Pith. sign in

REVIEW 5 major objections 5 minor 51 references

Graph Structure Learning for Tumor Microenvironment with Cell Type Annotation from non-spatial scRNA-seq data

T0 review · 5 major / 5 minor · reviewed 2026-08-09 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read A self-learned cell graph can annotate tumor cell types from ordinary scRNA-seq data and simultaneously reveal cell-cell interactions in the tumor microenvironment.

desk verdict A competent integration of existing GSL and graph domain adaptation for scRNA-seq cell type annotation, with solid benchmarks and a circular interpretability validation that needs to be redone. read the letter →

arxiv 2502.02629 v1 pith:IRVYLBLV submitted 2025-02-04 q-bio.GN cs.AIcs.LG

Yu-An Huang , Yue-Chao Li , Hai-Ru You , Jie Pan , Xiyue Cao , Xinyuan Li , Zhi-An Huang , Zhu-Hong You This is my paper · ORCID
classification q-bio.GNcs.AIcs.LG
keywords single-cellRNAsequencingtumormicroenvironmentcelltypeannotationgraphneuralnetworkstructurelearningdomainadaptationcell-cellcommunicationligand-receptorinteractions
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper claims that a graph neural network that learns its own cell-cell graph, rather than relying on fixed similarity graphs or precompiled ligand-receptor lists, can both annotate cell types in non-spatial tumor scRNA-seq data and produce an interpretable picture of cell-cell communication in the tumor microenvironment. Tested by transferring labels between patient samples of the same cancer, the model attains an average accuracy of 84.83%, precision of 86.23%, recall of 81.51%, and F1 of 80.92% across leukemia, breast cancer, and colorectal cancer datasets, consistently two to four points above ten established baselines. The same learned graph, built without supervision from gene interaction knowledge, flags literature-validated gene pairs as significantly differentially expressed in relevant cell types. If correct, the result would make patient-level cell annotation and communication analysis feasible from ordinary scRNA-seq, without spatial assays or complete ligand-receptor databases.

What carries the argument

The load-bearing object is the learned cell-cell adjacency matrix $S$ produced by graph structure learning. It is initialized as a KNN adjacency and refined by alternating optimization of a classification loss, a graph fidelity loss with $\ell^1$ and nuclear-norm penalties for sparsity and low rank, and a feature-smoothness Laplacian term. A second mechanism, pairwise alignment from graph domain adaptation, computes per-class-pair density ratios $\gamma$ that reweight source edges so the source conditional edge distribution matches the target, and a label-shift ratio $\beta$ reweights the classifier loss. Together these mechanisms make the model's predictions transferable across patients and give the final query graph a biological reading.

What would settle it

Take any reference-query patient pair, estimate the density-ratio matrix $\gamma$ with the paper's equations, and compare the reweighted source edge-type distribution against the observed target edge-type distribution; if the two diverge and the reweighting lowers accuracy below the best baseline on that pair, the core transfer assumption is falsified. A second decisive check would be to run scGSL on a dataset with matched spatial transcriptomics and ask whether the learned cell-cell edges coincide with spatial adjacency or experimentally validated ligand-receptor contacts.

Watch

Extended reading notes

Core claim

On the paper's own terms, scGSL combines graph structure learning with pairwise-alignment graph domain adaptation so that a classifier trained on one patient's labeled cells can annotate another patient's unlabeled cells of the same cancer type. For each source graph, the graph structure learning module refines a KNN adjacency matrix into a symmetric matrix $S$ supervised by cell-type labels, regularized toward low rank and sparsity, and smoothed over gene-expression features; the query graph is initially built by KNN alone. The pairwise alignment step estimates density ratios $\gamma$ between source and target edge-type distributions, reweights the source graph, and iterates with pseudo-labels to reduce conditional structure shift and label shift. After prediction, the query graph is re-fed through the same graph structure learning module, and the resulting edges are read as putative cell-cell interactions. Across all patient-pair experiments, the method reports average accuracy 84.83%, and for literature-validated gene pairs such as SPRY1-LAT, BSG-ATP2B4, ATP2A3-SP1, and BIRC5-CASP6, Mann-Whitney U tests find significant expression differences between edge-connected and non-edge cells in most experiments, supporting the claim that the learned graph is biologically meaningful rather than merely predictive.

Load-bearing premise

The transfer step assumes that two patients of the same cancer type have the same conditional distribution of cell-cell edges given cell type, differing only by per-class density ratios, and that the model's pseudo-labels are accurate enough to estimate those ratios.

