Olesia Kondrateva
06/25/2025, 10:21 AMvep_cache is empty, so VEP annotation technically runs but doesn’t generate meaningful output, so pVACseq also dont produce any output.
• I had to write a separate module to download custom VEP plugins.
Would it be better to use my working custom VCF or adapt the nf-core test data, knowing it won't produce output?
Also, a second question:
The tool depends on GitHub packages not available on PyPI, so the environment.yml includes:
- pip:
- tensorflow==2.15.1
- pvactools==5.3.1
- git+<https://github.com/griffithlab/bigmhc.git#egg=bigmhc>
- git+<https://github.com/griffithlab/deepimmuno.git#egg=deepimmuno>
- mhcflurry==2.0.6
What is the recommended way to handle this to comply with nf-core guidelines?Maxime Garcia
06/25/2025, 10:36 AMMaxime Garcia
06/25/2025, 10:38 AMNicolas Vannieuwkerke
06/25/2025, 10:45 AMMaxime Garcia
06/25/2025, 10:46 AMOlesia Kondrateva
06/25/2025, 11:04 AMMahesh Binzer-Panchal
06/25/2025, 11:05 AMOlesia Kondrateva
06/25/2025, 11:08 AMFelix Lenner
06/25/2025, 11:47 AMSimon Pearce
06/26/2025, 5:54 AMOlesia Kondrateva
06/30/2025, 11:49 AMwave-cli, and since it's a new tool for me, I just wanted to clarify: does it generate containers? I already have a container provided by the tool developers, so I don't need to build a new one. My goal is simply to add support for running the tool via a Conda environment as wellMahesh Binzer-Panchal
06/30/2025, 11:51 AM--freeze )