You need a working installation of Snakemake. Then:

git clone <this-repo>

The pipeline uses conda environments to manage dependencies, which are handled automatically if you run snakemake with the --use-conda flag.

It also relies on some tools (ThermoRawFileParser, SearchGui and PeptideShaker) which do not have conda packages or docker images available for the versions we used. These tools are downloaded on-the-fly by snakemake, so you do not need to install them separately.

There is a small test-data set, using a few assemblies from MGnify and two RAW files from PRIDE. To fetch the (~GB size) RAW files, which are too big for this git repository:

./test-data/pride/fetch-pride-test-data.sh

This downloads two RAW files into test-data/pride/.

Then:

conda activate snakemake # (assuming you installed snakemake with conda, into an env called snakemake)
cd MetaPUF
snakemake --cores 4 --use-conda

This will run the pipeline on the small dataset, and put results into ../test-run.

Edit the config/config.proteomics.yaml and sample_info.csv files to point the pipeline at real data. sample_info.csv is the mapping of MGnify to PRIDE datasets, and in the config parameters.input_dir and parameters.raw_dir refer to the MGnify and PRIDE data folders respectively.

• You can run a dry-run to check for any syntax errors

 Snakemake  -np

• To run the workflow

 Snakemake --cores 4 --use-conda

• Using LSF on an HPC cluster:

bsub -n 4 -R "rusage[mem=4096]" -J metapuf -u $USER -o job.log -e job.err snakemake --cores 4 --use-conda

• Tips: IF the pipeline got collapsed during running, you can always try to run a dry-run Snakemake -np first to check how many rules have been successful executed, and if you are sure that some files are generated correctly, you can use snakemake --cleanup-metadata <filenames> to skip these files to be re-generated. However, sometimes snakemake --cleanup-metadata <filenames> doesn't work, you can also try to manually delete the .snakemake/incomplete directory.

git clone https://github.com/PRIDE-reanalysis/MetaPUF.git
cd MetaPUF
conda activate snakemake  # or another conda/venv if you prefer
pip install ".[dev,docs]"
pre-commit install

This installs the development requirements, and installs the pre-commit hooks which format the code correctly while commiting changes. You can also manually format the code using black .. It also installs mkdocs, which is used to build the documentation.

Change the markdown files in the docs/ folder. Then mkdocs serve to view the documentation site locally.

As part of our efforts toward delivering open and inclusive science, we follow the Contributor Covenant Code of Conduct for Open Source Projects.