EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

scURL: uncertainty-sharpened representation learning for single-cell multi-omics clustering

Hongwei Liu iD, Donghui Yu, Wei Li, Tong Zhang, Ching-Yu Cheng, Huazhu Fu iD, Hong Liang, Shan Cong iD, Xiaohui Yao iD

Bioinformatics · 02.10.2026 · 42 (10)

DOI: 10.1093/bioinformatics/btag723 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioinformatics, online erschienen am 29.09.2026. Verfasst von Hongwei Liu und 8 weiteren Autor:innen. Beteiligte Einrichtung: College of Intelligent Systems Science and Engineering, Harbin Engineering University, Harbin, 150001, China. Hauptthemen laut MEDLINE: Multiomics, Single-Cell Analysis, Software, Computational Biology. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 29 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Shandong Provincial Natural Science Foundation, Hainan Provincial Natural Science Foundation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Motivation Single-cell multi-omics clustering requires integrating complementary omics layers while accounting for their unequal reliability across individual cells. However, cell-wise reliability can vary substantially within each omics layer, causing different omics to provide reliable signals for some cells but ambiguous or noisy signals for others. This challenge is further complicated by omics-level heterogeneity, where differences in sparsity, signal distribution, and biological resolution may introduce cross-omics conflicts and bias the fused representation. Results We present scURL, an uncertainty-sharpened representation learning framework for single-cell multi-omics clustering. scURL introduces representation uncertainty (RU) and establishes its connection to clustering generalization risk (GR), enabling the derivation of cell-wise omics weights for uncertainty-aware fusion (UAF). To support stable uncertainty estimation, scURL further introduces a multi-granular calibration module (MCM) that calibrates omics-specific representations from both cluster-level semantic and local-level structural perspectives before fusion. Experiments on ten datasets demonstrate that scURL outperforms existing clustering methods, remains robust under dropout noise, and identifies biologically meaningful cell populations, as supported by marker gene analysis. Availability and Implementation Source code is available at https://github.com/Yaolab-fantastic/scURL.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

MultiomicsSingle-Cell AnalysisSoftwareComputational BiologyClustering AlgorithmsRepresentation Machine LearningUncertaintyCluster AnalysisHumansAlgorithms

Autor:innen und Institutionen

  • College of Intelligent Systems Science and Engineering, Harbin Engineering University, Harbin, 150001, China.
  • Qingdao Innovation and Development Center of Harbin Engineering University, Qingdao, 266000, China.
  • Sanya Nanhai Innovation and Development Base of Harbin Engineering University, Sanya, 572024, China.
  • Stem Cell and Regenerative Biology, Genome Institute of Singapore, A*STAR, Singapore, 138672, Singapore.
  • Department of Ophthalmology, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, 119228, Singapore.
  • Institute of Advanced Intelligence and Computing (IAIC), A*STAR, Singapore, 138632, Singapore.

ORCID-Profile: Hongwei Liu, Huazhu Fu, Shan Cong, Xiaohui Yao

Förderung

  • Shandong Provincial Natural Science Foundation
  • Hainan Provincial Natural Science Foundation

Zitieren

APA

Liu, H., Yu, D., Li, W., Zhang, T., Cheng, C. Y., Fu, H., Liang, H., Cong, S., & Yao, X. (2026). scURL: uncertainty-sharpened representation learning for single-cell multi-omics clustering. Bioinformatics, 42(10). https://doi.org/10.1093/bioinformatics/btag723

Vancouver

Liu H, Yu D, Li W, Zhang T, Cheng CY, Fu H, et al. scURL: uncertainty-sharpened representation learning for single-cell multi-omics clustering. Bioinformatics. 2026;42(10). doi:10.1093/bioinformatics/btag723

BibTeX
@article{Liu2026,
  author = {Liu, Hongwei and Yu, Donghui and Li, Wei and Zhang, Tong and Cheng, Ching-Yu and Fu, Huazhu and Liang, Hong and Cong, Shan and Yao, Xiaohui},
  title = {scURL: uncertainty-sharpened representation learning for single-cell multi-omics clustering},
  journal = {Bioinformatics},
  year = {2026},
  volume = {42},
  number = {10},
  doi = {10.1093/bioinformatics/btag723}
}
Kostenlos starten →