pith. sign in

arxiv: 2509.12073 · v3 · submitted 2025-09-15 · 🧬 q-bio.GN · q-bio.QM

CEP-IP: An Explainable Framework for Cell Subpopulation Identification in Single-cell Transcriptomics

Kah Keng Wong This is my paper

Pith reviewed 2026-05-18 16:48 UTC · model grok-4.3

classification 🧬 q-bio.GN q-bio.QM
keywords single-cell RNA sequencingcell subpopulation identificationexplainable machine learninggene co-expressioninflection point analysisprostate cancerglioblastomapathway enrichment
0
0 comments X

The pith

CEP-IP identifies distinct cell subpopulations by splitting single-cell data at inflection points in the relationship between one gene and a module of co-expressed genes.

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

The paper introduces CEP-IP as an explainable framework that finds cell subpopulations showing strong pairwise monotonic relationships between a chosen gene of interest and its co-expressed partners. It identifies dual-filtered genes through combined Spearman and Kendall correlations, quantifies relationship strength per cell with generalized additive models using deviance explained, and ranks cells by their explanatory power. The transcriptional space is then divided at the inflection point into pre-IP and post-IP regions, each containing top-ranked cells whose pathway enrichments differ. This produces four subpopulations per sample in prostate cancer data and yields analogous splits in brain tissue and glioblastoma validation sets.

Core claim

CEP-IP applies inflection point analysis to the TRPM4-Ribo transcriptional space after CEP classification of top-ranked explanatory power cells, producing four subpopulations per prostate cancer patient where pre-IP TREP cells enrich for immune-related processes and post-IP TREP cells enrich for ribosomal, translation, and cell adhesion pathways; the same approach uncovers neuron-projection enrichments in the MTG dataset with the CARM1P1-DFG module and cell-division plus microtubule enrichments in the GBM dataset with the FOXM1-DFG module, along with continuous 3D trajectories for TREP cells.

What carries the argument

Inflection point analysis on the transcriptional space defined by deviance explained in generalized additive models of a gene-of-interest versus its dual-filtered co-expressed genes, after CEP classification of top-ranked explanatory power cells.

If this is right

  • TRPM4-Ribo modeling outperforms alternative gene-set modules at FDR below 0.05.
  • In each prostate cancer patient the four subpopulations separate into immune-enriched pre-IP TREP cells and ribosomal plus adhesion-enriched post-IP TREP cells.
  • In the middle temporal gyrus dataset the post-IP TREP cells using the CARM1P1-DFG module enrich for neuron projection ontologies.
  • In the glioblastoma dataset the FOXM1-DFG post-IP TREP cells enrich for cell division and microtubule pathways while 3D trajectory analysis shows continuous paths for TREP cells.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The ability to recover continuous trajectories in 3D that disappear in 2D embeddings indicates the method could complement standard dimensionality-reduction visualizations.
  • Because the same workflow succeeds with different GOI-DFG pairs across three independent datasets, it may be tried on additional tissues or disease states to expose relationship-driven cell states.
  • The framework's focus on monotonic gene-module strength per cell offers a route to link subpopulation identity directly to measurable changes in gene coordination rather than to overall expression levels alone.

Load-bearing premise

The inflection point detected in the gene-module transcriptional space marks a genuine biological transition between cell states whose pathway differences are not created by the modeling or filtering steps.

What would settle it

Re-running the pathway enrichment analysis on the pre-IP and post-IP groups after randomly reassigning cell labels or after using a non-inflection-based split yields no significant differences between the groups.

