Skip to content

A Python package to perform unsupervised discoveries of motifs from tandem mass spectrometry data.

License

Notifications You must be signed in to change notification settings

glasgowcompbio/pySubstructures

Repository files navigation

pySubstructures

A Python package to perform unsupervised discoveries of motifs from tandem mass spectrometry data.

Python TODO:

  • Add codes from the lda repo.
  • Tidy up codes, and keep only the ones we actually use.
  • MolnetEnhancer support
  • Gensim support
  • Online/offline MotifDB support when running LDA inference
  • Create a Python package

Maybe:

Visualisation:

  • Stand-alone viewer using Dash

MS2LDA.org TODO:

  • Replace the lda codes used on the server with this package

Environment Setup

This project uses a Conda environment to manage dependencies. To set up the environment, ensure you have Conda installed, then run the following command from the root directory of this project:

conda env create -f environment.yml

This will create a new Conda environment named pySubstructures and install all required dependencies, including Black for code formatting.

Activating the Environment

After installing the environment, activate it using:

conda activate pySubstructures

For Contributors

We use Black, the Python code formatter, to ensure code consistency. It's included in the project's Conda environment. Before submitting any code, activate the pySubstructures environment and run Black:

black path/to/your/python/file_or_directory

This will format your code according to the project's standards. Ensure you do this before creating a pull request. You can also configure your IDE to auto-reformat using black upon saving (recommended).

Testing

To run tests, ensure the pySubstructures environment is activated and run pytest from the project root:

pytest

About

A Python package to perform unsupervised discoveries of motifs from tandem mass spectrometry data.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published