Latent Dynamics · 2026-11-05
CellCanon: An Agentic Cellular World Model for Prediction, Explanation, and Observation-Driven Updating
Zhaokang Liang, Shuyang Zhuang, Xiaoran Jiao, Hao Chen
AI-assisted summary based on the paper. Consult the linked source for methods and limitations. Benchmark numbers appear after editor verification.
TL;DR
CellCanon is an agentic cellular world model that unifies single-cell expression prediction, evidence-grounded explanation, and observation-driven updating. It reuses condition-level biological evidence across gene-specific numerical readouts. On 339 held-out SciPlex3 conditions, the authors report improved differential-expression AUPRC and Macro-F1 over the best baseline, and reduced language-model token use for Predict/Explain. The Explain module is source-grounded, and Update revises predictions from newly measured responses without changing model weights.
Why it matters
The authors target biological discovery beyond observed conditions by linking predictions to source-linked evidence and acquired measurements in a shared world state. This supports inspectable, revisable biological judgments: Explain provides supported/contradicted/insufficient evidence with proofs, and Update transfers measured responses to revise predictions for the same drug in unmeasured cell contexts without updating predictor weights.
Method
- Shared world state linking evidence interpretation, prediction readouts, explanations (with claim dependencies), and an observation record.
- Predict: language-selected, source-linked evidence is reused across genes in a condition; continuous and DE/DIR readouts guide single-cell generation from matched DMSO controls.
- Explain/Update: Explain verifies source support with supported/contradicted/insufficient judgments; Update revises predictions for the same drug in unmeasured contexts using one measured source context per drug and fixed predictor weights.
Abstract (from arXiv)
Biological discovery requires using molecular knowledge and new measurements to infer responses beyond the conditions already observed. We introduce CellCanon, the first agentic cellular world model to unify single-cell expression prediction, evidence-grounded explanation, and observation-driven updating. Its predictive core reuses condition-level biological evidence across gene-specific numerical readouts to guide single-cell expression generation. A shared world state links predictions, acquired measurements, and source-linked claims, supporting explanation and revision. On 339 held-out SciPlex3 conditions across three cell lines, CellCanon improves differential-expression AUPRC and Macro-F1 over the best baseline for each metric by 63.3% and 17.3%, respectively, and direction Macro-F1 by 7.2%. Relative to gene-wise reasoning on matched drug-cell cohorts, measured language-model token use falls by 99.77% for Predict and 98.41% for Explain. On the source-grounded benchmark, Explain correctly answers all answerable questions with supporting proofs and identifies all insufficient-evidence cases. In simulated experimental feedback, Update uses one measured source context per drug to exceed the best fine-tuned baselines in differential-expression AUPRC and Macro-F1 without changing model weights. Together, these capabilities support a cumulative account of cellular responses, with predictions and biological judgments that can be revised as evidence accumulates.
Related papers
- VCLMU: Mechanism-Centric Virtual Cell World Modeling for Perturbation Response
- The Delta Gap: Cellular World Models for Perturbation Response Prediction Are Far Harder Than Global Correlation Implies
- NeuronDiscover: Agent-in-Twin for Mechanistic Discovery in Neuronal Microenvironments with World Action Models
- P2P: Cross-View Population Denoising for Unpaired Single-Cell Perturbation Response Prediction
- KoopCell: Koopman-Based Generative Model for Learning Single-Cell Dynamics from Distribution Snapshots
Publication record
Bibliographic evidence describes where and when a record appeared. It does not validate scientific claims.
First public release: 2026-11-04 · Indexed by Pub-AI: 2026-11-05 UTC
CellCanon: An Agentic Cellular World Model for Prediction, Explanation, and Observation-Driven Updating
Preprint posted: 2026-11-04 · Original record ↗
- arxiv
- 2611.05890
Explicit source versions (1)
- v1 · 2026-11-04
- Preprint available
Source date: 2026-11-04 · First detected: 2026-11-05 UTC · Historical baseline
Source evidence (1)
- Official evidence ↗ · atom:published (official metadata)
Abstract · arxiv
Biological discovery requires using molecular knowledge and new measurements to infer responses beyond the conditions already observed. We introduce CellCanon, the first agentic cellular world model to unify single-cell expression prediction, evidence-grounded explanation, and observation-driven updating. Its predictive core reuses condition-level biological evidence across gene-specific numerical readouts to guide single-cell expression generation. A shared world state links predictions, acquired measurements, and source-linked claims, supporting explanation and revision. On 339 held-out SciPlex3 conditions across three cell lines, CellCanon improves differential-expression AUPRC and Macro-F1 over the best baseline for each metric by 63.3% and 17.3%, respectively, and direction Macro-F1 by 7.2%. Relative to gene-wise reasoning on matched drug-cell cohorts, measured language-model token use falls by 99.77% for Predict and 98.41% for Explain. On the source-grounded benchmark, Explain correctly answers all answerable questions with supporting proofs and identifies all insufficient-evidence cases. In simulated experimental feedback, Update uses one measured source context per drug to exceed the best fine-tuned baselines in differential-expression AUPRC and Macro-F1 without changing model weights. Together, these capabilities support a cumulative account of cellular responses, with predictions and biological judgments that can be revised as evidence accumulates.
