CellLand for Cell dynamics on energy Landscapes

CellLand benchmarks principal component-based energy landscape mapping on EMT-MET, Melanoma, and SERGIO-simulated datasets with prior regulatory network knowledge.

---

CellLand

• CellLand (Cell dynamics on energy Landscapes) is a comprehensive pipeline for benchmarking attractor detection in Boolean and diffusion-based models under in silico driver-gene perturbation.
• Data contains all input and output files. Because some files exceed 100 MB, they cannot be uploaded to this repository. All data analyzed in this work are available on figshare.
• Boolean network tutorial (Python): PCA_visualization.ipynb provides fully reproducible workflow.
• Diffusion model tutorial (MATLAB): plot_Landscape.m, run with matlab -r "plot_Landscape".
• Questions about CellLand could be directed to the corresponding author, Prof. Zhi-Ping Liu (email: zpliu@sdu.edu.cn).

---

Boolean model

Conda environment

Use Python 3.10 or 3.11.

conda create -n CellLand python=3.11 -y
conda activate CellLand
conda install -c conda-forge numpy pandas scipy matplotlib scikit-learn seaborn

---

It allows users load Code/bmodel/ via sys.path without installing.

import sys
from pathlib import Path

for p in (Path("Code/bmodel"), Path("bmodel")):
    if (p / "bmodel" / "base.py").is_file():
        sys.path.insert(0, str(p.resolve()))
        break

from bmodel.base import Bmodel

Upstream package: ComplexityBiosystems/bmodel.

---

Diffusion model

MATLAB workflow

Launch MATLAB and run the diffusion-model landscape example from:

cd ALLIES/Code/diffusion/Melanoma_Landscape/code/
matlab -r "plot_Landscape"

Or start MATLAB interactively and execute:

plot_Landscape

Upstream package: chunhelilab/Melanoma_Landscape.

---

Citation

Lingyu Li, Liangjie Sun, Shumin Li, Wai-Ki Ching*, and Zhi-Ping Liu*. "Cell dynamics on energy landscapes: Comparing attractor detection in Boolean-network and diffusion-based models under in silico driver gene perturbations." Submitted and revised to Quantitative Biology.