), mesoderm progenitors of endothelial cells (multipotent) and differentiated endothelial and white blood cells. Wilcoxon rank-sum test P values between successive stages are given. (f) ROC curves and AUC values for discriminating the progenitor and differentiated cells within the mesoderm lineage for signalling entropy (SR) and the t-test pluripotency score (TPSC). (g) Signalling entropy (SR, y-axis) as a function of time in a single-cell time course differentiation experiment, starting from hESCs at time=0 h (time of differentiation induction) into definite endoderm (which occurs from 72 h onwards). Number of single cells measured at each time point is given. Wilcoxon rank-sum test P values between the first four time points and 72 h, and between 72 and 98 h are given. (h) Signalling entropy (SR, y-axis) as a function of developmental stage in the differentiation of the distal mouse lung epithelium. Number of single cells measured at each stage is given. Wilcoxon rank-sum test P values between embryonic day 14 (E14) and all other stages are given. (i) Comparison of the SRs in c) (left panel) to the case where expression values are randomly reshuffled before computation of SR (middle panel). Right panels compare the corresponding ROC curves and AUC values. (j) As c, but now splitting the hESCs into cells from H1 and H9 lines, and including an additional independent set of 90 single hESCs profiled with a different NGS platform.
Figure 2 Signalling entropy correlates with differentiation potency of single cells. (a) Violin plots of the signalling entropy (SR) against cell-type (DEP, definite endoderm progenitors; EC, endothelial cells (mesoderm progenitor derivatives); HFF, human foreskin fibroblasts; hESC, human embryonic stem cells; NPC, neural progenitor cells; TB, trophoblast cells). Number of single cells in each class is indicated. Total number is 1,018. Wilcoxon rank-sum test P values between each cell-type (ranked in decreasing order of SR) are given. Diamond shaped data points correspond to the matched bulk samples. (b) Scatterplot of the signalling entropy (SR, y-axis) against an independent mRNA expression based pluripotency score (TPSC, x-axis) for all 1,018 single cells. Cell-type is indicated by colour. Spearman Correlation Coefficient (SCC) and associated P value are given. (c) Violin plot comparing the signalling entropy (SR) between the hESCs and all other (non-pluripotent) cells. P value is from a Wilcoxon rank-sum test. Inlet figure is the associated ROC curve, which includes the AUC value. (d) Violin plot of signalling entropy (SR) values for non-malignant single cells found in the microenvironment of melanomas. Number of single cells of each cell-type are given (B, B-cells; CAF, cancer-associated fibroblasts; EndC, endothelial cells; MacPH, macrophages; NK, natural killer cells; T, T-cells). Wilcoxon rank-sum test P values between EndC and MacPH, and between MacPH and all lymphocytes are given. (e) Signalling entropy (SR) as a function of differentiation stage within the mesoderm lineage. Differentiation stages include hESCs (pluripotent), mesoderm progenitors of endothelial cells (multipotent) and differentiated endothelial and white blood cells. Wilcoxon rank-sum test P values between successive stages are given. (f) ROC curves and AUC values for discriminating the progenitor and differentiated cells within the mesoderm lineage for signalling entropy (SR) and the t-test pluripotency score (TPSC). (g) Signalling entropy (SR, y-axis) as a function of time in a single-cell time course differentiation experiment, starting from hESCs at time=0 h (time of differentiation induction) into definite endoderm (which occurs from 72 h onwards). Number of single cells measured at each time point is given. Wilcoxon rank-sum test P values between the first four time points and 72 h, and between 72 and 98 h are given. (h) Signalling entropy (SR, y-axis) as a function of developmental stage in the differentiation of the distal mouse lung epithelium. Number of single cells measured at each stage is given. Wilcoxon rank-sum test P values between embryonic day 14 (E14) and all other stages are given. (i) Comparison of the SRs in c) (left panel) to the case where expression values are randomly reshuffled before computation of SR (middle panel). Right panels compare the corresponding ROC curves and AUC values. (j) As c, but now splitting the hESCs into cells from H1 and H9 lines, and including an additional independent set of 90 single hESCs profiled with a different NGS platform.CC BY · PubMed Central
Figure 3 SCENT identifies single-cell subpopulations of biological significance. (a) Fitted Gaussian mixture model to the signalling entropies of 1,018 single cells (scRNA-Seq data from Chu et al.) using a logit scale for the signalling entropies (x-axis, log2[SR/(1−SR)]). Bayesian Information Criterion predicted only two-states: a high energy/entropy pluripotent state (magenta-PS1) and a lower-energy non-pluripotent state (cyan-PS2). Number of cells categorized into each state is indicated in plot. (b) Barplot comparing, for each cell-type, the probability that a cell from this cell population is in the pluripotent (prob(Pl)) or non-pluripotent state (probe(NonPl)). Cell-types include human embryonic stem cells (hESCs), neural progenitor cells (NPCs), definite endoderm progenitors (DEPs), trophoblast cells (TBs), human foreskin fibroblasts (HFFs) and endothelial cells (ECs). (c) Barplot of the corresponding Shannon Index for each cell-population type. (d) Distribution of single-cell numbers between inferred potency states and co-expression clusters, as predicted by SCENT. In brown, we indicate ‘landmark clusters' which contain at least 5% of the total number of single cells. (e) Distribution of single-cell-types among the seven landmark clusters. (f) Inferred lineage trajectories between the seven landmarks which map to cell-types. Border colour indicates potency state: magenta=PS1, cyan=PS2. (g) Left panel: scatterplot of signalling entropy (SR) vs mRNA expression level of a neural stem/progenitor cell marker, HES1, for all NPCs. NPCs categorized as pluripotent are shown in magenta, NPCs categorized into a non-pluripotent state are shown in cyan. NPCs of high and low HES1 expression (as inferred using a partition-around-medoids algorithm with k=2) are indicated with triangles and squares, respectively. Right panel: corresponding boxplot comparing the differentiation potency (SR) of NPCs with low versus high HES1 expression. P value is from a one-tailed Wilcoxon rank-sum test.
