ABMACT: Agent Based Model for Adaptive Cancer Therapy

This repository contains the code and data for the ABMACT model, an agent-based model for simulating adaptive cancer therapy. The model is implemented in Python and uses various libraries for numerical computations and data visualization.

Our paper is published at: https://www.nature.com/articles/s42003-026-09653-4

The primary framework for ABM simulation is Mesa. Detailed documentation of Mesa can be found at https://mesa.readthedocs.io/stable/getting_started.html

Mesa: ter Hoeven, E., Kwakkel, J., Hess, V., Pike, T., Wang, B., rht, & Kazil, J. (2025). Mesa 3: Agent-based modeling with Python in 2025. Journal of Open Source Software, 10(107), 7668. https://doi.org/10.21105/joss.07668

Setup

1. Install required packages by running bash setup.sh
2. Download data from ABMACT-data from Zenodo at: in to data/ folder. This includes data for generating figures for lymphoma and glioblastoma mouse model, benchmarking with ODE, and sensitivity analysis.

Run Simulations

1. conda activate ABMACT
2. Modify parameter file in params/ folder as needed.
3. Modify arguments in run.sh as needed (e.g., runname, paramfile, etc.).
4. bash run.sh
5. Output will be saved in the output/{runname} folder.

More advanced options:

• Modify step length, function parameters, and other simulation settings in cell agent and TME object in model/.

An Experimental GPT App for building your cell agents

Check out the Cell Therapy Agent Basied Model Wizard