Building virtual-cell dynamics
that evolve with biological measurements

Virtual cells must do more than represent observed states — they must predict what happens next. CellCompass determines when existing dynamics are sufficient and when evolving measurements require a new model to be designed, tested, and retained.

  • 100DynBench scenarios
  • 4measurement constraints
  • 4real single-cell studies
  • ∞long-horizon research
In motion

From biological measurements to designed dynamics.

A hand-drawn sketch of the idea — a cell on the bench, an atlas streaming into the machine, and the design cycle that draws new dynamics as new questions arrive.

a moving sketch · sound on for the score

Grounded in benchmark and real-data evidence

DynBench100 controlled tasks HCT116 pulse-labeled drug response LARRY clone-informed hematopoiesis CITE-seq RNA-to-protein dynamics CellCompass scientific agent + dynamical engine

The premise

As measurements evolve, the model must evolve with them.

New measurements reveal biology that existing dynamics cannot express. CellCompass turns those measurements into new models—and retains them for the next study.

Explore the system overview
Figure 1 system overview showing biological questions, evolving measurements, the CellCompass model-design loop, and reusable dynamics families.
Figure 1 · system overview Experimental evidence reshapes the dynamics base.
01

Prediction beyond observed states

Virtual cells must forecast cellular change under biological context and intervention.

02

The predictive task is not fixed

Each new experiment can expose a relation that existing dynamics cannot express.

03

Model design must accumulate

Scientific capacity should grow across studies instead of restarting from one-off solutions.

The modeling principle

New measurements change what a faithful cellular future must preserve.

New measurements do more than enrich a cellular representation. CellCompass translates the biological relation they reveal into an explicit constraint on future dynamics — movement, population change, lineage divergence, perturbation response, or future molecular phenotype.
01
Time course

Velocity through state space

Sampled marginals constrain how populations move between observed time points.

transport
02
Clone record

Branching future fate

Clone barcodes link early progenitors to their descendants and constrain future fate probabilities.

fate mass
03
Growth / death

Changing population mass

Changes in observed abundance motivate models that allow population mass to vary over time.

unbalanced
04
Multi-omics

Coupled regulatory state

RNA and protein readouts require trajectories that carry molecular programs across modalities.

RNA → protein
The dynamical engine

Three model components connect measurements to dynamics.

Transport, bridge and flow-matching models share a modular structure. CellCompass changes the components needed to represent a new experimental constraint while reusing the remaining training and simulation procedures.

Coupling

Which cells can be connected across observation times? RNA state, metabolic directions, clone barcodes or paired protein measurements can inform these connections.

Conditional path

How do connected cells change between observations? Paths can be deterministic or stochastic and can incorporate measured directions of molecular change.

Transported quantities and readouts

What changes along the path? Models can carry effective population mass, represent fate branches or predict surface-protein states along RNA trajectories.

The engine includes balanced and unbalanced formulations, with flow-matching methods for learning velocity, growth and score-based diffusion fields. Spatial interactions and history-dependent processes are directions for future extension.

Measurement-grounded evaluation

Six tasks test recovery of cellular dynamics.

DynBench100 uses controlled ground truth to test dynamical reasoning across velocity, growth-driver recovery, held-out distributions, fate probabilities, perturbation response, and gene-regulatory recovery. CellCompass achieved the highest aggregate performance among the tested coding and biomedical agents.

100scenarios across easy, medium, hard, and adversarial settings
0.712highest aggregate score among tested agents
6dynamical objectives evaluated per agent
20 × 3scenarios × runs for the repeated-trial comparison

CellCompass scored 0.712 with GPT-5.5 high and 0.667 with MiMo-v2.5-pro. In the repeated-trial comparison, GPT-5.5 high reached a mean score of 0.672.

Contribution of the dynamical engine

Across all 40 medium-complexity scenarios, with two runs per condition, the mean six-task score was 0.736 with the engine and 0.672 without it. The complete system scored higher in 35 of 40 scenarios, with the largest gains in fate and regulatory recovery.

DynBench100 evaluates six dynamical tasks, agent performance, repeated-run stability and the contribution of the dynamical engine.
CellCompass achieved the highest aggregate score across six dynamical tasks. Panel g compares the complete system with a version without the dynamical engine. View full-resolution figure ↗
Cumulative model building

Construct. Test. Retain. Repeat.