read the original abstract

Single-cell RNA sequencing (scRNA-seq) frameworks lack explainable approaches for identifying cell subpopulations harboring strong pairwise monotonic gene-module relationships between a gene of interest (GOI) and its co-expressed genes. CEP-IP is introduced as a novel explainable machine learning framework to address this gap. In the primary dataset, TRPM4 served as the GOI and its co-expressed ribosomal genes (Ribo) were identified via Spearman-Kendall dual-filter (i.e., dual-filtered gene, DFG). Generalized additive modeling quantified TRPM4-Ribo relationship strength via deviance explained (DE), which was then mapped to individual cells via CEP classification to identify top-ranked explanatory power (TREP) cells. TRPM4-Ribo transcriptional space was then stratified into pre-IP and post-IP regions using inflection point (IP) analysis, producing four subpopulations per patient for pathway analysis. TRPM4-Ribo modeling outperformed alternative gene set modules (FDR<0.05). In each prostate cancer (PCa) patient, CEP-IP yielded four cell subpopulations, where pre-IP TREP cells showed enrichment of immune-related processes, and post-IP TREP cells were enriched for ribosomal, translation, and cell adhesion pathways. Validation was performed in the Allen middle temporal gyrus (MTG) and Neftel glioblastoma (GBM) datasets. In the MTG dataset (CARM1P1-DFG module), post-IP TREP cells showed enrichment of neuron projection ontologies. In the GBM dataset, FOXM1 was the sole GOI yielding mesenchymal-state DFGs, with FOXM1-DFG post-IP TREP cells enriched for cell division and microtubule pathways; 3D trajectory analysis demonstrated continuous trajectories of TREP cells that were obscured in 2D embeddings. CEP-IP identifies biologically distinct cell subpopulations in three independent scRNA-seq datasets, and it may be applicable to other pairwise GOI-DFG modules in single-cell transcriptomics.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript introduces CEP-IP, a novel explainable framework for identifying cell subpopulations in scRNA-seq data that harbor strong pairwise monotonic relationships between a gene of interest (GOI) and its co-expressed dual-filtered genes (DFGs). The approach applies a Spearman-Kendall dual filter to select DFGs, fits a generalized additive model (GAM) to quantify relationship strength via deviance explained (DE), uses CEP classification to map DE to individual cells and identify top-ranked explanatory power (TREP) cells, then performs inflection point (IP) analysis to stratify the GOI-DFG transcriptional space into pre-IP and post-IP regions. This yields four subpopulations per patient for pathway enrichment analysis. The framework is demonstrated on TRPM4-Ribo modules in prostate cancer (PCa) data, CARM1P1-DFG in Allen MTG data, and FOXM1-DFG in Neftel GBM data, with reported outperformance over alternative modules (FDR<0.05), distinct enrichments (e.g., immune in pre-IP TREP, ribosomal/translation in post-IP TREP), and 3D trajectory analysis revealing continuous patterns obscured in 2D.

Significance. If the central results hold after robustness checks, CEP-IP would offer a useful explainable complement to standard clustering or trajectory methods by directly tying subpopulation labels to interpretable monotonic gene-module relationships. Strengths include explicit validation across three independent datasets, identification of biologically plausible pathway differences, and the 3D trajectory analysis that highlights continuous TREP cell dynamics. The potential extension to other GOI-DFG pairs could aid targeted studies of cellular heterogeneity in cancer and brain tissue. Credit is due for the sequential, non-circular application of standard statistical filters and models without self-referential fitting loops.

major comments (2)
  1. [Methods (GAM and IP stratification)] Methods section on GAM fitting and inflection point analysis: No sensitivity analysis is reported for the GAM smoothing parameter, the precise numerical method for locating the inflection point (second-derivative zero-crossing or change-point detection), or modest perturbations to the Spearman-Kendall dual-filter thresholds. This is load-bearing for the central claim because the reported immune enrichment in pre-IP TREP cells and ribosomal/translation enrichment in post-IP TREP cells could arise from smoothing artifacts or IP placement rather than stable biological transitions in the TRPM4-Ribo (or equivalent) space.
  2. [Results (PCa, MTG, GBM)] Results (PCa and validation datasets): The abstract states FDR<0.05 for module comparisons and pathway enrichments, yet no details are provided on error bars, multiple-testing correction across patients, or direct head-to-head comparison against established subpopulation methods (e.g., Seurat clustering or Monocle trajectories) on the identical data. This undermines the strength of the claim that CEP-IP yields biologically distinct subpopulations superior to alternatives.
minor comments (2)
  1. [Methods] Notation for CEP classification and TREP cell ranking is introduced without a clear equation or pseudocode; a small methods box or explicit formula would improve reproducibility.
  2. [Figures] Figure legends for the 3D trajectory plots should explicitly state the embedding method and how pre/post-IP labels are overlaid to avoid ambiguity in interpreting continuous trajectories.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive and detailed review of our manuscript. Their comments have highlighted important areas for improving the robustness and transparency of our methods and results. We address each major comment point by point below, with revisions incorporated where they strengthen the work without altering the core claims.

read point-by-point responses

  1. Referee: Methods section on GAM fitting and inflection point analysis: No sensitivity analysis is reported for the GAM smoothing parameter, the precise numerical method for locating the inflection point (second-derivative zero-crossing or change-point detection), or modest perturbations to the Spearman-Kendall dual-filter thresholds. This is load-bearing for the central claim because the reported immune enrichment in pre-IP TREP cells and ribosomal/translation enrichment in post-IP TREP cells could arise from smoothing artifacts or IP placement rather than stable biological transitions in the TRPM4-Ribo (or equivalent) space.

    Authors: We agree that sensitivity analyses are necessary to substantiate the stability of the reported enrichments. In the revised manuscript we have added a dedicated subsection to the Methods describing these checks. We varied the GAM smoothing parameter over a grid centered on the REML-selected value, implemented both second-derivative zero-crossing (via finite differences) and change-point detection for IP identification, and applied modest perturbations to the dual-filter thresholds (Spearman 0.35–0.55, Kendall 0.25–0.45). The pathway enrichments remain consistent across these ranges, as documented in new Supplementary Figures S6–S8. We have also clarified the exact numerical procedure used for IP detection in the main text. revision: yes

  2. Referee: Results (PCa and validation datasets): The abstract states FDR<0.05 for module comparisons and pathway enrichments, yet no details are provided on error bars, multiple-testing correction across patients, or direct head-to-head comparison against established subpopulation methods (e.g., Seurat clustering or Monocle trajectories) on the identical data. This undermines the strength of the claim that CEP-IP yields biologically distinct subpopulations superior to alternatives.

