kyuhyongpark / cubewalkers-analysis

Model analysis using cubewalkers
1 stars 0 forks source link

cubewalkers-analysis

This repository contains all analysis result published on

Park KH, Costa FX, Rocha LM, Albert R, Rozum JC. Models of Cell Processes are Far from the Edge of Chaos. PRX Life 1, 023009 (2023)

which can be accessed here: https://doi.org/10.1103/PRXLife.1.023009.

Documentation

models

Here we have the original models from the Cell Collective (https://cellcollective.org), and the models we corrected. The corrections are made to better resemble the model in the original publications. For example, some models are only valid in a particular cellular context. See the list of alterations at corrected_models_list.csv.

analysis

Files in this directory cacalculates various dynamic parameters of Boolean models, such as the Derrida coefficient or the converged average node values. Requires cubewalkers, cupy and GPU supporting CUDA.

notebooks

We have notebooks that generate figures and tables in the paper, either by processing the data in the analysis folder or by running simple analyses directly.
perturbation_output_plots Figures 1, 4-6, 14-19
node_average_values_differences_plots Figure 3
simple_benchmark_plots Figures 7, 8
convergence_of_averages Figure 9
analysis_of_specific_models Some results are used to generate Figures 10 and 11
source_nodes_plots Figure 12
correction_comparison_plots Figure 13
models_description_tables Table II-XIV

data

All data (except data generated during the analysis phase), figures, and tables generated are stored here.

Requirements

cubewalkers (v1.3.7+) https://github.com/jcrozum/cubewalkers

cupy (v12.2.0+) https://cupy.dev/
Note: cupy requires cuda11x or other version depending on installed CUDA version

pyboolnet (v3.0.9+) https://github.com/hklarner/pyboolnet
Note 1: pyboolnet requires pyeda, which can be difficult to install in Windows; it is recommended to obtain a pyeda Windows wheel from https://www.lfd.uci.edu/~gohlke/pythonlibs/#pyeda
Note 2: pyboolnet also requires clasp and gringo to be installed separately on Linux systems

pystablemotifs (v3.0.4+) https://github.com/jcrozum/pystablemotifs

cana (v0.2.0+) https://github.com/CASCI-lab/CANA

pandas (v2.0.1+) https://pandas.pydata.org/

numy (v1.19.2+) https://numpy.org/

scipy (v1.10.1+) https://scipy.org/

matplotlib (v3.2.1+) https://matplotlib.org/

jinja2 (v3.1.2+) https://jinja.palletsprojects.com/en/2.10.x/

setuptools (v65.5.0+) https://github.com/pypa/setuptools