Editorial extensions

If this is right

  • Cell type annotation can be carried out from one patient's labeled cells to another's unlabeled cells of the same cancer, at accuracies roughly 2 to 4 points above ten existing methods.
  • Adding graph structure learning to the GNN raises average accuracy from 87.55% to 90.89% on the GSE132509 leukemia dataset, indicating that graph learning itself, not just the GNN, drives the improvement.
  • The final cell-cell graph can be inspected for cell-type-specific gene pairs, and known interactions show significant expression differences, implying the graph carries communication-relevant signal.
  • Because the query graph is learned after prediction, the same pipeline yields both labels and a candidate interaction map from a single non-spatial scRNA-seq run.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the learned graph edges really track communication, the method could be used to propose novel ligand-receptor or gene-pair hypotheses from non-spatial data alone; a prospective test would require validating newly discovered pairs with spatial or experimental evidence rather than checking already-known ones.
  • The pairwise alignment reweighting is the most fragile link: batch effects that change the conditional edge distribution beyond a density-ratio rescaling, or unreliable pseudo-labels for rare cell types, would silently distort the target graph; diagnostic plots of estimated versus observed edge distributions would make this visible.
  • The interpretability results are retrospective, testing genes already known for each cancer; a stronger claim would need the pipeline to surface reproducible new interactions across independent patient cohorts.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 5 minor

Summary. The paper proposes scGSL, a GNN-based method that combines graph structure learning (GSL) with graph domain adaptation (pairwise alignment, PA) to annotate cell types in non-spatial scRNA-seq data across patients of the same cancer type, and to construct cell-cell interaction graphs for tumor microenvironment analysis. The model is evaluated on 49,024 cells from 19 patients across three cancers (leukemia, breast cancer, colorectal cancer) in pairwise patient-transfer experiments, reporting an average accuracy of 84.83% against ten baseline methods. The authors further claim that the learned graphs identify biologically meaningful gene interactions in an unsupervised manner, validated by Mann-Whitney tests on six literature-known gene pairs.

Significance. If the reported accuracy gains and the biological interpretability claims were robust, scGSL would be a useful addition to single-cell analysis methods that extract cell-type annotations and cell communication hypotheses from non-spatial data. The paper provides a public code repository and uses a patient-pair transfer benchmark that is relevant to clinical applications. However, the current evidence is not fully convincing: the breast-cancer average accuracy is internally inconsistent, hyperparameters are selected on the benchmark data without a held-out protocol, no uncertainty estimates accompany the accuracy comparisons, and the biological validation is circular because graph edges are built from expression similarity. The interpretability claim, which is a core claimed novelty, is not supported by the presented tests. The idea of adapting graph-domain alignment to cross-patient cell graphs is plausible, but the paper needs stronger validation before its central claims can be accepted.

