AI for Bio · Model

STATE

Virtual cell model predicting transcriptional response to drug or genetic perturbation across cell lines

Single-cell & perturbationNonprofit instituteUnited StatesActive中文

Overview

NameSTATE · aliases: STATE model
Full nameState Transition + State Embedding
Ambiguous names (not auto-linked)STATE, State
FamilyNot a family entry. The developers say a next generation is in development but unreleased; no family entry
OrganizationArc Institute
Organization typeNonprofit institute
CountryUnited States
Model classSingle-cell perturbation prediction / virtual cell
DomainSingle-cell & perturbation
First release† partly verified · paper revision
Basis: Preprint (10.1101/2025.06.26.661135) and the developer's launch page
Note: The site also records a 2025-06-27 posting; the day-level difference is unresolved source
Other milestones
  • First public evidence: † · Preprint and launch page in the same window source
  • Preprint v1: † · Preprint v1 source
  • Journal online: † · Journal issue (registration is month-level; no online-first day found) source
  • Code release: † · Repository v0.11.x releases (2026-06 to 07) source
  • Official announcement: † · Developer launch page source
Latest update† partly verified · journal
Basis: Journal publication (issue is month-level)
Journal version published; the repository's v0.11.x releases (2026-06 to 07) are code releases. The developers say a next generation is in development but unreleased
Note: No online-first day was found; only month-level evidence source
The most recent sourced event, not a claim that nothing changed since.
StatusActive · Journal publication and code releases within 12 months
OpennessCode: Open
Weights: Non-commercial · non-commercial licence
Code is public and weights carry a non-commercial licence; the full licence text was not retrieved this round
Benchmarks reported1 counted (modified protocol 1)
Wet-lab validation1 recorded campaign
Unverified fields† first_release, latest_update, release_milestones.journal_online, benchmarks_reported[0], open_status.code_license
As of2026-11-16

What it does

Learns state transitions between pre- and post-perturbation cells to predict transcriptional response to drugs or genetic perturbation across cell lines, with a developer-built evaluation tool alongside. Unlike the institute's Stack, which does prompt-style prediction by in-context learning, this model trains on perturbation data for response prediction; it shares the same third-party perturbation corpus as Tahoe-x1.

Benchmarks reported

Every row is one pairing from the survey's relation file; a score belongs to the benchmark page's ranking, where it is shown with metric, eval set, source type and date. 1 of 1 pairings are counted.

BenchmarkRelationEval set / protocolSource typeNote and evidence
24 Virtual Cell Challenge and perturbation benchmarksDeveloper report, modified protocol—Self-reportedSelf-reported with cell-eval on Tahoe-100M; its VCC blind score is unpublished. · evidence

Other evaluations (outside the 44)

  • Developer-built evaluation tool (Own task set) Self-reported: The official page and the paper abstract frame the same result two ways (about 50% and over 30%) and claim it is the first model to beat linear baselines consistently
  • Linear-baseline findings in independent evaluations (Comparison against baselines only) Third-party: Several independent evaluations and a challenge wrap-up find simple baselines hard to beat, cutting against the developer's claim
  • scIB (Public benchmark not among the 44) Self-reported: Confirmed not a pairing: neither the paper nor the official page reports that benchmark

Developers report: learning state transitions substantially improves separation of perturbation effects and recovery of truly differential genes, and beats linear baselines consistently for the first time.

Independent evidence: No independent replication of this model was found; related independent evaluations and the perturbation-prediction challenge wrap-up conclude that simple baselines are hard to beat.

Training data

Group in Figure 7: Public data only.

ClassValueDetail
a. Public databasesYesAbout 167 million observational cells (including the institute's atlas and a public count collection), over 100 million perturbed cells across 70 cell lines, plus a public Perturb-seq dataset source
b. Private or licensed dataNot disclosed
c. Own wet-lab data used for trainingNoNone for the developer: the main perturbation corpus was produced by a data company and donated to the institute's atlas (about 100 million cells, 50 cancer cell lines, 1,100 perturbations) and another set came from a reagent company; the institute's own challenge data is used only for evaluation. To avoid leakage, training used 20 of the cell lines with 5 held out source
d. Wet lab used only for validationNo source

Wet-lab validation

Every rate carries its numerator and denominator. Claims that could not be verified are shown as unverified, without numbers; different paper versions are separate entries.

Perturbation data description (not an own experiment) (primary paper)
Targets — · Designs tested — · Hits — · Success — · Best affinity —

Denominator basis: The same third-party perturbation corpus; the perturbation count is unified to 1,100 this round (the press release's ~1,200 is not used)
This item exists only to correct the published perturbation count and is not wet-lab validation of the model
Executor: Data produced by a third-party company, not a wet-lab campaign by the developer

Self-reported · † · verified · source

Background

  • Organization: A California non-profit research institute founded in 2021 with over $650M committed at launch; core investigators hold multi-year renewable terms alongside university appointments. No citable official headcount, so none is given.
  • Funding: Non-profit and endowment-funded; 2024 budget figures differ between sources, so no single number is given.
  • Products and partnerships: The institute's virtual-cell initiative includes a cell atlas and an annual perturbation-prediction challenge (first launched 2025-06-26 with a $100k grand prize and three sequencing and compute sponsors; a 2026 edition is announced); sibling models are Stack and Evo 2; the main perturbation corpus was donated by a data company that also builds Tahoe-x1.
  • Related models: Stack, Tahoe-x1, Evo 2

Reading notes

  1. The same result is framed two ways on the official page and in the abstract; always say which framing is cited.
  2. Training and evaluation share one perturbation corpus; even with held-out cell lines this stays a within-distribution comparison.
  3. The main wet-lab data came from a third-party company and must not be attributed to the developer.
  4. No blind score for this model is published for the challenge, so its blind-test standing is unknown.

Caveats

  1. The only site pairing is self-reported under a modified protocol, and the blind-challenge part is unconfirmed.
  2. The journal online date has month-level evidence only.
  3. The perturbation count conflicts between sources; the lower framing is adopted this round.
  4. Weights carry a non-commercial licence.

Sources

  1. Journal paper 2026 (journal)
  2. Preprint (preprint)
  3. Developer launch page (official site)
  4. Code repository releases (repository)
  5. Challenge wrap-up (official site)
  6. Perturbation data donation announcement (official site)

Corrections

2 corrections
FieldWasNowBasisFound
training_data.c_own_wetlab_for_trainingPerturbation data on about 1,200 drugs1,100 perturbations (about 1,200 appears only in press releases)source2026-11-16
benchmarks_reportedChallenge results (treated as a board)Modified-protocol self-report; the challenge's blind-test scores are not published, marked unconfirmedsource2026-11-16

Part of the AI for Bio benchmarks survey · all 112 models · open in the explorer. Record as of 2026-11-16.