EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

scTimeBench: a streamlined benchmarking platform for single-cell time-series analysis

Adrien Osakwe iD, Eric H Huang, Yue Li iD

Bioinformatics · 2026

DOI: 10.1093/bioinformatics/btag414 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioinformatics, 2026. Verfasst von Adrien Osakwe und 2 weiteren Autor:innen. Beteiligte Einrichtung: Quantitative Life Sciences Program, McGill University, 550 Rue Sherbrooke O., Montreal, Quebec, H3A 1E3, Canada. Hauptthemen laut MEDLINE: Single-Cell Analysis, Software, Computational Biology. Die Arbeit nennt 29 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Canada Research Chair, Natural Sciences and Engineering Research Council of Canada, CIHR.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Summary Temporal modelling of single-cell gene expression is essential for capturing dynamic cellular processes, yet a systematic framework for evaluating time-aware trajectory inference methods has not yet been established. Here, we present a modular and scalable benchmark designed to assess methods across three critical tasks: forecast accuracy (temporal cell alignment) for projecting cells to unseen time points, embedding coherence between original and projected data, and cell-type lineage fidelity. We evaluated ten state-of-the-art methods, which are broadly categorized into 8 forecasting-based and 2 optimal transport (OT)-based methods across eight diverse datasets spanning four species. Our results show that while several methods achieve high forecast accuracy, they often fail to preserve biological signals, both in their latent spaces and in cell lineage reconstruction. Notably, most methods confer low lineage fidelity and often underperform compared to a correlation baseline. We further demonstrate that integrating pseudotime can effectively denoise trajectories by aligning the data snapshots with the intrinsic biological clock in each cell. Finally, to streamline benchmarking for temporal single-cell analysis, we built one of the first self-contained Python packages for the research community. Availability & Implementation https://github.com/li-lab-mcgill/scTimeBench.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Single-Cell AnalysisSoftwareComputational BiologyBenchmarkingSingle-Cell Gene Expression AnalysisAnimalsHumansAlgorithms

Autor:innen und Institutionen

  • Quantitative Life Sciences Program, McGill University, 550 Rue Sherbrooke O., Montreal, Quebec, H3A 1E3, Canada.
  • School of Computer Science, McGill University, 3480 Rue University, Montreal, Quebec H3A 2A7, Canada.

ORCID-Profile: Adrien Osakwe, Yue Li

Förderung

  • Canada Research Chair
  • Natural Sciences and Engineering Research Council of Canada
  • CIHR
  • CIHR Canada Graduate Research Scholarship-Doctoral
  • NSERC Canada Graduate Research Scholarship-Master's program
  • Canadian Institutes of Health Research
  • CIHR Canada Graduate Research Scholarship—Doctoral
  • NSERC Canada Graduate Research Scholarship—Master’s program

Zitieren

APA

Osakwe, A., Huang, E. H., & Li, Y. (2026). scTimeBench: a streamlined benchmarking platform for single-cell time-series analysis. Bioinformatics. https://doi.org/10.1093/bioinformatics/btag414

Vancouver

Osakwe A, Huang EH, Li Y. scTimeBench: a streamlined benchmarking platform for single-cell time-series analysis. Bioinformatics. 2026. doi:10.1093/bioinformatics/btag414

BibTeX
@article{Osakwe2026,
  author = {Osakwe, Adrien and Huang, Eric H and Li, Yue},
  title = {scTimeBench: a streamlined benchmarking platform for single-cell time-series analysis},
  journal = {Bioinformatics},
  year = {2026},
  doi = {10.1093/bioinformatics/btag414}
}
Kostenlos starten →