major comments (5)
  1. [Section 3.1, Table 2, Section 3.2] There is an internal inconsistency in the reported breast cancer average accuracy. Section 3.1 states the average accuracy for the Breast Invasive Carcinoma dataset as 83.56%, whereas the 'Average' row of Table 2 and the text in Section 3.2 both report 82.14%. Recomputing the average of the 'Ours' column in Table 2 gives 83.56%, which is also the value consistent with the overall unweighted average of 84.83% across all 40 patient-pair experiments. The manuscript should be corrected to use one consistent number, and the aggregate averaging scheme should be stated explicitly.
  2. [Section 3.2, Tables 1-3] The accuracy comparisons are reported as single point estimates with no variance or significance testing. Many of the claimed advantages over the second-best method are 2-4 percentage points, and with one run per patient pair it is unclear whether these differences are above noise. Provide standard deviations, confidence intervals, or paired significance tests (e.g., Wilcoxon signed-rank over the patient-pair experiments) for the comparisons.
  3. [Section 3.3, Fig. 5] Hyperparameters, including KNN neighbor fraction K=0.2% and embedding size ES=128, are selected by comparing performance on the same datasets that are later used in the benchmark (Fig. 5a). As no held-out validation or nested cross-validation is described, the reported gains over baselines may be inflated by selection on the test data. Please either use a separate validation set for hyperparameter tuning or demonstrate that all baselines were tuned under the same protocol.
  4. [Section 3.4, Tables 4-6] The Mann-Whitney validation is circular and does not support the claim that the model identifies biologically meaningful gene interactions. Cell-cell graphs are constructed from expression similarity via KNN and the feature smoothness loss (Eq. 4), so cells connected by an edge are transcriptionally similar by construction; any gene whose expression varies across cell states will differ between the connected and background groups, independent of whether the gene pair actually mediates cell-cell communication. A control with random gene pairs, or a permutation test that shuffles edge assignments while preserving expression, is needed. Additionally, the statement that the graphs are generated 'in an unsupervised manner' is misleading: the GSL objective in Eq. (1) explicitly uses cell-type labels, so the graph is supervised by cell type even if the specific validated gene pairs were not used in training.
  5. [Section 2.4, Eqs. (6)-(16)] The graph domain adaptation procedure relies on the pairwise alignment assumptions from ref. [25]: the conditional edge distribution P(A|Y) of the target is assumed to be a density-ratio reweighting of the source, and pseudo-labels from the classifier are used to estimate the quantities Sigma, nu, and C. On cross-patient tumor scRNA-seq data these assumptions may be violated by batch effects and biological shifts, but no diagnostics are provided. Please report experiments that assess whether the estimated gamma and beta actually improve target accuracy relative to an ablated version without reweighting, and evaluate pseudo-label reliability on the source domain.
minor comments (5)
  1. [Fig. 2 caption] The caption reads 'Flowchat of scGCN model' but should read 'Flowchart of scGSL model'.
  2. [Eq. (3), Section 2.3] Equation (3) defines the weights as ws and wl, but the following sentence states that 'the values of α and β were respectively set at 5e-4 and 1.5', introducing undefined symbols α and β. Please align the notation.
  3. [Abstract vs Section 2.1] The abstract reports a dataset of 49,020 cells, while Section 2.1 states 49,024 cells. These numbers should be reconciled.
  4. [Section 2.4, introduction] The paper states 'we introduce for the first time the concept of graph domain adaptation' in this context, but the pairwise alignment method is taken from ref. [25]. Please clarify that the contribution is the application to single-cell cell-cell graphs, not the introduction of the GDA concept itself.
  5. [Section 3.2, Tables 1-3] The tables report performance for each patient pair, but no information is given about cell type class distributions or which cell types are present. Adding the number of cells per class and the label distributions would help interpret the per-pair variability.

Circularity Check

1 steps flagged · score 6.0 of 10

Partial circularity: the biological interaction claim is validated by a Mann-Whitney test on groups defined by expression-derived edges, so the test restates graph construction rather than demonstrating gene-pair interactions.

  1. self definitional [Section 3.4 (Mann-Whitney validation paragraph); Eqs. (1)-(4) in Section 2.3]
    "We then conducted the Mann -Whitney U test to compare the expression levels of the two genes in the gene pair between these groups. If the p-value from this test is less than 0.05, it indicates that the differences in gene expression are statistically significant. Such a significant result would support the claim that the gene pair's expression differences, as learned by the scGSL from the cell-cell graph, are biologically meaningful."

    Edges are constructed from the expression matrix X: S is initialized as the KNN adjacency A and learned with min ||A-S||_F^2 + w_s||S||_1 + w_l||S||_* + w_n tr(X^T L X) (Eqs. 2-4). Thus connected cells are expression-similar by construction. The validation partitions cells into edge-connected vs background groups using these same expression-derived edges and tests individual gene expression differences between those groups. Any variable gene will tend to differ between the groups regardless of whether the tested pair mediates a real interaction, so the p-values are a restatement of the graph's construction objective, not evidence for gene-pair interactions. Additionally, the graph is not unsupervised: Eq. (1) supervises learning with cell-type labels Y via L_Cls(theta, A, X, Y).

full rationale

The cell type annotation claim is an empirical benchmark against ten baselines and is not circular: the accuracy, precision, recall, and F1 numbers compare predictions to held-out patient cell type labels, and the pairwise-alignment domain adaptation is imported from an external reference [25]. No fitted parameter is renamed as a prediction in the annotation pipeline. However, the abstract's second central claim, that scGSL identifies biologically meaningful gene interactions in an unsupervised manner, is supported only by the Section 3.4 Mann-Whitney validation, which is tautological. The graph S is initialized from KNN on X and explicitly optimized to preserve feature smoothness tr(X^T L X) and closeness to the KNN adjacency A; therefore the edge/no-edge groups are defined by expression similarity, and the Wilcoxon/Mann-Whitney test on individual genes is forced by construction rather than by genuine pair-level interaction. The 'unsupervised' wording is also contradicted by Eq. (1), where cell type labels Y supervise the GCN and the learned graph. Because this invalid interaction-validation is load-bearing for the paper's stated novelty, the circularity score is 6: one central claimed result reduces by construction, while the annotation benchmark itself remains independent.

