Skip to content

Contributing

Alpha

HOMERun is in alpha. Interfaces and layout may change, so check with the maintainer before building anything large on top of the current API.

Thanks for helping improve HOMERun.

Development setup

git clone https://github.com/NguyenThomas986/homerun.git
cd homerun
python -m venv venv
source venv/bin/activate        # Windows: venv\Scripts\activate
pip install -e ".[test]"

The editable install puts the homerun command on your PATH and lets the test suite import the package. (The test suite also adds the repo root to sys.path itself, so pytest works even before an editable install.)

Before opening a pull request

  • Run the full suite and make sure it's green:
pytest

  • Add tests for new behavior. The suite runs without any external tools, so new logic should be unit-tested against the tmp_path fixtures rather than a live cluster.

  • Keep output directory and file names stable. Single-command reruns and SLURM array indexing depend on the existing layout (RawData/, Trimmed/, Aligned/, TagDirs/, bedGraphs/, TSS/, QC/) and on tag-directory / bedGraph naming (<sample>_<assay>_<replicate>, <sample>_<assay>-combo). Renaming any of these breaks reruns, so avoid it unless that's the explicit point of the change.

Reporting issues

See Report an Issue.

Code style

Scope notes

  • HOMERun currently identifies every sample purely from its FASTQ filename. Manifest input (CSV / TSV / Excel) is out of scope for now.