Welcome to the GitHub repository for the following publication: An extension of the Walsh-Hadamard transform to calculate and model epistasis in genetic landscapes of arbitrary shape and complexity (Faure AJ, et al., 2023)
Here you'll find the Python code to reproduce the figures and results from the computational analyses described in the paper.
- 1. whmatrixextms-benchmarkin.ipynb Jupyter Notebook to perform benchmarking analyses and plot heat map representations of matrices.
- 2. whmatrixextms-validations.ipynb Jupyter Notebook to reformat published background-averaged epistatic coefficients and simulate multiallelic genetic landscape.
- 3. whmatrixextms.py Python script to fit sparse models to the fitness landscapes.
- 4. whmatrixextms_plot.py Python script to plot model results.
You will need the following dependencies installed:
- Python >=v3.9.9 (NumPy, pandas, scikit-learn, SciPy, Matplotlib, seaborn>=v0.12)
DMS data (fitness estimates) and additional files required to run the above analyses can be downloaded from here.
Bash scripts with command-line options for fitting sparse models to fitness landscape for each dataset are also included in this repository.