Assumptions & free parameters 7 free parameters · 5 assumptions · 0 invented entities

The central model is an engineering combination of published components, so the ledger is dominated by hyperparameters and domain assumptions rather than invented entities. The model contributes no new biological entity; its correctness depends on the transferability of reference labels and on the validity of treating learned graph edges as candidate interactions.

free parameters (7)
  • KNN neighbor fraction K = 0.2% of total cell count
    Selected in ablation (Fig. 5a) because this setting delivered the highest ACC and F1; the choice is tuned on the same data rather than fixed by a biological prior.
  • Embedding size ES = 128
    Selected in the same ablation (Fig. 5a) as the best-performing configuration; other tested sizes gave lower ACC and F1.
  • Sparsity regularization weight (ws or alpha) = 5e-4
    Eq. (3) uses ws, but the text says alpha is set to 5e-4; the mapping is never defined. The value is chosen by hand and the central graph learning objective depends on it.
  • Low-rank regularization weight (wl or beta) = 1.5
    Eq. (3) uses wl, but the text says beta is set to 1.5. This weight controls the nuclear norm penalty that shapes the learned adjacency matrix.
  • Classification loss weight wc = 0.1
    Set by hand in Section 2.3, Eq. (5); controls how strongly cell type labels supervise the learned graph.
  • Node smoothness weight wn = 1
    Set by hand in Section 2.3, Eq. (5); controls feature smoothness regularization on the learned graph.
  • Number of highly variable genes = 3000
    Top 3000 HVGs are used as features; the number is a standard choice but is not justified for these datasets and affects the graph construction.
assumptions (5)
  • domain assumption Reference cell type labels YS are accurate, and the class-conditional graph structure transfers across patients.
    Supervised GSL and the GDA training in Sections 2.3 and 2.4 treat reference labels as ground truth. If the labels are noisy or the source graph structure does not transfer, the alignment in Eqs. (6) to (16) can distort predictions.
  • domain assumption A KNN graph with K=0.2% plus low-rank, sparsity, and smoothness regularization yields edges that correspond to real cell-cell communication.
    Section 3.4 treats edges in the learned matrix S as candidate biological interactions. This equivalence between graph proximity and signaling interaction is assumed, not measured.
  • domain assumption Pairwise alignment graph domain adaptation from ref. [25] remains valid when applied to cell-cell graphs from scRNA-seq.
    Section 2.4 applies the PA algorithm directly to reference-query patient graphs without modification or verification of its assumptions on this data type.
  • domain assumption Mann-Whitney U tests of gene expression differences between edge-connected and background cells are a valid proxy for biological gene interactions.
    Section 3.4 uses this test to validate known gene pairs. The test measures expression differences, not interaction strength, and is not benchmarked against random gene pairs.
  • standard math Standard proximal splitting and convex regularizers for the l1 and nuclear norms are applicable to the graph learning objective.
    Supplementary File 1 invokes Forward-Backward splitting and proximal operators for the two non-differentiable regularizers; these are accepted mathematical tools.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Graph Structure Learning for Tumor Microenvironment with Cell Type Annotation from non-spatial scRNA-seq data." pith.science (2026). https://pith.science/paper/IRVYLBLV

@misc{pith2026250202629,
  author       = {Pith},
  title        = {Pith review of: Graph Structure Learning for Tumor Microenvironment with Cell Type Annotation from non-spatial scRNA-seq data},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IRVYLBLV}},
  note         = {Machine review of arXiv:2502.02629}
}
read the original abstract

