search

jts / ncov-tools

Small collection of tools for performing quality control on coronavirus sequencing data and genomes
MIT License
47 stars 16 forks source link

Issue building SnpEFF database #74

Closed DarianHole closed 3 years ago

DarianHole commented 3 years ago

SnpEFF seems to have updated to a new version5.0-1 which is causing the database build to fail when it is installed instead of version 5.0-0.

Likely due to the line here as given to me by @kdmack87: https://github.com/jts/ncov-tools/blob/21fd71f4fc9207a592d6684997d6b66bc2d4ad3a/workflow/scripts/build_db.py#L9

I am not sure if the best solution would be to specify multiple options in the above file or to set the environment dependency to be exactly 5.0-0 (not sure if conda allows the - in the environment files for a version number?)

I worked around this by going into my conda environment and re-naming the folder from 5.0-1 to 5.0-0, building the database, and then re-naming it back but that is not super ideal. Once I did this everything else still worked as intended

jts commented 3 years ago

@rdeborja could you look at this?

rdeborja commented 3 years ago

@DarianHole Previously the workflow would download the MN908947.3 version of the snpEff files from their repository of db build leaving the build_snpeff_db workflow redundant. Since the latest snpEff update, it looks like the db file is corrupt or the link is broken. I removed the fixed version of snpEff from the script and annotation workflow so that the build will work again. Also, it supports multiple installed versions of snpEff in a single conda environment. I have added branch fix_snpeff_build to address this and submitted a PR. Feel free to try out the branch and let me know if you run into any issues.

rdeborja commented 3 years ago

@DarianHole the branch has been merged and should resolve the db build issue.

DarianHole commented 3 years ago

Awesome, thanks! I'll run it through right now and update you with the results

DarianHole commented 3 years ago

Works as intended! Thanks again for the quick fixes to this and the snpEFF logs