Official repository for ProMetaInfer: A tool for inferring missing metadata in proteomics data using machine learning and database search.
-
Python >= 3
-
Install OpenMS
-
Install Docker
-
Download Swissprot database as fasta and speficy path to it (https://www.uniprot.org/uniprotkb?query=reviewed:true)
-
Download Comet and specify comet.exe file (make sure it is executable, https://github.com/UWPR/Comet/releases/tag/v2025.01.0)
-
Pull docker container for param-medic predictions with (https://hub.docker.com/r/elisamaske/predict_tolerances_container)
docker pull elisamaske/predict_tolerances_container
We recommend running ProMetaInfer in a dedicated Conda environment for better reproducibility and dependency management.
To set it up, run the following:
conda create -n prometa_env python=3.10
conda activate prometa_env
pip install -r requirements.txt
Make sure all additional tools and databases (OpenMS, Docker, SwissProt FASTA, Comet) are properly installed and configured as described above.
python infer_metadata.py --mzml_dir <path_to_mzml_files> \
--output_dir <path_to_output_directory> \
--idxml_dir <path_to_idxml_directory> \
--database <path_to_protein_database> \
--comet_exe <path_to_comet_executable>
Important
Give complete path to comet executable