The exploration of cellular heterogeneity within the tumor microenvironment (TME) via single-cell RNA sequencing (scRNA-seq) is essential for understanding cancer progression and response to therapy. Current scRNA-seq approaches, however, lack spatial context and rely on incomplete datasets of ligand-receptor interactions (LRIs), limiting accurate cell type annotation and cell-cell communication (CCC) inference. This study addresses these challenges using a novel graph neural network (GNN) model that enhances cell type prediction and cell interaction analysis. Our study utilized a dataset consisting of 49,020 cells from 19 patients across three cancer types: Leukemia, Breast Invasive Carcinoma, and Colorectal Cancer. The proposed scGSL model demonstrated robust performance, achieving an average accuracy of 84.83%, precision of 86.23%, recall of 81.51%, and an F1 score of 80.92% across all datasets. These metrics represent a significant enhancement over existing methods, which typically exhibit lower performance metrics. Additionally, by reviewing existing literature on gene interactions within the TME, the scGSL model proves to robustly identify biologically meaningful gene interactions in an unsupervised manner, validated by significant expression differences in key gene pairs across various cancers. The source code and data used in this paper can be found in https://github.com/LiYuechao1998/scGSL.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

51 extracted references · 50 canonical work pages

  1. [25]

    Single-cell analysis of childhood leukemia reveals a link between developmental states and ribosomal protein expression as a source of intra -individual heterogeneity,

    M. Caron et al., "Single-cell analysis of childhood leukemia reveals a link between developmental states and ribosomal protein expression as a source of intra -individual heterogeneity," Scientific reports, vol. 10, no. 1, p. 8079, 2020

    work page 2020

  2. [1]

    Introduction Single-cell RNA-sequencing (scRNA-seq) technology has revolutionized our understanding of the cellular heterogeneity within the TME, a complex ecosystem integral to cancer progression and therapeutic response. The TME consists of a diverse array of cellular components including immune cells, endothe lial cells, and stromal cells, alongside mo...

    work page

  3. [2]

    gold standard

    and NicheNet [3]; (ii) machine learning prediction techniques, such as PyMINEr [4], 4 SoptSC [5], and SingleCellSignalR [6]; and (iii) spatial information-based inference tools like CellTalker [7], SpaOTsc [8], and histoCAT [9]. The computational pipeline for these methods typically begins with cell type annotation or cell clustering, followed by the orga...

    work page

  4. [3]

    NicheNet: modeling intercellular communication by linking ligands to target genes,

    R. Browaeys, W. Saelens, and Y. Saeys, "NicheNet: modeling intercellular communication by linking ligands to target genes," Nature methods, vol. 17, no. 2, pp. 159-162, 2020

    work page 2020

  5. [4]

    training set

    Methodology 2.1 scRNA-seq dataset In this study, we utilized a comprehensive scRNA -seq dataset to predict cell types across various diseases. The dataset encompasses a total of 49,024 cells derived from 19 individuals, averaging approximately 2,580 cells per individual. This rich dataset i ncludes cells from patients suffering from three distinct disease...

    work page

  6. [5]

    Cell –cell communication inference and analysis in the tumour microenvironments from single-cell transcriptomics: data resources and computational strategies,

    L. Peng et al. , "Cell –cell communication inference and analysis in the tumour microenvironments from single-cell transcriptomics: data resources and computational strategies," Briefings in bioinformatics, vol. 23, no. 4, p. bbac234, 2022

    work page 2022

  7. [6]

    Transcriptome analysis of individual stromal cell populations identifies stroma-tumor crosstalk in mouse lung cancer model,

    H. Choi et al., "Transcriptome analysis of individual stromal cell populations identifies stroma-tumor crosstalk in mouse lung cancer model," Cell reports, vol. 10, no. 7, pp. 1187-1201, 2015

    work page 2015

  8. [7]

    Immune landscape of viral -and carcinogen-driven head and neck cancer,

    A. R. Cillo et al., "Immune landscape of viral -and carcinogen-driven head and neck cancer," Immunity, vol. 52, no. 1, pp. 183-199. e9, 2020

    work page 2020

Show all 51 references
  1. [8]

    PyMINEr finds gene and autocrine-paracrine networks from human islet scRNA-Seq,

    S. R. Tyler et al., "PyMINEr finds gene and autocrine-paracrine networks from human islet scRNA-Seq," Cell reports, vol. 26, no. 7, pp. 1951-1964. e8, 2019

    1951

  2. [9]

    Cell lineage and communication network inference via optimization for single -cell transcriptomics,

    S. Wang, M. Karikomi, A. L. MacLean, and Q. Nie, "Cell lineage and communication network inference via optimization for single -cell transcriptomics," Nucleic acids research, vol. 47, no. 11, pp. e66-e66, 2019

    2019

  3. [10]

    SingleCellSignalR: inference of intercellular networks from single -cell transcriptomics,