Figure 3 SCENT identifies single-cell subpopulations of biological significance. (a) Fitted Gaussian mixture model to the signalling entropies of 1,018 single cells (scRNA-Seq data from Chu et al.) using a logit scale for the signalling entropies (x-axis, log2[SR/(1−SR)]). Bayesian Information Criterion predicted only two-states: a high energy/entropy pluripotent state (magenta-PS1) and a lower-energy non-pluripotent state (cyan-PS2). Number of cells categorized into each state is indicated in plot. (b) Barplot comparing, for each cell-type, the probability that a cell from this cell population is in the pluripotent (prob(Pl)) or non-pluripotent state (probe(NonPl)). Cell-types include human embryonic stem cells (hESCs), neural progenitor cells (NPCs), definite endoderm progenitors (DEPs), trophoblast cells (TBs), human foreskin fibroblasts (HFFs) and endothelial cells (ECs). (c) Barplot of the corresponding Shannon Index for each cell-population type. (d) Distribution of single-cell numbers between inferred potency states and co-expression clusters, as predicted by SCENT. In brown, we indicate ‘landmark clusters' which contain at least 5% of the total number of single cells. (e) Distribution of single-cell-types among the seven landmark clusters. (f) Inferred lineage trajectories between the seven landmarks which map to cell-types. Border colour indicates potency state: magenta=PS1, cyan=PS2. (g) Left panel: scatterplot of signalling entropy (SR) vs mRNA expression level of a neural stem/progenitor cell marker, HES1, for all NPCs. NPCs categorized as pluripotent are shown in magenta, NPCs categorized into a non-pluripotent state are shown in cyan. NPCs of high and low HES1 expression (as inferred using a partition-around-medoids algorithm with k=2) are indicated with triangles and squares, respectively. Right panel: corresponding boxplot comparing the differentiation potency (SR) of NPCs with low versus high HES1 expression. P value is from a one-tailed Wilcoxon rank-sum test.CC BY · PubMed Central
Figure 4 SCENT dissects distinct lineage trajectories in human myoblast differentiation. (a) Signalling entropy (SR) versus time point (0 h, 24 h, 48 h, 72 h) for a total of 372 single cells, collected during a time-course differentiation experiment of human myoblasts (scRNA-Seq from Trapnell et al.). Violin plots show the density distribution of SR values at each time point. P value is from a one-tailed Wilcox rank-sum test comparing timepoint 0 h to 24 h. (b) SCENT Gaussian Model fit to SR values predicts three potency states (PS1, PS2, PS3). (c) Probability distribution of potency states at each timepoint. (d) Co-expression heatmap of highly variable genes obtained by SCENT predicting three main clusters. Single cells have been ordered, first by cluster, then by potency state and finally by their time of sampling, as indicated. Landmarks are indicated by rectangular boxes, and distribution of single cells across landmarks and time points is provided in table. Genes have been clustered using hierarchical clustering. Genes that are markers of the different landmarks have been highlighted. (e) Inferred lineage trajectories between landmarks. Diagram illustrates an inferred two-phase trajectory, with one trajectory describing myoblasts of high potency (t=0, cyan circle) differentiating into skeletal muscle cells of intermediate potency (t=24 and 48) (blue circles) and a mixture of terminally differentiated and intermediate potency skeletal muscle cells (t=72) (grey and blue circle, respectively). A second trajectory/landmark describes a different cell-type (interstitial mesenchymal cells) whose intermediate potency state does not change during the time-course (blue stars).
Figure 4 SCENT dissects distinct lineage trajectories in human myoblast differentiation. (a) Signalling entropy (SR) versus time point (0 h, 24 h, 48 h, 72 h) for a total of 372 single cells, collected during a time-course differentiation experiment of human myoblasts (scRNA-Seq from Trapnell et al.). Violin plots show the density distribution of SR values at each time point. P value is from a one-tailed Wilcox rank-sum test comparing timepoint 0 h to 24 h. (b) SCENT Gaussian Model fit to SR values predicts three potency states (PS1, PS2, PS3). (c) Probability distribution of potency states at each timepoint. (d) Co-expression heatmap of highly variable genes obtained by SCENT predicting three main clusters. Single cells have been ordered, first by cluster, then by potency state and finally by their time of sampling, as indicated. Landmarks are indicated by rectangular boxes, and distribution of single cells across landmarks and time points is provided in table. Genes have been clustered using hierarchical clustering. Genes that are markers of the different landmarks have been highlighted. (e) Inferred lineage trajectories between landmarks. Diagram illustrates an inferred two-phase trajectory, with one trajectory describing myoblasts of high potency (t=0, cyan circle) differentiating into skeletal muscle cells of intermediate potency (t=24 and 48) (blue circles) and a mixture of terminally differentiated and intermediate potency skeletal muscle cells (t=72) (grey and blue circle, respectively). A second trajectory/landmark describes a different cell-type (interstitial mesenchymal cells) whose intermediate potency state does not change during the time-course (blue stars).CC BY · PubMed Central
Figure 5 Increased signalling entropy in cancer cells and identification of drug resistant cancer stem cells. (a) Boxplots of the signalling entropy (SR) for single melanoma cancer cells (c) compared to non-malignant (NotC) cells for three different melanoma patients (patient IDs given above each plot). Numbers of single cells are given below each boxplot. P value is from a Wilcoxon rank-sum test. (b) As a, but now pooled across all 12 patients. (c) Comparison of signalling entropy (SR) of 19 diagnostic acute myeloid leukaemia bulk samples to relapsed samples from the same patients. Wilcox rank sum test P value (one-tailed paired) is given. (d) Sorting of 96 single-AML cells from one patient according to signalling entropy and comparison of mRNA expression of AML CSC markers between low and high SR groups. P values from a one-tailed Wilcox test. (e) Comparison of signalling entropy (SR) of circulating tumour cells from metastatic prostate cancer patients who did not receive AR inhibitor treatment (UNTR) to those which developed resistance (RESIST). P value from a one-tailed Wilcox test. (f) Sorting of 73 single CTCs according to SCENT (signalling entropy, SR) into low and high SR groups. Correlation of gene expression of one putative CSC marker (ALDH7A1) with SR.
