Bioconda distribution

[jonasscheid]

Dear authors,

thanks, for the great tool. I'm concidering integrating SpecHLA as a model in https://github.com/nf-core/hlatyping. For easier usability and reproducibility it would be great to have your tool packaged up in bioconda. I could try assisting in that regard. Are there any plans to support this? Obviously there needs to be a solution found for novoalign. Maybe falling back to another mapper or so. Would be great to hear your thoughts on this!

[wshuai294]

Hi,

Thank you so much for your interest in SpecHLA! It would be fantastic to see it integrated into the nf-core/hlatyping pipeline.

Regarding Bioconda, I haven't had the chance to prioritize packaging it yet, so your assistance in that regard would be greatly appreciated. >
As for the mapper, we have already implemented Bowtie2 as a fallback. The tool is designed to detect the Novoalign license automatically—if it’s missing, it will seamlessly switch to Bowtie2.

Looking forward to hearing your thoughts on the next steps!

Best regards,
Shuai

[jonasscheid]

Great to know! Unfortunately, not straight forward in this case. I think a couple of things need to be changed. I'll put a claude plan here on changes that should be made. Note that binaries that are available on conda are not needed!

CONDA_PACKAGING_PROPOSAL.md

[wshuai294]

Thanks for your effort. Seems a lot of works are required to do. Really appreciate if you could help with this.

One point I must mention, the version of samtools DO matter. We use SamTools-1.3.1. The samtools depth parameter is different from new versions, which might cause bugs with new versions.

[jonasscheid]

A rough idea from my side: #69
Does this make sense to you while not breaking any functionality with the current code? :)

[wshuai294]

Great! I left some comments in the PR #69 .