    S. Cabello-Aguilar, M. Alame, F. Kon-Sun-Tack, C. Fau, M. Lacroix, and J. Colinge, "SingleCellSignalR: inference of intercellular networks from single -cell transcriptomics," Nucleic acids research, vol. 48, no. 10, pp. e55-e55, 2020

    2020

  4. [11]

    Comprehensive integration of single -cell data,

    T. Stuart et al., "Comprehensive integration of single -cell data," cell, vol. 177, no. 7, pp. 1888-1902. e21, 2019

    1902

  5. [12]

    Inferring spatial and signaling relationships between cells from single cell transcriptomic data,

    Z. Cang and Q. Nie, "Inferring spatial and signaling relationships between cells from single cell transcriptomic data," Nature communications, vol. 11, no. 1, p. 2084, 2020

    2020

  6. [13]

    histoCAT: analysis of cell phenotypes and interactions in multiplex image cytometry data,

    D. Schapiro et al., "histoCAT: analysis of cell phenotypes and interactions in multiplex image cytometry data," Nature methods, vol. 14, no. 9, pp. 873-876, 2017

    2017

  7. [14]

    scPriorGraph: constructing biosemantic cell–cell graphs with prior gene set selection for cell type identification from scRNA-seq data,

    X. Cao et al., "scPriorGraph: constructing biosemantic cell–cell graphs with prior gene set selection for cell type identification from scRNA-seq data," Genome Biology, vol. 25, no. 1, p. 207, 2024

    2024

  8. [15]

    scGCN is a graph convolutional networks algorithm for knowledge transfer in single cell omics,

    Q. Song, J. Su, and W. Zhang, "scGCN is a graph convolutional networks algorithm for knowledge transfer in single cell omics," Nature communications, vol. 12, no. 1, p. 3826, 2021

    2021

  9. [16]

    Reference -based analysis of lung single -cell sequencing reveals a transitional profibrotic macrophage,

    D. Aran et al. , "Reference -based analysis of lung single -cell sequencing reveals a transitional profibrotic macrophage," Nature immunology, vol. 20, no. 2, pp. 163-172, 26 2019

    2019

  10. [17]

    Despite the differing methodologies in constructing graphs, these approaches have confirmed that establishing links between cells is beneficial for predicting cell types

    calculates edge weights using AUCell [18] scoring of gene signaling pathways from databases like KEGG [19]. Despite the differing methodologies in constructing graphs, these approaches have confirmed that establishing links between cells is beneficial for predicting cell types...

  11. [18]

    Probabilistic cell -type assignment of single -cell RNA -seq for tumor microenvironment profiling,

    A. W. Zhang et al. , "Probabilistic cell -type assignment of single -cell RNA -seq for tumor microenvironment profiling," Nature methods, vol. 16, no. 10, pp. 1007 -1015, 2019

    2019

  12. [19]

    scCATCH: automatic annotation on cell types of clusters from single-cell RNA sequencing data,

    X. Shao, J. Liao, X. Lu, R. Xue, N. Ai, and X. Fan, "scCATCH: automatic annotation on cell types of clusters from single-cell RNA sequencing data," Iscience, vol. 23, no. 3, 2020

    2020

  13. [20]

    Inference and analysis of cell -cell communication using CellChat,

    S. Jin et al. , "Inference and analysis of cell -cell communication using CellChat," Nature communications, vol. 12, no. 1, p. 1088, 2021

    2021

  14. [21]

    scPML: pathway-based multi-view learning for cell type annotation from single-cell RNA-seq data,

    Z.-H. Du et al., "scPML: pathway-based multi-view learning for cell type annotation from single-cell RNA-seq data," Communications Biology, vol. 6, no. 1, p. 1268, 2023

    2023

  15. [22]

    SCENIC: single -cell regulatory network inference and clustering,

    S. Aibar et al. , "SCENIC: single -cell regulatory network inference and clustering," Nature methods, vol. 14, no. 11, pp. 1083-1086, 2017

    2017

  16. [23]

    KEGG: kyoto encyclopedia of genes and genomes,

    M. Kanehisa and S. Goto, "KEGG: kyoto encyclopedia of genes and genomes," Nucleic acids research, vol. 28, no. 1, pp. 27-30, 2000

    2000

  17. [24]

    C/EBPα and GATA-2 mutations induce bilineage acute erythroid leukemia through transformation of a neomorphic neutrophil -erythroid progenitor,