    Authors: The FDR<0.05 reported in the abstract applies to the outperformance of the TRPM4-Ribo module versus other tested gene modules, with Benjamini-Hochberg correction applied across modules. We have revised the Results to state this explicitly and added error bars (standard error across patients) to the relevant bar plots. Pathway enrichment was performed per patient with FDR control within each patient; no cross-patient multiple-testing correction was applied because enrichments were interpreted at the individual-patient level. CEP-IP is presented as an explainable complement to clustering or trajectory methods rather than a claim of overall superiority. To address the referee’s concern we have added a supplementary comparison (Supplementary Note 3) showing that Seurat clusters on the same PCa data do not recover the same monotonic-relationship-driven subpopulations or their distinct pathway signatures. This supports the unique contribution of CEP-IP while remaining within the scope of the original claims. revision: partial

Circularity Check

0 steps flagged

No significant circularity in the derivation chain

full rationale

The paper describes a sequential pipeline: Spearman-Kendall dual-filtering to select DFGs, GAM fitting to compute per-cell deviance explained, CEP classification to rank TREP cells, and inflection-point stratification of the resulting transcriptional coordinate followed by external GO enrichment. No equation or step reduces the final subpopulation labels or reported enrichments to quantities defined by the same fitted parameters used to generate the coordinate; the enrichments are computed independently against external databases after stratification. No self-citation is invoked as a load-bearing uniqueness theorem or ansatz, and the method relies on standard, externally verifiable statistical procedures without self-referential fitting or renaming of known results. The derivation remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

2 free parameters · 2 axioms · 2 invented entities

The framework rests on standard correlation and modeling assumptions plus several method-specific choices whose justification is not visible from the abstract.

free parameters (2)
  • Spearman-Kendall dual-filter thresholds
    Cutoffs used to define DFGs are chosen but not numerically specified in the abstract.
  • Deviance explained mapping parameters
    Parameters controlling how DE is mapped to individual cells via CEP classification.
axioms (2)
  • domain assumption Spearman and Kendall correlations reliably capture monotonic gene relationships in scRNA-seq count data.
    Invoked when applying the dual-filter to identify co-expressed genes.
  • domain assumption Generalized additive models provide a valid measure of relationship strength between GOI and DFGs.
    Used to compute deviance explained for cell ranking.
invented entities (2)
  • TREP cells no independent evidence
    purpose: Cells ranked highest by explanatory power of the GOI-DFG relationship
    New label introduced by the CEP classification step.
  • pre-IP and post-IP regions no independent evidence
    purpose: Stratification of transcriptional space at the inflection point
    New division created by the IP analysis step.

pith-pipeline@v0.9.0 · 5896 in / 1601 out tokens · 32869 ms · 2026-05-18T16:48:22.079105+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

87 extracted references · 87 canonical work pages

  1. [1]

    Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries

    Sung H, Ferlay J, Siegel RL et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 71:209- 249 (2021)

    work page 2020

  2. [2]

    Cancer statistics, 2023

    Siegel RL, Miller KD, Wagle NS et al. Cancer statistics, 2023. CA Cancer J Clin 73:17-48 (2023)

    work page 2023

  3. [3]

    Significantly reduced incidence and improved survival from prostate cancer over 25 years

    Serda-Ferrer BC, Sanvisens A, Fuentes -Raspall R et al. Significantly reduced incidence and improved survival from prostate cancer over 25 years. BMC Public Health 23:2552 (2023). CEP-IP to identify cell subpopulations 33

    work page 2023

  4. [4]

    No improvement noted in overall or cause-specific survival for men presenting with metastatic prostate cancer over a 20 -year period

    Wu JN, Fish KM, Evans CP et al. No improvement noted in overall or cause-specific survival for men presenting with metastatic prostate cancer over a 20 -year period. Cancer 120:818-823 (2014)

    work page 2014

  5. [5]

    Navigating the evolving therapeutic landscape in advanced prostate cancer

    Crawford ED, Petrylak D, Sartor O. Navigating the evolving therapeutic landscape in advanced prostate cancer. Urol Oncol 35S:S1-S13 (2017)

    work page 2017

  6. [6]

    The Ca2+-activated cation channel TRPM4 is regulated by phosphatidylinositol 4,5-biphosphate

    Nilius B, Mahieu F, Prenen J et al. The Ca2+-activated cation channel TRPM4 is regulated by phosphatidylinositol 4,5-biphosphate. EMBO J 25:467-478 (2006)

    work page 2006

  7. [7]

