State (Arc Institute perturbation model)
Predicts how cells will respond to a drug or gene knockout, trained on over 100 million perturbed cells.
State (2025) pairs a state-transition model trained on 100M+ perturbed cells (including Tahoe-100M) with a cell-embedding model from 167M human cells; the reference entry for Arc's Virtual Cell Challenge.
How it works
Transformer predicting expression shifts conditioned on perturbation and cell context.
- Largest perturbation training set
- Context generalisation
- Cell-line data; in vivo transfer unproven
Latest papers
topQuery for this technology: (TITLE:"State" OR ABSTRACT:"State" OR TITLE:"Arc Institute perturbation model" OR ABSTRACT:"Arc Institute perturbation model") AND (cancer OR tumor OR tumour OR oncology OR carcinoma OR lymphoma OR leukemia OR leukaemia OR myeloma OR sarcoma OR melanoma OR glioma). Results are unfiltered search hits about State (Arc Institute perturbation model), not a curated reading list.