    C. Di Genua et al., "C/EBPα and GATA-2 mutations induce bilineage acute erythroid leukemia through transformation of a neomorphic neutrophil -erythroid progenitor," Cancer Cell, vol. 37, no. 5, pp. 690-704. e8, 2020

    2020

  18. [26]

    Risk-associated alterations in marrow T cells in pediatric leukemia,

    J. K. Bailur et al., "Risk-associated alterations in marrow T cells in pediatric leukemia," JCI insight, vol. 5, no. 16, 2020

    2020

  19. [27]

    A pan-cancer blueprint of the heterogeneous tumor microenvironment revealed by single-cell profiling,

    J. Qian et al., "A pan-cancer blueprint of the heterogeneous tumor microenvironment revealed by single-cell profiling," Cell research, vol. 30, no. 9, pp. 745-762, 2020

    2020

  20. [28]

    Cancer cells deploy lipocalin-2 to collect limiting iron in leptomeningeal metastasis,

    Y. Chi et al., "Cancer cells deploy lipocalin-2 to collect limiting iron in leptomeningeal metastasis," Science, vol. 369, no. 6501, pp. 276-282, 2020

    2020

  21. [29]

    Pairwise Alignment Improves Graph Domain Adaptation,

    S. Liu, D. Zou, H. Zhao, and P. Li, "Pairwise Alignment Improves Graph Domain Adaptation," arXiv preprint arXiv:2403.01092, 2024

    2024 arXiv

  22. [30]

    Exact matrix completion via convex optimization,

    E. Candes and B. Recht, "Exact matrix completion via convex optimization," Communications of the ACM, vol. 55, no. 6, pp. 111-119, 2012

    2012

  23. [31]

    MarkerCount: A stable, count-based cell type identifier for single -cell RNA -seq experiments,

    H. Kim, J. Lee, K. Kang, and S. Yoon, "MarkerCount: A stable, count-based cell type identifier for single -cell RNA -seq experiments," Computational and Structural Biotechnology Journal, vol. 20, pp. 3120-3132, 2022

    2022

  24. [32]

    CHETAH: a selective, hierarchical cell type identification method for single -cell RNA sequencing,

    J. K. De Kanter, P. Lijnzaad, T. Candelli, T. Margaritis, and F. C. Holstege, "CHETAH: a selective, hierarchical cell type identification method for single -cell RNA sequencing," Nucleic acids research, vol. 47, no. 16, pp. e95-e95, 2019

    2019

  25. [33]

    SingleCellNet: a computational tool to classify single cell RNA- Seq data across platforms and across species,

    Y. Tan and P. Cahan, "SingleCellNet: a computational tool to classify single cell RNA- Seq data across platforms and across species," Cell systems, vol. 9, no. 2, pp. 207-213. e2, 2019. 27

    2019

  26. [34]

    Learning for single-cell assignment,

    B. Duan et al., "Learning for single-cell assignment," Science advances, vol. 6, no. 44, p. eabd0855, 2020

    2020

  27. [35]

    scPred: accurate supervised method for cell-type classification from single-cell RNA-seq data,

    J. Alquicira -Hernandez, A. Sathe, H. P. Ji, Q. Nguyen, and J. E. Powell, "scPred: accurate supervised method for cell-type classification from single-cell RNA-seq data," Genome biology, vol. 20, pp. 1-17, 2019

    2019

  28. [36]

    scmap: projection of single -cell RNA-seq data across data sets,

    V. Y. Kiselev, A. Yiu, and M. Hemberg, "scmap: projection of single -cell RNA-seq data across data sets," Nature methods, vol. 15, no. 5, pp. 359-362, 2018

    2018

  29. [37]

    SciBet as a portable and fast single cell type identifier,

    C. Li et al. , "SciBet as a portable and fast single cell type identifier," Nature communications, vol. 11, no. 1, p. 1818, 2020

    2020

  30. [38]

    scClassify: sample size estimation and multiscale classification of cells using single and multiple reference,

    Y. Lin et al., "scClassify: sample size estimation and multiscale classification of cells using single and multiple reference," Molecular systems biology, vol. 16, no. 6, p. e9389, 2020

    2020

  31. [39]

    The reactome pathway knowledgebase 2022,

    M. Gillespie et al. , "The reactome pathway knowledgebase 2022," Nucleic acids research, vol. 50, no. D1, pp. D687-D692, 2022

    2022

  32. [40]

