Understanding cell fate decisions with temporal attention
5 affiliations
- 1Institute for Biomedical Informatics, Faculty of Medicine, University of Cologne and University Hospital Cologne, Cologne, Germany
- 2Faculty of Mathematics and Natural Sciences, University of Cologne, Cologne, Germany
- 3Cologne Excellence Cluster on Cellular Stress Responses in Aging-Associated Diseases (CECAD), University of Cologne, Cologne, Germany
- 4Center for Molecular Medicine Cologne (CMMC), University of Cologne and University Hospital Cologne, Cologne, Germany
- 5Granada Lab, Berlin, Germany
Abstract
Understanding non-genetic determinants of cell fate is critical for developing and improving cancer therapies, as genetically identical cells can exhibit divergent outcomes under the same treatment conditions. In this work, we present a deep learning approach for cell fate prediction from raw long-term live-cell recordings of cancer cell populations under chemotherapeutic treatment. Our Transformer model is trained to predict cell fate directly from raw image sequences, without relying on predefined morphological or molecular features. Beyond classification, we introduce a comprehensive explainability framework for interpreting the temporal and morphological cues guiding the model’s predictions. We demonstrate that prediction of cell outcomes is possible based on the video only, our model achieves balanced accuracy of 0.94 and an F1-score of 0.93. Attention and masking experiments further indicate that the signal predictive of the cell fate is not uniquely located in the final frames of a cell trajectory, as reliable predictions are possible up to 8 h before the event. Our analysis reveals distinct temporal distribution of predictive information in the mitotic and apoptotic sequences, as well as the role of cell morphology and p53 signaling in determining cell outcomes. Together, these findings demonstrate that attention-based temporal models enable accurate cell fate prediction while providing biologically interpretable insights into non-genetic determinants of cellular decision-making. Author summary: Cancer treatment often fails because not all cells respond in the same way. Even when exposed to the same drug, some cells die while others survive and continue to grow, which can lead to relapse. Understanding why genetically identical cells make different decisions is therefore an important challenge. In this study, we investigate whether it is possible to predict the fate of individual cells directly from time-lapse microscopy videos. We develop a computational approach that learns from raw image sequences without relying on manually defined features. This allows us to capture subtle changes in cell behavior over time. We find that cell fate can be predicted with high accuracy based on the video data alone, and that informative signals appear well before the final outcome is visible. By analyzing the model, we identify when during a cell’s lifetime predictive information emerges and which aspects of cell behavior are most relevant. Our results show that dynamic patterns in cell behavior contain early indicators of fate decisions, providing new opportunities to better understand variability in treatment response.
Related
- Running to Their Fates: Neural Crest and Placode Migratory Behavior and Cell Fate Decisions
- Interpretable machine learning coupled to gene regulatory networks uncovers subcircuits underlying cell fate decisions
- Post-transcriptional regulation of Profilin-2 by microRNAs and RNA-binding proteins forms a critical regulatory node for early embryonic cell fate decisions
- Spatial, temporal and Notch determination of terminal selector expression controls neuronal cell fate in the Drosophila optic lobe
- Early hematopoietic fate decisions are independent of bone marrow niche
BibTeX
@article{burger2026understanding,
title = {{Understanding cell fate decisions with temporal attention}},
author = {Florian Bürger and Martim Dias Gomes and Adrián E. Granada and Noémie Moreau and Katarzyna Bozek},
journal = {PLOS Computational Biology},
year = {2026},
month = oct,
doi = {10.1371/journal.pcbi.1014813},
url = {https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1014813},
note = {CC BY},
}
RIS
TY - JOUR
TI - Understanding cell fate decisions with temporal attention
AU - Florian Bürger
AU - Martim Dias Gomes
AU - Adrián E. Granada
AU - Noémie Moreau
AU - Katarzyna Bozek
PY - 2026
DA - 2026/10/01
JO - PLOS Computational Biology
DO - 10.1371/journal.pcbi.1014813
UR - https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1014813
AB - Understanding non-genetic determinants of cell fate is critical for developing and improving cancer therapies, as genetically identical cells can exhibit divergent outcomes under the same treatment conditions. In this work, we present a deep learning approach for cell fate prediction from raw long-term live-cell recordings of cancer cell populations under chemotherapeutic treatment. Our Transformer model is trained to predict cell fate directly from raw image sequences, without relying on predefined morphological or molecular features. Beyond classification, we introduce a comprehensive explainability framework for interpreting the temporal and morphological cues guiding the model’s predictions. We demonstrate that prediction of cell outcomes is possible based on the video only, our model achieves balanced accuracy of 0.94 and an F1-score of 0.93. Attention and masking experiments further indicate that the signal predictive of the cell fate is not uniquely located in the final frames of a cell trajectory, as reliable predictions are possible up to 8 h before the event. Our analysis reveals distinct temporal distribution of predictive information in the mitotic and apoptotic sequences, as well as the role of cell morphology and p53 signaling in determining cell outcomes. Together, these findings demonstrate that attention-based temporal models enable accurate cell fate prediction while providing biologically interpretable insights into non-genetic determinants of cellular decision-making. Author summary: Cancer treatment often fails because not all cells respond in the same way. Even when exposed to the same drug, some cells die while others survive and continue to grow, which can lead to relapse. Understanding why genetically identical cells make different decisions is therefore an important challenge. In this study, we investigate whether it is possible to predict the fate of individual cells directly from time-lapse microscopy videos. We develop a computational approach that learns from raw image sequences without relying on manually defined features. This allows us to capture subtle changes in cell behavior over time. We find that cell fate can be predicted with high accuracy based on the video data alone, and that informative signals appear well before the final outcome is visible. By analyzing the model, we identify when during a cell’s lifetime predictive information emerges and which aspects of cell behavior are most relevant. Our results show that dynamic patterns in cell behavior contain early indicators of fate decisions, providing new opportunities to better understand variability in treatment response.
SN - 1553-7358
SN - 1553-734X
ER -