    TRPM4, a Ca2+ -activated nonselective cation channel in mouse sino-atrial node cells

    Demion M, Bois P, Launay P et al. TRPM4, a Ca2+ -activated nonselective cation channel in mouse sino-atrial node cells. Cardiovasc Res 73:531-538 (2007)

    work page 2007

  8. [8]

    Four TRPM4 cation channel mutations found in cardiac conduction diseases lead to altered protein stability

    Bianchi B, Ozhathil LC, Medeiros -Domingo A et al. Four TRPM4 cation channel mutations found in cardiac conduction diseases lead to altered protein stability. Front Physiol 9:177 (2018)

    work page 2018

  9. [9]

    Non-invasive multimodality imaging directly shows TRPM4 inhibition ameliorates stroke reperfusion injury

    Chen B, Ng G, Gao Y et al. Non-invasive multimodality imaging directly shows TRPM4 inhibition ameliorates stroke reperfusion injury. Transl Stroke Res 10:91-103 (2019)

    work page 2019

  10. [10]

    Knockdown of the TRPM4 channel alters cardiac electrophysiology and hemodynamics in a sex - and age-dependent manner in mice

    Arullampalam P, Essers MC, Boukenna M et al. Knockdown of the TRPM4 channel alters cardiac electrophysiology and hemodynamics in a sex - and age-dependent manner in mice. Physiol Rep 11:e15783 (2023)

    work page 2023

  11. [11]

    New role of TRPM4 channel in the cardiac excitation - contraction coupling in response to physiological and pathological hypertrophy in mouse

    Hedon C, Lambert K, Chakouri N et al. New role of TRPM4 channel in the cardiac excitation - contraction coupling in response to physiological and pathological hypertrophy in mouse. Prog Biophys Mol Biol 159:105-117 (2021)

    work page 2021

  12. [12]

    TRPM4 is overexpressed in breast cancer associated with estrogen response and epithelial-mesenchymal transition gene sets

    Wong KK, Hussain FA. TRPM4 is overexpressed in breast cancer associated with estrogen response and epithelial-mesenchymal transition gene sets. PLoS One 15:e0233884 (2020)

    work page 2020

  13. [13]

    TRPM4 channels contribute to Adriamycin chemoresistance in breast cancer cells

    Guo J, Liu F, Yang X et al. TRPM4 channels contribute to Adriamycin chemoresistance in breast cancer cells. Eur J Pharmacol 998:177637 (2025)

    work page 2025

  14. [14]

    Expression profiling analysis of the CD5+ diffuse large B- cell lymphoma subgroup: development of a CD5 signature

    Suguro M, Tagawa H, Kagami Y et al. Expression profiling analysis of the CD5+ diffuse large B- cell lymphoma subgroup: development of a CD5 signature. Cancer Sci 97:868-874 (2006)

    work page 2006

  15. [15]

    TRPM4 expression is associated with activated B cell subtype and poor survival in diffuse large B cell lymphoma

    Loo SK, Ch'ng ES, Md Salleh MS et al. TRPM4 expression is associated with activated B cell subtype and poor survival in diffuse large B cell lymphoma. Histopathology 71:98-111 (2017)

    work page 2017

  16. [16]

    A novel approach to identify driver genes involved in androgen- independent prostate cancer

    Schinke EN, Bii V, Nalla A et al. A novel approach to identify driver genes involved in androgen- independent prostate cancer. Mol Cancer 13:120 (2014)

    work page 2014

  17. [17]

    TRPM4 protein expression in prostate cancer: a novel tissue biomarker associated with risk of biochemical recurrence following radical prostatectomy

    Berg KD, Soldini D, Jung M et al. TRPM4 protein expression in prostate cancer: a novel tissue biomarker associated with risk of biochemical recurrence following radical prostatectomy. Virchows Arch 468:345-355 (2016)

    work page 2016

  18. [18]

    Transient receptor potential melastatin 4 channel contributes to migration of androgen -insensitive prostate cancer cells

    Holzmann C, Kappel S, Kilch T et al. Transient receptor potential melastatin 4 channel contributes to migration of androgen -insensitive prostate cancer cells. Oncotarget 6:41783- 41793 (2015)

    work page 2015

  19. [19]

    MicroRNA -150 suppresses epithelial -mesenchymal transition, invasion, and metastasis in prostate cancer through the TRPM4 -mediated beta-catenin signaling pathway

    Hong X, Yu JJ. MicroRNA -150 suppresses epithelial -mesenchymal transition, invasion, and metastasis in prostate cancer through the TRPM4 -mediated beta-catenin signaling pathway. Am J Physiol Cell Physiol 316:C463-C480 (2019)

    work page 2019

  20. [20]

    TRPM4 channel is involved in regulating epithelial to mesenchymal transition, migration, and invasion of prostate cancer cell lines

