XgenePy is a Python port of the XgeneR package for modeling cis and trans regulatory differences between homozygous strains. It mirrors the core XgeneR workflow while replacing the R edgeR dependency with the Python edgepython package.

The package centers on:

• building the same allele-aware design matrices used by XgeneR
• fitting negative binomial GLMs with edgepython
• testing cis and trans hypotheses
• generating the same assignment and proportion-cis summaries
• producing diagnostic plots in matplotlib

Create the conda environment:

conda env create -f environment.yaml
conda activate xgenepy

Then install the package in editable mode:

pip install -e .

import pandas as pd
from xgenepy import FitObject, fit_edgepython, get_assignments_and_plot

counts = pd.read_csv("../XgeneR/inst/extdata/BATcold_ballinger_counts.csv", index_col=0)
metadata = pd.read_csv("../XgeneR/inst/extdata/BATcold_ballinger_metadata.csv", index_col=0)

fit_obj = FitObject(counts=counts, metadata=metadata, trans_model="log_additive")
fit_obj = fit_edgepython(fit_obj)

results = get_assignments_and_plot(fit_obj, make_plot=False)
print(results.dataframe.head())

• src/xgenepy/model.py: model construction, design matrices, contrasts, and fitting
• src/xgenepy/plotting.py: plotting and assignment helpers
• data/: copied example datasets from XgeneR, including BATcold, cold, and full Ballinger files
• notebooks/: Jupyter notebook equivalents of the XgeneR vignettes
• tests/: pytest coverage for design and fitting smoke tests
• scripts/: example script for running the BATcold dataset