Biodivine library for control of parametrised (partially specified) Boolean networks

A library to solve one-step, temporary and permanent source-target control of parametrised (partially-specified) Boolean networks.

The directory structure:

.
├── Cargo.toml
├── Cargo.lock
├── README.md
├── auxiliary_scripts/    # Scripts to do & process experiments
├── models/               # Base experimental models
├── results/              # Raw measured results from experiments
└── src/                  # Library source code
    ├── aeon/             # Simplified algorithms taken over from aeon-server
    ├── bin/              # Alternative console entry-points
    ├── control/          # Source-target control algorithms
    ├── perturbation/     # Perturbed state-transition graph implementation
    ├── phenotype_control # Phenotype control algorithms
    ├── lib.rs            # Library declaration
    └── main.rs           # Main console entry point

Auxiliary scripts

• analyse_results.py - A script showing quick statistics about the obtained experiment results
• networks_sampler.py - A script generating partially-specified samples of witness models
• plot_results.ipynb - A Jupyter notebook for visualization of the experiment results
• run_groups.py - A script for obtaining the experiment results, running the methods from library on the generated methods. Allows timeout specification.
• run_phenotype.py - A script to run phenotype-control experiments on parametrised Boolean-network models; supports timeouts and exports results for later analysis.

Models

Base models for testing the library. Contains witness models from CellCollective platform and some parametrised version of the models.

results

The raw unprocessed outputs of experiments for both performance comparison and robustness metric.

src

Source code of the library. Consists of following rust modules:

aeon module

Operations to perform base state-transition graph manipulations.

control module

Implementations of control algorithm on the perturbable graph.

perturbation module

Data structure representing state transition graph of Boolean network which is viable for perturbations.

phenotype_control module

Implementations of phenotype control algorithm on the perturbable graph.

To run the basic experiments, execute cargo run --release