Manipulate counts matrix files and cook scRNA-seq data from command line
You can install a local version of the package (along with python dependencies) by cloning this repository and running the following command from the main directory:
pip install -e .This package is intended to automate manipulation and processing of scRNA-seq counts matrices from the command line, eliminating the need for interactive python sessions with repetitive, manual scanpy functions.
# print out all kitchen command options
kitchen -h
# ex1: print information about anndata object (saved in .h5ad format) to console
kitchen info <path/to/.h5ad>
# ex2: process a filtered .h5ad file from raw counts, performing unsupervised clustering,
# cell cycle inference, and UMAP embedding colored by genes, mito percentage,
# and cell cycle phase, along with leiden clusters and PAGA graph
kitchen recipe <path/to/.h5ad> -p -cc -c arcsinh_n_genes_by_counts pct_counts_mito phaseFull documentation is available at codyheiser.github.io/kitchen/.
After making changes, lint, format and document code before committing:
make format # black-formatting
make lint # lint Python code
make doc # pdoc3 documentationThen, following git commit, create new version tag and push to remote:
git tag -a vX.X.X -m "tag message"
git push --follow-tags