    Sagredo AI, Sagredo EA, Pola V et al. TRPM4 channel is involved in regulating epithelial to mesenchymal transition, migration, and invasion of prostate cancer cell lines. J Cell Physiol 234:2037-2050 (2019)

    work page 2037

  21. [21]

    The oncogenic roles of TRPM ion channels in cancer

    Wong KK, Banham AH, Yaacob NS et al. The oncogenic roles of TRPM ion channels in cancer. J Cell Physiol 234:14556-14573 (2019)

    work page 2019

  22. [22]

    TRPM4 contributes to cell death in prostate cancer tumor spheroids, and to extravasation and metastasis in a zebrafish xenograft model system

    Bochen F, Subedi S, La Manna F et al. TRPM4 contributes to cell death in prostate cancer tumor spheroids, and to extravasation and metastasis in a zebrafish xenograft model system. Mol Oncol 19:1299-1309 (2025)

    work page 2025

  23. [23]

    Small molecular inhibitors block TRPM4 currents in prostate cancer cells, with limited impact on cancer hallmark functions

    Borgstrom A, Hauert B, Kappel S et al. Small molecular inhibitors block TRPM4 currents in prostate cancer cells, with limited impact on cancer hallmark functions. J Mol Biol 433:166665 (2021). CEP-IP to identify cell subpopulations 34

    work page 2021

  24. [24]

    On confidence intervals for generalized additive models based on penalized regression splines

    Wood SN. On confidence intervals for generalized additive models based on penalized regression splines. Aust N Z J Stat 48:445-464 (2006)

    work page 2006

  25. [25]

    Fast stable restricted maximum likelihood and marginal likelihood estimation of semiparametric generalized linear models

    Wood SN. Fast stable restricted maximum likelihood and marginal likelihood estimation of semiparametric generalized linear models. J R Stat Soc Ser B Stat Methodol 73:3-36 (2011)

    work page 2011

  26. [26]

    Generalized additive models: an introduction with R

    Wood SN. Generalized additive models: an introduction with R. Second Edition Boca Raton: CRC Press (2017)

    work page 2017

  27. [27]

    Generalized additive models for large data sets

    Wood SN, Goude Y, Shaw S. Generalized additive models for large data sets. J R Stat Soc Series C Appl Stat 64:139-155 (2015)

    work page 2015

  28. [28]

    Hierarchical generalized additive models in ecology: an introduction with mgcv

    Pedersen EJ, Miller DL, Simpson GL et al. Hierarchical generalized additive models in ecology: an introduction with mgcv. PeerJ 7:e6876 (2019)

    work page 2019

  29. [29]

    Combined generalized additive model and random forest to evaluate the influence of environmental factors on phytoplankton biomass in a large eutrophic lake

    Zhang J, Zhi M, Zhang Y. Combined generalized additive model and random forest to evaluate the influence of environmental factors on phytoplankton biomass in a large eutrophic lake. Ecol Indic 130:108082 (2021)

    work page 2021

  30. [30]

    Electrical energy prediction of combined cycle power plant using gradient boosted generalized additive model

    Pachauri N, Ahn CW. Electrical energy prediction of combined cycle power plant using gradient boosted generalized additive model. IEEE Access 10:24566-24577 (2022)

    work page 2022

  31. [31]

    Regression and generalized additive model to enhance the performance of photovoltaic power ensemble predictors

    Sundararajan A, Ollis B. Regression and generalized additive model to enhance the performance of photovoltaic power ensemble predictors. IEEE Access 9:111899-111914 (2021)

    work page 2021

  32. [32]

    Generalized additive models for gigadata: modeling the UK black smoke network daily data

    Wood SN, Li Z, Shaddick G et al. Generalized additive models for gigadata: modeling the UK black smoke network daily data. JASA 112:1199-1210 (2017)

    work page 2017

  33. [33]

    Huang X, Ma W, Law C et al. Importance of applying mixed generalized additive model (MGAM) as a method for assessing the environmental health impacts: Ambient temperature and acute myocardial infarction (AMI), among elderly in Shanghai, China. PLoS One 16:e0255767 (2021)

    work page 2021

  34. [34]

    Study on the correlation between the number of mushroom poisoning cases and meteorological factors based on the generalized additive model in Guizhou Province, 2023

    Xiong S, Wu A, Weng L et al. Study on the correlation between the number of mushroom poisoning cases and meteorological factors based on the generalized additive model in Guizhou Province, 2023. BMC Public Health 24:2628 (2024)

    work page 2023

  35. [35]

    Trajectory-based differential expression analysis for single-cell sequencing data

    Van den Berge K, Roux de Bézieux H, Street K et al. Trajectory-based differential expression analysis for single-cell sequencing data. Nat Commun 11:1201 (2020)

    work page 2020

  36. [36]

    PseudotimeDE: inference of differential gene expression along cell pseudotime with well -calibrated p -values from single -cell RNA sequencing data

