Skip to main content
model-family passport · Review date not recorded

State / Stack

The first virtual-cell model predicting how cells react to drugs.

4/7Evidence fields documented
60-SECOND EVALUATION VIEW

What should a scientist know before using State / Stack?

SupportedEvidence supports the stated context with explicit boundaries
Best suited forRepresentation · Prediction
Evidence supportsPerturbation prediction datasets: Preprint / open evaluation
Evidence does not establishUniversal superiority, therapeutic success, clinical utility or regulatory acceptance.
Major limitationPerformance depends on the evaluation dataset and operating conditions.
Current registry recordStack2 recorded releases · Review date not recorded. A newer version is not assumed to be universally better.

What it is

State is Arc Institute's first virtual-cell model: given a starting transcriptome and a perturbation (drug, gene edit or cytokine), it predicts how gene expression will shift. Trained on ~170M observational and 100M+ perturbational cells across 70 contexts, it pairs a State Embedding module with a State Transition transformer. Its 2026 successor, Stack, learns cell biology in-context to generalize to unseen conditions.

Evidence trail

BioAtlas keeps the path from source to decision visible. A connection records provenance; it does not imply that evidence is sufficient for every context.

Sources4 connectedPrimary resources and normalized claims
Claims1 normalizedCell perturbation prediction
EntityState / Stackmodel-family · Stack
ReviewReview date not recordedReview date not claimed
ConclusionContext requiredAdd to an evaluation before operational use

Model passport

Entity typemodel-family
OrganizationArc Institute
Model family introduced2026
AccessOpen source
Commercial useRestricted / verify terms
DeploymentSelf-hosted
ComputeGPU recommended
Domainscells
Biology → representation → computation → evidence

How State / Stack represents biology

model-familycells

Category is navigation. These fields describe the model-specific computational transformation and deliberately override broad category defaults.

1 · Biological inputs
Starting cell stateDrug, gene-edit or cytokine perturbation
2 · Input representation
Gene-expression / cell-state representationPerturbation representation
3 · Internal representation
Cell-state embeddingTransition representation
4 · Architecture
State embedding + transition modelIn-context successor lineage (Stack)
5 · Learning objective
Perturbation-response modelling
6 · Output representation
Gene-expression state

Biological scale

cellgene

Modalities & tasks

CellTranscriptomicsPerturbationRepresentationPredictionSimulation

Registry, claims and frontier intelligence

Versioned registry

Stack

2 version records · latest curated year 2026. Model-family identity remains separate from capability and access changes.

Explore version lineage →
State
Stack

Connected research frontiers

These records describe active research directions, not guaranteed capabilities of this model. Evidence stages and unresolved questions are preserved separately.

Virtual cells

Virtual cells that predict perturbation response

Arc Institute · Virtual Cell research community · 2026-04-30
Recent preprint

Can models forecast how cell populations respond to unseen drugs, gene edits, cytokines and environmental changes across biological contexts?

Evidence boundary and unresolved questions

Recent strict evaluations show marked performance drops under unseen contexts and metric-dependent rankings; simple baselines remain competitive on some global trends.

  • Can models recover perturbation-specific mechanisms rather than average expression shifts?
  • How should cell distributions, dose and time be represented?
  • Which metrics predict prospective experimental usefulness?
virtual cells · perturbation · single cell · OOD generalization · world modelsOpen frontier record →

Inputs and outputs

Inputs

Starting cell stateDrug, gene-edit or cytokine perturbation

Outputs

Predicted post-perturbation cell states

Scientific and technical profile

Scientific principles

Perturbation response predictionRepresentation learning of cell stateIn-context learning (Stack)

Technology

State Embedding (SE) moduleState Transition (ST) transformerSet-of-cells modeling
Ideas before algorithms

Scientific lineage

Explore all foundations

These are transparent concept matches—not claims that one scientist alone caused this model. Each connection is based on the model’s recorded domain, scientific principles, technical terms or an explicit lineage link.

Genomics & cell systems

Gene regulation and the operon model

François Jacob & Jacques Monod

Target biology, perturbation models, transcriptomic response prediction and virtual cells all require an explicit model of regulated gene programs.

Matched concepts: expression, transcript, perturbation
Genomics & cell systems

The epigenetic landscape and cell-fate trajectories

Conrad H. Waddington

Single-cell embeddings, trajectory inference, reprogramming and virtual-cell models often represent cell identity as movement through a learned state landscape.

Matched concepts: cell state, single-cell
Biologics & genome engineering

Programmable CRISPR–Cas genome editing

Jennifer A. Doudna & Emmanuelle Charpentier

CRISPR enables target validation, disease models, perturbation atlases, functional genomics and gene-editing therapeutics.

Matched concepts: gene edit, perturbation
Molecular recognition

Cooperative ligand binding

Archibald V. Hill

Dose–response curves, receptor occupancy, multisite binding and systems pharmacology still use Hill-type models.

Matched concepts: response
Computational intelligence

Information, entropy and communication

Claude E. Shannon

Sequence modelling, cross-entropy training, language models, mutual information and representation learning all use Shannon’s framework.

Matched concepts: representation

Evaluation evidence

Dataset or evaluationPerturbation prediction datasets
Task or metricCell-state prediction
Evidence statusPreprint / open evaluation
Open source ↗

Task-specific evidence only; not comparable as a universal leaderboard score.

Cell perturbation prediction

Perturbation prediction datasets

Stack · Perturbation-prediction datasets
developer-reported

Open evaluation of cell-state prediction under perturbation.

Claim caveats
  • Cell line, tissue, dose and timepoint shifts can dominate performance.
  • In-silico perturbations require experimental validation.

Known limitations

  • Performance depends on the evaluation dataset and operating conditions.
  • Task-specific benchmark results should not be compared across unlike domains.
  • Outputs require task-specific scientific and experimental validation.

Milestones

2026

Trained on 270M+ cells — largest single-cell perturbation model at release.

Evidence

Built on the Arc Virtual Cell Atlas.