# State (Arc Institute perturbation model)

Source: https://onco.cc/technologies/state-arc/  
OnCo record `state-arc` (Technology). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

Predicts how cells will respond to a drug or gene knockout, trained on over 100 million perturbed cells.

## Summary

State is the Arc Institute's perturbation model, a transformer that predicts how a cell's gene expression shifts in response to a drug or genetic knockout, conditioned on both the perturbation and the cell's context. The 2025 bioRxiv preprint pairs a state-transition model trained on more than 100M perturbed cells, including the Tahoe-100M dataset, with a cell-embedding model trained on 167M human cells, and it serves as the reference entry for Arc's Virtual Cell Challenge. It is aimed at researchers who want to prioritise which perturbations to test experimentally, including in cancer cell lines. The training data come from cell lines, so transfer to tissues and patients in vivo is unproven, and benchmark work in the field has shown perturbation prediction is hard. For a newcomer: State tries to predict what a drug does to a cell before anyone runs the experiment.

## Fields

- Kind: Technology
- Status: emerging
- Last checked: 2026-09-08
- Tags: foundation-model; virtual-cell
- Principle: Transformer predicting expression shifts conditioned on perturbation and cell context.
- Since: 2025
- Strengths: Largest perturbation training set; Context generalisation
- Limitations: Cell-line data; in vivo transfer unproven

## Sources

- State bioRxiv 2025: https://www.biorxiv.org/content/10.1101/2025.06.26.661135v1

## Connected records

- fronts: [AI & Computation](https://onco.cc/fronts/ai-computation/), [Drug Discovery Platforms](https://onco.cc/fronts/drug-discovery/)
- technologies: [CRISPR functional genomics](https://onco.cc/technologies/crispr-screens/), [Single-cell & spatial profiling](https://onco.cc/technologies/single-cell-spatial/)
- institutions: [Arc Institute](https://onco.cc/institutions/arc-institute/)
- collections: [Tahoe-100M](https://onco.cc/collections/tahoe-100m/)
- roadmaps: [AI in oncology roadmap: pattern readers → foundation models → agents in the workflow](https://onco.cc/roadmaps/ai-oncology-roadmap/), [Drug discovery roadmap: screening in mice → maps of dependency → designing in silico](https://onco.cc/roadmaps/drug-discovery-roadmap/), [Virtual cell roadmap: from bulk omics to a predictive model of a cancer cell](https://onco.cc/roadmaps/virtual-cell/)

---
JSON: https://onco.cc/api/v1/entities/state-arc.json