When an experiment reveals a relation that existing models cannot express, CellCompass proposes and implements a new formulation, tests it against the measurements and retains the validated method.

01 · 07 Motivation

A real biological gap

Experimental measurements reveal a biological relationship that existing dynamics do not represent.

    CellCompass in action

    Watch new dynamics take shape.

    Three experiments. Three new models. Follow the design choices and the measured results.

    ·
    Recorded optimization

    TrialRetained model
    Model design · conceptual illustration

    Terrain and positions illustrate design choices, not a measured loss surface.

    What this revealed

    Explore the biology and paper figures

    Open full-resolution figure ↗
    Model reuse, not just invention

    Sometimes the right model already exists.

    Across four cancer cell lines and three stimuli, CellCompass reused velocity–growth dynamics to reconstruct context-dependent EMT routes.

    • 12 trajectoriesseparate velocity–growth dynamics for every cell-line × stimulus pair — 34,994 cells, no new algorithm required.
    • Context-gated routesthe same stimulus yields mesenchymal conversion, partial/hybrid EMT, or epithelial retention depending on the initial cell-line state.
    • Route-specific programsstimulus-proximal signaling, cell-state identity and downstream remodeling distinguish the inferred routes.
    Generated endpoint route atlas across twelve cell-line and stimulus conditions, showing mesenchymal conversion, partial or hybrid EMT, epithelial-retaining and mixed routes.
    Context-specific EMT trajectories, gene programs and effective abundance across twelve cell-line × stimulus conditions.
    Using CellCompass

    Start with your data and a biological question.

    Launch the web interface, configure a supported model provider and describe the question you want to investigate. CellCompass prepares the data, proposes a dynamical model and evaluates the generated trajectories.

    Algorithm-design settings

    For algorithm-design studies, enable Planner stop hook under Advanced settings → Review and stop hook before initializing the agent. Select LLM prompt evaluator to continue unfinished model development.

    Codex sign-in · Algorithm-design settings

    start CellCompass
    cellcompass --llm-model gpt-5.5 --llm-thinking-level high \
      web --host 127.0.0.1 --port 8000 --device cpu

    Open http://localhost:8000, configure the provider and click Initialize Agent.

    My time-course scRNA-seq data are in /path/to/data.h5ad. How does this cell population change over time? Check the data and sampling times, then propose a dynamical model and explain how you would evaluate it.
    The research behind it

    Built on a line of published theory.

    npj Syst. Biol. Appl. · 2026 Deciphering cell-fate trajectories using spatiotemporal single-cell transcriptomic data
    ICLR · 2025 · Oral Learning stochastic dynamics from snapshots through regularized unbalanced optimal transport
    NeurIPS · 2025 Modeling cell dynamics and interactions with the Unbalanced Mean-Field Schrödinger Bridge
    NeurIPS · 2025 Joint Velocity-Growth Flow Matching for single-cell dynamics modeling
    NeurIPS · 2025 Variational Regularized Unbalanced Optimal Transport: single network, least action
    Nature Methods stVCR: spatiotemporal dynamics of single cells
    Core team

    The core team behind CellCompass.

    The initial CellCompass team brings together single-cell dynamics, mathematical modeling, and AI-assisted model design.

    Peking University
    University of California, Irvine
    Qing Nie

    Qing Nie

    University of California, Irvine

    Bring a new measurement.
    Let the dynamics evolve.

    Reuse an existing model when it is adequate, or begin a measurement-grounded design campaign.

    Requirements: Conda, Node.js 22.12 or later, and a supported provider’s API key or Codex sign-in. Small examples can run on CPU. An NVIDIA GPU is recommended for biological model training.

    install
    git clone https://github.com/zhenyiizhang/CellCompass-review.git
    cd CellCompass-review
    conda create -n cellcompass python=3.10 -y
    conda activate cellcompass
    python -m pip install -c requirements-runtime-constraints.txt -r requirements-agent.txt -e ./CytoBridge-main -e .
    npm --prefix web ci
    npm --prefix web run build