Arc Institute · State ↗
Cell representation and perturbation prediction are distinct tasks; the model page links to papers, code and weights.
AI VIRTUAL CELL · REFERENCES
Grounded in public research. Independently written.
The following public websites informed our understanding of research directions and industry communication. Accessed October 31, 2026.
Tahoe · Research & data ↗
Perturbation data, models and research tools connect biological questions with experimental evidence.
CellVoyant · FateView ↗
Label-free live-cell image analysis for counting, morphology and selected state readouts.
Rivercell · Virtual cell ↗
Describes a developing approach combining perturbations, multimodal longitudinal measurements and modeling.
Nature Methods · Perturbation prediction and simple baselines ↗
In the tested datasets and tasks, evaluated deep learning models did not outperform simple baselines; this does not establish failure of all models on all tasks.
Nature Methods · Generalizable perturbation prediction ↗
Evaluation across datasets, perturbations and cellular contexts supports task-specific model selection.
These sources support the research discussion. Their products, results and collaborations are their own; inclusion does not imply a relationship with AI Virtual Cell. Reviewed October 31, 2026.