Figure 5 Increased signalling entropy in cancer cells and identification of drug resistant cancer stem cells. (a) Boxplots of the signalling entropy (SR) for single melanoma cancer cells (c) compared to non-malignant (NotC) cells for three different melanoma patients (patient IDs given above each plot). Numbers of single cells are given below each boxplot. P value is from a Wilcoxon rank-sum test. (b) As a, but now pooled across all 12 patients. (c) Comparison of signalling entropy (SR) of 19 diagnostic acute myeloid leukaemia bulk samples to relapsed samples from the same patients. Wilcox rank sum test P value (one-tailed paired) is given. (d) Sorting of 96 single-AML cells from one patient according to signalling entropy and comparison of mRNA expression of AML CSC markers between low and high SR groups. P values from a one-tailed Wilcox test. (e) Comparison of signalling entropy (SR) of circulating tumour cells from metastatic prostate cancer patients who did not receive AR inhibitor treatment (UNTR) to those which developed resistance (RESIST). P value from a one-tailed Wilcox test. (f) Sorting of 73 single CTCs according to SCENT (signalling entropy, SR) into low and high SR groups. Correlation of gene expression of one putative CSC marker (ALDH7A1) with SR.CC BY · PubMed Central
Figure 6 Signalling entropy predicts regulated expression heterogeneity of single-cell populations. (a) Definition of the measure of regulated expression heterogeneity (MRH). The MRH is a z-statistic, obtained by measuring the deviation of the signalling entropy (SR) of the bulk expression profile from the mean of single-cell entropies, taking into account the variability of single-cell entropies in the population. (b) Barplots of MRH for each cell-type population from Chu et al., representing the degree to which the signalling entropy of the cell population is higher than that of single cells. P values are from a one-tailed normal-deviation test. Dashed line indicates the line P=0.05. AvgBulkS compares the signalling entropy of the average expression over all bulk samples to that of the individual bulk samples, indicating that although the RHM is positive (signalling entropy increases), that it is not significantly higher than that of the individual bulk samples. (c) Scatterplot of the signalling entropy of bulk samples (y-axis), representing six cell-types (hESCs, NPCs, DEPs, TBs, HFFs, ECs) against the corresponding signalling entropies of these cell populations obtained by first averaging the expression profiles of single cells (‘Simulated Bulk', x-axis). R2 value and P value are given with green dashed line representing the fitted regression. Observe how the signalling entropy of bulk samples is always higher than that obtained from first averaging expression of single cells, in line with the fact that the assayed single cells are a subpopulation of the full bulk sample. (d) Left panel: comparison of the signalling entropy of an acute myeloid leukemia (AML) bulk sample (red line and point) to the signalling entropies of 96 single-AML cells (blue) from that bulk sample. P value is from a one-tailed normal deviation test. Right panel: comparison of the MRH value for the matched 96 single cells and bulk AML sample (SCs) to the MRH values obtained by comparing the signalling entropy of the average expression over 19 AML bulk samples to the signalling entropies of each individual AML bulk sample. The 19 AML bulk samples come in pairs, obtained at diagnosis (dgn) and relapse (rel), which are shown separately. P values are from a one-tailed normality deviation test.
Figure 6 Signalling entropy predicts regulated expression heterogeneity of single-cell populations. (a) Definition of the measure of regulated expression heterogeneity (MRH). The MRH is a z-statistic, obtained by measuring the deviation of the signalling entropy (SR) of the bulk expression profile from the mean of single-cell entropies, taking into account the variability of single-cell entropies in the population. (b) Barplots of MRH for each cell-type population from Chu et al., representing the degree to which the signalling entropy of the cell population is higher than that of single cells. P values are from a one-tailed normal-deviation test. Dashed line indicates the line P=0.05. AvgBulkS compares the signalling entropy of the average expression over all bulk samples to that of the individual bulk samples, indicating that although the RHM is positive (signalling entropy increases), that it is not significantly higher than that of the individual bulk samples. (c) Scatterplot of the signalling entropy of bulk samples (y-axis), representing six cell-types (hESCs, NPCs, DEPs, TBs, HFFs, ECs) against the corresponding signalling entropies of these cell populations obtained by first averaging the expression profiles of single cells (‘Simulated Bulk', x-axis). R2 value and P value are given with green dashed line representing the fitted regression. Observe how the signalling entropy of bulk samples is always higher than that obtained from first averaging expression of single cells, in line with the fact that the assayed single cells are a subpopulation of the full bulk sample. (d) Left panel: comparison of the signalling entropy of an acute myeloid leukemia (AML) bulk sample (red line and point) to the signalling entropies of 96 single-AML cells (blue) from that bulk sample. P value is from a one-tailed normal deviation test. Right panel: comparison of the MRH value for the matched 96 single cells and bulk AML sample (SCs) to the MRH values obtained by comparing the signalling entropy of the average expression over 19 AML bulk samples to the signalling entropies of each individual AML bulk sample. The 19 AML bulk samples come in pairs, obtained at diagnosis (dgn) and relapse (rel), which are shown separately. P values are from a one-tailed normality deviation test.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

