This repository contains the code for the analyses in the manuscript "Quantifying higher-order epistasis: beware the chimera". In these analyses, we compare the multiplicative and the "chimeric" epistasis measures for quantifying higher-order interactions between mutations, drug, or other perturbations.
Required packages:
- Numpy (>= 1.23.4)
- Matplotlib (>= 3.8.0)
- Pandas (>= 2.1.1)
- Scipy (>= 1.11.2)
- Seaborn (>= 0.12.2)
Code tested using standard Macbook Pro.
To run each of the analyses, see the following notebooks. Runtime should not take more than 20 minutes. See notebooks for expected output.
The notebook Kuzmin_data_analysis.ipynb in the directory yeast_screens contains all the analyses to reproduce our results on three-way interactions in yeast.
WARNING: refactoring of this code is in progress.
The notebook drug_interaction_reanalysis_additive.ipynb in the directory drug_data contains all the analyses to reproduce our results on higher-order interactions between antibiotics.
WARNING: refactoring of this code is in progress.
The notebook protein_epistasis.ipynb contains code to reproduce the analyses in the manuscript. We have included all data files in the folder protein_data, one file for each protein analyzed. We obtained these files from the original publications; we append the name of the protein to the beginning of the name of the original data file.
The manuscript is available on bioRxiv. If you use our code for your work, please cite our paper.
@article {Chitra2024.07.17.603976,
author = {Chitra, Uthsav and Arnold, Brian J and Raphael, Benjamin},
title = {Quantifying higher-order epistasis: beware the chimera},
elocation-id = {2024.07.17.603976},
year = {2024},
doi = {10.1101/2024.07.17.603976},
publisher = {Cold Spring Harbor Laboratory},
URL = {https://www.biorxiv.org/content/early/2024/07/19/2024.07.17.603976},
eprint = {https://www.biorxiv.org/content/early/2024/07/19/2024.07.17.603976.full.pdf},
journal = {bioRxiv}
}