Comprehensively resolving the cell state landscape requires integrating single-cell omics data from diverse studies. We developed CONCORD, a contrastive learning framework that leverages principled mini-batch sampling to learn denoised, batch-integrated and high-resolution representations of cells, capturing intricate structures such as differentiation trajectories and cell-cycle loops across numerous biological contexts.
This is a preview of subscription content, access via your institution
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
$32.99 / 30 days
cancel any time
Subscribe to this journal
Receive 12 print issues and online access
$259.00 per year
only $21.58 per issue
Buy this article
- Purchase on SpringerLink
- Instant access to the full article PDF.
USD 39.95
Prices may be subject to local taxes which are calculated during checkout
References
Flores-Bautista, E. & Thomson, M. Unraveling cell differentiation mechanisms through topological exploration of single-cell developmental trajectories. Preprint at bioRxiv https://doi.org/10.1101/2023.07.28.551057 (2023). This preprint demonstrates that complex topological structures in gene expression space reflect meaningful biological processes.
Luecken, M. D. et al. Benchmarking atlas-level data integration in single-cell genomics. Nat. Methods 19, 41–50 (2022). This paper systematically evaluates single-cell data integration methods using standardized metrics.
Chen, T., Kornblith, S., Norouzi, M. & Hinton, G. A simple framework for contrastive learning of visual representations. In International Conference on Machine Learning 1597–1607 (2020). This paper introduces the SimCLR framework that underpins many modern contrastive learning methods.
Robinson, J., Chuang, C.-Y., Sra, S. & Jegelka, S. Contrastive learning with hard negative samples. In International Conference on Learning Representations (2021). This paper shows that modifying mini-batch composition with hard-negative sampling improves contrastive learning performance.
Large, C. R. et al. Lineage-resolved analysis of embryonic gene expression evolution in C. elegans and C. briggsae. Science 388, eadu8249 (2025). This paper presents a lineage-resolved cross-species atlas of embryogenesis, providing a ground-truth resource for benchmarking data integration methods.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This is a summary of: Zhu, Q. et al. Revealing a coherent cell-state landscape across single-cell datasets with CONCORD. Nat. Biotechnol. https://doi.org/10.1038/s41587-025-02950-z (2026).
Rights and permissions
About this article
Cite this article
Machine learning framework reveals a concordant cell-state landscape across single-cell datasets. Nat Biotechnol (2026). https://doi.org/10.1038/s41587-025-02978-1
Published:
Version of record:
DOI: https://doi.org/10.1038/s41587-025-02978-1