    WikiPathways: connecting communities,

    M. Martens et al., "WikiPathways: connecting communities," Nucleic acids research, vol. 49, no. D1, pp. D613-D621, 2021

    2021

  33. [41]

    Pathway Commons, a web resource for biological pathway data,

    E. G. Cerami et al., "Pathway Commons, a web resource for biological pathway data," Nucleic acids research, vol. 39, no. suppl_1, pp. D685-D690, 2010

    2010

  34. [42]

    CancerSEA: a cancer single-cell state atlas,

    H. Yuan et al., "CancerSEA: a cancer single-cell state atlas," Nucleic acids research, vol. 47, no. D1, pp. D900-D908, 2019

    2019

  35. [43]

    GCNG: graph convolutional networks for inferring gene interaction from spatial transcriptomics data,

    Y. Yuan and Z. Bar-Joseph, "GCNG: graph convolutional networks for inferring gene interaction from spatial transcriptomics data," Genome biology, vol. 21, pp. 1-16, 2020

    2020

  36. [44]

    Recruitment of Sprouty1 to immune synapse regulates T cell receptor signaling,

    J. S. Lee et al., "Recruitment of Sprouty1 to immune synapse regulates T cell receptor signaling," The Journal of Immunology, vol. 183, no. 11, pp. 7178-7186, 2009

    2009

  37. [45]

    Association of CD147 and calcium exporter PMCA4 uncouples IL- 2 expression from early TCR signaling,

    V. Supper et al., "Association of CD147 and calcium exporter PMCA4 uncouples IL- 2 expression from early TCR signaling," The Journal of Immunology, vol. 196, no. 3, pp. 1387-1399, 2016

    2016

  38. [46]

    Nifetepimine, a dihydropyrimidone, ensures CD4+ T cell survival in a tumor microenvironment by maneuvering sarco (endo) plasmic reticulum Ca2+ ATPase (SERCA),

    S. Ghosh et al., "Nifetepimine, a dihydropyrimidone, ensures CD4+ T cell survival in a tumor microenvironment by maneuvering sarco (endo) plasmic reticulum Ca2+ ATPase (SERCA)," Journal of Biological Chemistry, vol. 287, no. 39, pp. 32881 - 32896, 2012

    2012

  39. [47]

    Zap70 controls the interaction of talin with integrin to regulate the chemotactic directionality of T -cell migration,

    Y.-P. Lin, Y.-J. Cheng, J.-Y. Huang, H.-C. Lin, and B.-C. Yang, "Zap70 controls the interaction of talin with integrin to regulate the chemotactic directionality of T -cell migration," Molecular immunology, vol. 47, no. 11-12, pp. 2022-2029, 2010

    2022

  40. [48]

    Survivin-3B potentiates immune escape in cancer but also inhibits the toxicity of cancer chemotherapy,

    F. Vé gran et al., "Survivin-3B potentiates immune escape in cancer but also inhibits the toxicity of cancer chemotherapy," Cancer research, vol. 73, no. 17, pp. 5391-5401, 2013

    2013

  41. [49]

    Downregulation of SIRT7 by 5 -fluorouracil induces radiosensitivity in human colorectal cancer,

    M. Tang et al., "Downregulation of SIRT7 by 5 -fluorouracil induces radiosensitivity in human colorectal cancer," Theranostics, vol. 7, no. 5, p. 1346, 2017. 28 Supplementary File 1 Updating θ. To update θ, we fix S and eliminate terms that do not pertain to θ. Consequently, t...

    2017

  42. [50]

    Proximal splitting methods in signal processing

    Combettes, Patrick L., and Jean-Christophe Pesquet. "Proximal splitting methods in signal processing." Fixed-point algorithms for inverse problems in science and engineering (2011): 185-212

    2011

  43. [51]

    A generalized forward-backward splitting

    Raguet, Hugo, Jalal Fadili, and Gabriel Peyré. "A generalized forward-backward splitting." SIAM Journal on Imaging Sciences 6.3 (2013): 1199-1226. Initialize S←A Randomize θ while the error has not converged do 𝑆 ← 𝑆 − 𝜂 ∙ ∇𝑆(‖𝐴 − 𝑆‖𝐹 2 + 𝛾ℒ𝐶𝑙𝑠 + 𝜆ℒ𝑔𝑟𝑎𝑝ℎ) 𝑆 ← prox𝜂𝛽‖∙‖∗(𝑆) 𝑆 ←...

    2013

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.