    Song D, Li JJ. PseudotimeDE: inference of differential gene expression along cell pseudotime with well -calibrated p -values from single -cell RNA sequencing data. Genome Biol 22:124 (2021)

    work page 2021

  37. [37]

    Single cell analysis of cribriform prostate cancer reveals cell intrinsic and tumor microenvironmental pathways of aggressive disease

    Wong HY, Sheng Q, Hesterberg AB et al. Single cell analysis of cribriform prostate cancer reveals cell intrinsic and tumor microenvironmental pathways of aggressive disease. Nat Commun 13:6036 (2022)

    work page 2022

  38. [38]

    Spatial reconstruction of single-cell gene expression data

    Satija R, Farrell JA, Gennert D et al. Spatial reconstruction of single-cell gene expression data. Nat Biotechnol 33:495-502 (2015)

    work page 2015

  39. [39]

    Hamid SS

    Wong KK, Ab. Hamid SS. Multiomics in silico analysis identifies TM4SF4 as a cell surface target in hepatocellular carcinoma. PLoS One 20:e0307048 (2025)

    work page 2025

  40. [40]

    ToppGene Suite for gene list enrichment analysis and candidate gene prioritization

    Chen J, Bardes EE, Aronow BJ et al. ToppGene Suite for gene list enrichment analysis and candidate gene prioritization. Nucleic Acids Res 37:W305-311 (2009)

    work page 2009

  41. [41]

    Inference and computation with generalized additive models and their extensions

    Wood SN. Inference and computation with generalized additive models and their extensions. TEST 29:307-339 (2020)

    work page 2020

  42. [42]

    Package ‘mgcv’ : Mixed GAM Computation Vehicle with Automatic Smoothness Estimation

    Wood SN. Package ‘mgcv’ : Mixed GAM Computation Vehicle with Automatic Smoothness Estimation. R package version 1.9.1. https://cran.r- project.org/web/packages/mgcv/mgcv.pdf (2025)

    work page 2025

  43. [43]

    Using Cliff’s delta as a non -parametric effect size measure: an accessible web app and R tutorial

    Meissel K, Yao ES. Using Cliff’s delta as a non -parametric effect size measure: an accessible web app and R tutorial. PARE 29 (2024)

    work page 2024

  44. [44]

    Canonical androgen response element motifs are tumor suppressive regulatory elements in the prostate

    Chen X, Augello MA, Liu D et al. Canonical androgen response element motifs are tumor suppressive regulatory elements in the prostate. Nat Commun 15:10675 (2024). CEP-IP to identify cell subpopulations 35

    work page 2024

  45. [45]

    Design, synthesis, and biological evaluation of 2 -(naphthalen-1-yloxy)- N-phenylacetamide derivatives as TRPM4 inhibitors for the treatment of prostate cancer

    Niu L, Liu H, Li X et al. Design, synthesis, and biological evaluation of 2 -(naphthalen-1-yloxy)- N-phenylacetamide derivatives as TRPM4 inhibitors for the treatment of prostate cancer. Bioorg Med Chem 98:117584 (2024)

    work page 2024

  46. [46]

    TRPM4 regulates Akt/GSK3 -beta activity and enhances beta -catenin signaling and cell proliferation in prostate cancer cells

    Sagredo AI, Sagredo EA, Cappelli C et al. TRPM4 regulates Akt/GSK3 -beta activity and enhances beta -catenin signaling and cell proliferation in prostate cancer cells. Mol Oncol 12:151-165 (2018)

    work page 2018

  47. [47]

    Decreased expression of TRPM4 is associated with unfavorable prognosis and aggressive progression of endometrial carcinoma

    Li XC, Cheng Y, Yang X et al. Decreased expression of TRPM4 is associated with unfavorable prognosis and aggressive progression of endometrial carcinoma. Am J Transl Res 12:3926- 3939 (2020)

    work page 2020

  48. [48]

    Aberrant TRPM4 expression in MLL -rearranged acute myeloid leukemia and its blockade induces cell cycle arrest via AKT/GLI1/Cyclin D1 pathway

    Wang F, Wu P, Gong S et al. Aberrant TRPM4 expression in MLL -rearranged acute myeloid leukemia and its blockade induces cell cycle arrest via AKT/GLI1/Cyclin D1 pathway. Cell Signal 72:109643 (2020)

    work page 2020

  49. [49]

    Deregulation of ribosomal proteins in human cancers

    El Khoury W, Nasr Z. Deregulation of ribosomal proteins in human cancers. Biosci Rep 41:BSR20211577 (2021)

    work page 2021

  50. [50]

    Ribosomal protein L22-like1 promotes prostate cancer progression by activating PI3K/Akt/mTOR signalling pathway

    Yi X, Zhang C, Liu B et al. Ribosomal protein L22-like1 promotes prostate cancer progression by activating PI3K/Akt/mTOR signalling pathway. J Cell Mol Med 27:403-411 (2023)

    work page 2023

  51. [51]

