CFMF: A Clustering‐Free Cell Marker Finder for Single‐Cell Transcriptomic Data
IET Systems Biology · 2026-01 · 20 (1)
DOI: 10.1049/syb2.70065 ↗Kurz gefasst
Fachartikel in der Zeitschrift IET Systems Biology, online erschienen am 30.03.2026. Verfasst von Shumin Yin und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Medical Dataology, School of Public Health, Cheeloo College of Medicine, Shandong University, Jinan, China. Hauptthemen laut MEDLINE: Single-Cell Analysis, Transcriptome, Computational Biology. Die Arbeit nennt 54 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: National Natural Science Foundation of China, Science Fund for Distinguished Oversea Young Scholars of Shandong Province, Taishan Young Scholar Program of Shandong Province.
Abstract (Originalsprache)
The discovery of novel cell types and their marker genes is a fundamental goal of single‐cell RNA sequencing (scRNA‐seq). Nevertheless, most conventional methods rely on cell clustering, a process in which parameter selection is often non‐trivial and subjective. To address such limitations, we developed Clustering‐Free Cell Marker Finder (CFMF), a novel computational framework that enables the discovery of marker genes without clustering. We first validated CFMF using the PBMC3K dataset and found that it not only recovered canonical marker genes, but also uncovered novel marker genes. When validating with glioblastoma datasets, CFMF successfully recapitulated established cellular state signatures. Using human normal lung scRNA‐seq datasets, we demonstrated its superior sensitivity in rare cell type detection even at very low prevalence. We utilised CFMF to perform integrative analysis on colorectal cancer scRNA‐seq datasets and defined seven distinct transcriptional states including an LGR5 + state that is strongly associated with metastasis. Finally, we systematically benchmarked CFMF against widely used gene selection methods across multiple scRNA‐seq datasets and revealed that CFMF appeared to more effectively identify biologically meaningful genes. In summary, we provide a powerful framework for the discovery of marker genes and the analysis of tumour heterogeneity, which may facilitate our understanding of complex biological systems.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- Department of Medical Dataology, School of Public Health, Cheeloo College of Medicine, Shandong University, Jinan, China.
ORCID-Profile: Shumin Yin
Förderung
- National Natural Science Foundation of China
- Science Fund for Distinguished Oversea Young Scholars of Shandong Province
- Taishan Young Scholar Program of Shandong Province
- Cheeloo Young Scholar Program of Shandong University
Zitieren
APA
Yin, S., & Liu, K. (2026). CFMF: A Clustering‐Free Cell Marker Finder for Single‐Cell Transcriptomic Data. IET Systems Biology, 20(1). https://doi.org/10.1049/syb2.70065
Vancouver
Yin S, Liu K. CFMF: A Clustering‐Free Cell Marker Finder for Single‐Cell Transcriptomic Data. IET Systems Biology. 2026;20(1). doi:10.1049/syb2.70065
BibTeX
@article{Yin2026,
author = {Yin, Shumin and Liu, Ke},
title = {CFMF: A Clustering‐Free Cell Marker Finder for Single‐Cell Transcriptomic Data},
journal = {IET Systems Biology},
year = {2026},
volume = {20},
number = {1},
doi = {10.1049/syb2.70065}
}
EUVIMED