biobalm is a Python library for exploring the attractor landscape of large-scale Boolean networks with hundreds or thousands of variables. It combines symbolic (BDD) and automated (ASP) reasoning to efficiently construct a succession diagram of a Boolean network: an inclusion-based acyclic graph of the network's trap spaces. biobalm can then use this succession diagram to accelerate attractor search and infer control strategies for target trap spaces.
biobalmis accompanied by an analysis artefact that benchmarks it againstpystablemotifs,mts-nfvsandaeon.py. The artefact also compares the succession diagrams generated by published biological Boolean networks to random network ensembles and finds significant differences in structure. The artefact is available at Zenodo and Github.
biobalm is on PyPI:
pip install biobalm
The base installation should enable all core functionalities (generate succession diagrams and control strategies, find attractor seeds states and attractor sets). Optionally, you can also use pint during attractor identification as a static analysis step:
- Native binaries of
pintcan be obtained here.
You can also install the latest version of biobalm directly from github:
pip install git+https://github.com/jcrozum/biobalm.git@main
The manuscript introducing and benchmarking biobalm is freely available in Bioinformatics.
Please cite the following if you use biobalm in your publication:
@article{trinhBiobalmMappingAttractorLandscape2025,
title = {Mapping the attractor landscape of {{Boolean}} networks with biobalm},
author = {Trinh, Van-Giang and Park, Kyu Hyong and Pastva, Samuel and Rozum, Jordan C},
year = {2025},
month = may,
journal = {Bioinformatics},
volume = {41},
number = {5},
pages = {btaf280},
issn = {1367-4811},
doi = {10.1093/bioinformatics/btaf280}
}
To learn more about how biobalm functions, API documentation is available online. You can also explore the analysis artefact mentioned above for more in-depth examples of biobalm usage.
Usage examples are also available in the example directory. First, a simple usage example is provided in Jupyter notebook example/tutorial.ipynb. You can also run python3 example/attractors.py PATH_TO_NETWORK to compute all attractor seeds of a Boolean network. This scripts demonstrates basic configuration options for the attractor detection process (limits the succession diagram size and shows how to deal with succession diagrams that exceed this node limit).