Repository to progressively translate the Wolfram Mathematica code from the original repository (Biomarker-Guided scRNA-Seq Cancer Attractor Analysis
) to Python. It also aims to include analyses from previous studies (article) and additional features.
- Construct and visualize gene regulatory networks (GRNs).
- Generate marker gene combinations and visualize dispersion.
- Create histograms and t-SNE plots for dimensionality reduction.
- Apply clustering algorithms (KMeans, DBSCAN) on reduced/markers dimensions and analyze clusters.
- Investigate biomarkers' potential for finding cancer attractors.
- Perform stochastic simulations to refine attractor identification.
- Identify potential clusters' multistability as sources of cancer recurrence.
- Others: Additional analyses and features from previous and upcoming studies.
-
Gene expression data:
- "inputs/datapoints_seurat_BT_ALL.xlsx"
This dataset contains gene expression data. The data can be obtained from the link.
-
Activation and inhibition adjacency matrices:
- "inputs/act_BT_ALL_seurat.xlsx"
- "inputs/sup_BT_ALL_seurat.xlsx"
These are the activation and inhibition adjacency matrices corresponding to the GRN dynamic model.
- Data: Input scRNA-seq data in a normalized count matrix format with adjacency matrices corresponding to the GRNs’ activation and inhibition interactions.
- Dimensionality reduction and clustering: Select the desired biomarker dimensions to proceed with the analysis.
- Python 3.x
- Jupyter Notebook
- NumPy
- Pandas
- Matplotlib
- SciPy
- Scikit-learn
- NetworkX
- SymPy
- TQDM
- Joblib
All necessary packages are pre-installed in Google Colab.
To view and run the notebook:
- Clone the repository to your local machine.
- Upload the repository to Google Colab.
- Run the cells as needed.
This project is licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License (CC BY-NC-SA 4.0).
This license allows others to remix, adapt, and build upon this work non-commercially, as long as they credit the original author and license their new creations under the same terms. For more details, please visit the license page.
Any contributions are welcome.