    CRISPR screening reveals that RNA helicase DDX41 triggers ribosome biogenesis and cancer progression through R -loop-mediated RPL/RPS transcription

    Li H, He Y, Jiang J et al. CRISPR screening reveals that RNA helicase DDX41 triggers ribosome biogenesis and cancer progression through R -loop-mediated RPL/RPS transcription. Nat Commun 16:7409 (2025)

    work page 2025

  52. [52]

    Reduction of QM protein expression correlates with tumor grade in prostatic adenocarcinoma

    Altinok G, Powell IJ, Che M et al. Reduction of QM protein expression correlates with tumor grade in prostatic adenocarcinoma. Prostate Cancer Prostatic Dis 9:77-82 (2006)

    work page 2006

  53. [53]

    RPS2: a novel therapeutic target in prostate cancer

    Wang M, Hu Y, Stearns ME. RPS2: a novel therapeutic target in prostate cancer. J Exp Clin Cancer Res 28:6 (2009)

    work page 2009

  54. [54]

    TRPM4 mRNA stabilization by METTL3 -mediated m6A modification promotes calcific aortic valve inflammation

    Wu J, Huang H, Yang W et al. TRPM4 mRNA stabilization by METTL3 -mediated m6A modification promotes calcific aortic valve inflammation. Heliyon 10:e31871 (2024)

    work page 2024

  55. [55]

    New insights in osteogenic differentiation of human valvular interstitial cells: Implication of transient receptor potential melastatin 4 (TRPM4)

    Aize M, Roussel BD, Boileve A et al. New insights in osteogenic differentiation of human valvular interstitial cells: Implication of transient receptor potential melastatin 4 (TRPM4). Arch Cardiovasc Dis 117:S196-S197 (2024)

    work page 2024

  56. [56]

    Decapping and decay of messenger RNA occur in cytoplasmic processing bodies

    Sheth U, Parker R. Decapping and decay of messenger RNA occur in cytoplasmic processing bodies. Science 300:805-808 (2003)

    work page 2003

  57. [57]

    Attard G, Murphy L, Clarke NW et al. Abiraterone acetate and prednisolone with or without enzalutamide for high-risk non-metastatic prostate cancer: a meta-analysis of primary results from two randomised controlled phase 3 trials of the STAMPEDE platform protocol. Lancet 399:447-460 (2022)

    work page 2022

  58. [58]

    MYC-mediated ribosomal gene expression sensitizes enzalutamide-resistant prostate cancer cells to EP300/CREBBP inhibitors

    Furlan T, Kirchmair A, Sampson N et al. MYC-mediated ribosomal gene expression sensitizes enzalutamide-resistant prostate cancer cells to EP300/CREBBP inhibitors. Am J Pathol 191:1094-1107 (2021)

    work page 2021

  59. [59]

    Caloric restriction enhances the efficacy of anti - androgen therapy in prostate cancer by inhibiting androgen receptor translation

    Cordova RA, Elbanna M, Rupert C et al. Caloric restriction enhances the efficacy of anti - androgen therapy in prostate cancer by inhibiting androgen receptor translation. Cancer Res (2025)

    work page 2025

  60. [60]

    The impact of androgen -induced translation in modulating androgen receptor activity

    Israel JS, Marcelin LM, Mehralivand S et al. The impact of androgen -induced translation in modulating androgen receptor activity. Biol Direct 19:111 (2024)

    work page 2024

  61. [61]

    TRPM4 Is a novel component of the adhesome required for focal adhesion disassembly, migration and contractility

    Caceres M, Ortiz L, Recabarren T et al. TRPM4 Is a novel component of the adhesome required for focal adhesion disassembly, migration and contractility. PLoS One 10:e0130540 (2015)

    work page 2015

  62. [62]

    EB1- and EB2-dependent anterograde trafficking of TRPM4 regulates focal adhesion turnover and cell invasion

    Blanco C, Morales D, Mogollones I et al. EB1- and EB2-dependent anterograde trafficking of TRPM4 regulates focal adhesion turnover and cell invasion. FASEB J 33:9434-9452 (2019)

    work page 2019

  63. [63]

    siRNA knockdown of ribosomal protein gene RPL19 abrogates the aggressive phenotype of human prostate cancer

    Bee A, Brewer D, Beesley C et al. siRNA knockdown of ribosomal protein gene RPL19 abrogates the aggressive phenotype of human prostate cancer. PLoS One 6:e22672 (2011)

    work page 2011

  64. [64]

    METTL3 enhances cell adhesion through stabilizing integrin beta1 mRNA via an m6A -HuR-dependent mechanism in prostatic carcinoma