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 57 references

Research works printed but not linked: 1

  1. Chu, L.-F. et al. Single-cell RNA-seq reveals novel regulators of human embryonic stem cell differentiation to definite endoderm. Genome Biol. 17, 173 (2016). · DOI · Paper page

Cited by

Items 301–355 of 355

Metadata

Authors
Teschendorff, A. E., & Enver, T.
Affiliations
  • Chinese Academy of Sciences (CN)
  • University College London (GB)
Corresponding author
Teschendorff, Andrew E.
Venue
Nature Communications
Published
2017-06-01
Volume
8
Issue
1
Pages
15599-15599
ISSN
2041-1723 (ISSN-L)
OpenAlex
W2620676965
Cited by
346
References
57
Citations by year
2026: 36 · 2025: 43 · 2024: 47 · 2023: 46 · 2022: 56 · 2021: 37 · 2020: 36 · 2019: 24 · 2018: 18 · 2017: 2
FWCI
8.94
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature Communications (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • RePEc: Research Papers in Economics (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • UCL Discovery (University College London) (submittedVersion)
Funders
  • Cancer Research UK (12796)
  • National Natural Science Foundation of China (:31571359, 31401120)
  • Medical Research Council (MR/M009033/1, MR/N000838/1)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Single-cell and spatial transcriptomics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Gene Regulatory Network Analysis
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Bioinformatics and Genomic Networks
Keywords
transcriptome, gene interaction networks, connectome, cancer stem cells, drug resistance, circulating tumor cells, cellular plasticity, cell subpopulations
MeSH terms
Cell Differentiation, Cell Line, Humans, RNA, Neoplastic Stem Cells, Signal Transduction, Sequence Analysis, RNA, Entropy, Computational Biology, Single-Cell Analysis, Transcriptome, Datasets as Topic

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

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.