This repo was used during the development of pycom. The current repo can be found at:
Current branch: https://github.com/scdantu/pycom
Website: https://pycom.brunel.ac.uk/
Python Library & server-side API for accessing the PyCom database.
PyCom is an open-source, queryable database of Protein Residue-Residue Contacts (Coevolution Matrices) and annotation data derived from UniProt (Swiss-Prot to be precise!) using HHBlits/HHFilter/CCmpred.
More info can be found here:
It was primarly developed by me over the summers of 2022/2023. My work was financed by two grants by the Dept. of CompSci at Brunel University London.
This repo was used during development and now serves as an archive. The main repo can be found at: https://github.com/scdantu/pycom
The code for the creation of the database, that was deployed on the (Jade2)[https://www.jade.ac.uk/] HPC can be found at: https://github.com/cemiu/pycom_generator
Install the PyCom library with:
pip install git+https://github.com/scdantu/pycom
Requirements:
- Python 3.8+
- numpy / pandas / h5py / requests, installed automatically
Get the pycom.db and pycom.mat files (115 GB total) from:
https://pycom.brunel.ac.uk/downloads/
Not necessary for remote use.
Examples can be found in example.ipynb.
This requires the pycom.db and pycom.mat files.
The library can be used remotely without downloading the pycom.db and pycom.mat files.
A tutorial can be found in example_remote.ipynb.
It is also possible to use the API directly, hosted at https://pycom.brunel.ac.uk/api/.