    Li E, Wei B, Wang X et al. METTL3 enhances cell adhesion through stabilizing integrin beta1 mRNA via an m6A -HuR-dependent mechanism in prostatic carcinoma. Am J Cancer Res 10:1012-1025 (2020). CEP-IP to identify cell subpopulations 36

    work page 2020

  65. [65]

    Targeting MHC-I molecules for cancer: function, mechanism, and therapeutic prospects

    Wu X, Li T, Jiang R et al. Targeting MHC-I molecules for cancer: function, mechanism, and therapeutic prospects. Mol Cancer 22:194 (2023)

    work page 2023

  66. [66]

    Targeting MHC-I inhibitory pathways for cancer immunotherapy

    Wang J, Lu Q, Chen X et al. Targeting MHC-I inhibitory pathways for cancer immunotherapy. Trends Immunol 45:177-187 (2024)

    work page 2024

  67. [67]

    Identification of a binding site for small molecule inhibitors targeting human TRPM4

    Ekundayo B, Arullampalam P, Gerber CE et al. Identification of a binding site for small molecule inhibitors targeting human TRPM4. Nat Commun 16:833 (2025)

    work page 2025

  68. [68]

    Identification and development of TRPM4 antagonists to counteract neuronal excitotoxicity

    Binkle-Ladisch L, Pironet A, Zaliani A et al. Identification and development of TRPM4 antagonists to counteract neuronal excitotoxicity. iScience 27:111425 (2024)

    work page 2024

  69. [69]

    tRF-19-W4PU732S promotes breast cancer cell malignant activity by targeting inhibition of RPL27A (ribosomal protein -L27A)

    Zhang Z, Liu Z, Zhao W et al. tRF-19-W4PU732S promotes breast cancer cell malignant activity by targeting inhibition of RPL27A (ribosomal protein -L27A). Bioengineered 13:2087-2098 (2022)

    work page 2087

  70. [70]

    Detecting novel associations in large data sets

    Reshef DN, Reshef YA, Finucane HK et al. Detecting novel associations in large data sets. Science 334:1518-1524 (2011)

    work page 2011

  71. [71]

    A fast algorithm for computing distance correlation

    Chaudhuri A, Hu W. A fast algorithm for computing distance correlation. CSDA 135:15-24 (2019)

    work page 2019

  72. [72]

    A new coefficient of correlation

    Chatterjee S. A new coefficient of correlation. JASA 116:2009-2022 (2021)

    work page 2009

  73. [73]

    Advanced R statistical programming and data models: Analysis, machine learning, and visualization

    Wiley M, Wiley JF. Advanced R statistical programming and data models: Analysis, machine learning, and visualization. Berkeley, CA: Apress (2019)

    work page 2019

  74. [74]

    Noise regularization removes correlation artifacts in single -cell RNA-seq data preprocessing

    Zhang R, Atwal GS, Lim WK. Noise regularization removes correlation artifacts in single -cell RNA-seq data preprocessing. Patterns 2:100211 (2021)

    work page 2021

  75. [75]

    Enhancing biological signals and detection rates in single-cell RNA-seq experiments with cDNA library equalization

    Bacher R, Chu LF, Argus C et al. Enhancing biological signals and detection rates in single-cell RNA-seq experiments with cDNA library equalization. Nucleic Acids Res 50:e12 (2022)

    work page 2022

  76. [76]

    FOXP2-positive diffuse large B -cell lymphomas exhibit a poor response to R -CHOP therapy and distinct biological signatures

    Wong KK, Gascoyne DM, Soilleux EJ et al. FOXP2-positive diffuse large B -cell lymphomas exhibit a poor response to R -CHOP therapy and distinct biological signatures. Oncotarget 7:52940-52956 (2016)

    work page 2016

  77. [77]

    DNMT1 is associated with cell cycle and DNA replication gene sets in diffuse large B-cell lymphoma

    Loo SK, Ab Hamid SS, Musa M et al. DNMT1 is associated with cell cycle and DNA replication gene sets in diffuse large B-cell lymphoma. Pathol Res Pract 214:134-143 (2018)

    work page 2018

  78. [78]

    FOXP1 suppresses immune response signatures and MHC class II expression in activated B-cell-like diffuse large B-cell lymphomas

    Brown PJ, Wong KK, Felce SL et al. FOXP1 suppresses immune response signatures and MHC class II expression in activated B-cell-like diffuse large B-cell lymphomas. Leukemia 30:605-616 (2016)

    work page 2016

  79. [79]

    Associating transcription factors to single-cell trajectories with DREAMIT

    Maulding ND, Seninge L, Stuart JM. Associating transcription factors to single-cell trajectories with DREAMIT. Genome Biol 25:220 (2024)

    work page 2024

  80. [80]

    A cell type and state specific gene regulation network inference method for immune regulatory analysis

    Li X, Rao K, Chen C et al. A cell type and state specific gene regulation network inference method for immune regulatory analysis. NPJ Syst Biol Appl 11:94 (2025)

    work page 2025

Showing first 80 references.