search

frankligy / SNAF

Splicing Neo Antigen Finder (SNAF) is an easy-to-use Python package to identify splicing-derived tumor neoantigens from RNA sequencing data, it further leverages both deep learning and hierarchical Bayesian models to prioritize certain candidates for experimental validation
MIT License
44 stars 9 forks source link

Incompatibility Issue with mhcgnomes version in SNAF Pipeline #24

Closed yiolino closed 10 months ago

yiolino commented 10 months ago

Environment: OS: Ubuntu 20.04 Python Version: 3.7.17 Virtual Environment: Python venv

Description: I encountered an error while progressing through the tutorial for the SNAF pipeline. The issue occurred when executing the command jcmq.run(hlas=hlas, outdir='./result').

Error Message: The following exception was raised during the execution:

Exception: binding prediction error: cannot import name ‘Class2Pair’ from ‘mhcgnomes’ (/snaf-pipeline/.venv/lib/python3.7/site-packages/mhcgnomes/__init__.py)

Steps to Reproduce: Set up a Python virtual environment in Ubuntu 20.04. Follow the installation guide to install SNAF. Proceed with the tutorial steps until reaching the jcmq.run command.

Observed Behavior: Upon the installation of SNAF, mhcgnomes version 1.8.6 was automatically installed. However, this version led to the aforementioned exception.

Resolution Attempt: Downgrading mhcgnomes to version 1.7.0 resolved the error, and the jcmq.run command executed successfully.

Suggestion: It seems that the current version of mhcgnomes (1.8.6) is incompatible with the SNAF pipeline.
A possible cause could be the absence of a fixed version requirement for mhcgnomes in the mhcflurry requirements.txt.
It may be beneficial to specify a compatible version of mhcgnomes to ensure the stability of the SNAF pipeline.

frankligy commented 10 months ago

Thanks for spotting that, I'll update the readme once I fixed some other issues at my end.

yiolino commented 9 months ago

Thank you for accepting my Pull Request.