Cell-DINO: Self-supervised image-based embeddings for cell fluorescent microscopy
PLOS Computational Biology · 29.12.2025 · 21 (12) · e1013828
DOI: 10.1371/journal.pcbi.1013828 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS Computational Biology, online erschienen am 29.12.2025. Verfasst von Théo Moutakanni und 12 weiteren Autor:innen. Beteiligte Einrichtung: Meta Platforms Inc., FAIR, Paris, France. Hauptthemen laut MEDLINE: Image Processing, Computer-Assisted, Single-Cell Analysis. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 88 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NHGRI NIH HHS, Muscular Dystrophy Association, National Institutes of Health.
Abstract (Originalsprache)
Accurately quantifying cellular morphology at scale could substantially empower existing single-cell approaches. However, measuring cell morphology remains an active field of research, which has inspired multiple computer vision algorithms over the years. Here, we show that DINOv2, a vision-transformer based, self-supervised algorithm, has a remarkable ability for learning rich representations of cellular morphology without manual annotations or any other type of supervision. We apply DINOv2 to cell phenotyping problems, and compare the performance of resulting models, called Cell-DINO models, on a wide variety of tasks across two publicly available imaging datasets of diverse specifications and biological focus. Compared to supervised and other self-supervised baselines, Cell-DINO models demonstrate improved performance, especially in low annotation regimes. For instance, to classify protein localization using only 1% of annotations on a challenging single-cell dataset, Cell-DINO performs 70% better than a supervised strategy, and 24% better than another self-supervised alternative. The results show that Cell-DINO can support the study of unknown biological variation, including single-cell heterogeneity and relationships between experimental conditions, making it an excellent tool for image-based biological discovery.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Meta Platforms Inc., FAIR, Paris, France.
- MICS, CentraleSupelec, Université Paris-Saclay, Gif-sur-Yvette, France.
- Broad Institute, Cambridge, Massachusetts, United States of America.
- Helmholtz Munich, Institute of Computational Biology, Neuherberg, Germany.
- Google LLC DeepMind, Paris, France.
- Columbia University Irving Medical Center, Neurology, New York, New York, United States of America.
ORCID-Profile: Zitong S Chen, Nikita Moshkov, Elouan Gardes, Juan C Caicedo
Förderung
- NHGRI NIH HHS
- Muscular Dystrophy Association
- National Institutes of Health
- Hungarian Ministry of Innovation
- National Science Foundation
Zitieren
APA
Moutakanni, T., Couprie, C., Yi, S., Doron, M., Chen, Z. S., Moshkov, N., Gardes, E., Caron, M., Touvron, H., Joulin, A., Bojanowski, P., Pernice, W. M., & Caicedo, J. C. (2025). Cell-DINO: Self-supervised image-based embeddings for cell fluorescent microscopy. PLOS Computational Biology, 21(12), e1013828. https://doi.org/10.1371/journal.pcbi.1013828
Vancouver
Moutakanni T, Couprie C, Yi S, Doron M, Chen ZS, Moshkov N, et al. Cell-DINO: Self-supervised image-based embeddings for cell fluorescent microscopy. PLOS Computational Biology. 2025;21(12):e1013828. doi:10.1371/journal.pcbi.1013828
BibTeX
@article{Moutakanni2025,
author = {Moutakanni, Théo and Couprie, Camille and Yi, Seungeun and Doron, Michael and Chen, Zitong S and Moshkov, Nikita and Gardes, Elouan and Caron, Mathilde and Touvron, Hugo and Joulin, Armand and Bojanowski, Piotr and Pernice, Wolfgang M and Caicedo, Juan C},
title = {Cell-DINO: Self-supervised image-based embeddings for cell fluorescent microscopy},
journal = {PLOS Computational Biology},
year = {2025},
volume = {21},
number = {12},
pages = {e1013828},
doi = {10.1371/journal.pcbi.1013828}
}
EUVIMED