search

KosinskiLab / AlphaPulldown

https://doi.org/10.1093/bioinformatics/btac749
GNU General Public License v3.0
176 stars 39 forks source link

Replace this with non-cctbx solution #343

Closed jkosinski closed 2 days ago

jkosinski commented 1 month ago

https://github.com/KosinskiLab/AlphaPulldown/blob/fea07b690c311f1b3fa88ac92522ff591e4d9bbe/alphapulldown/analysis_pipeline/af2_3dmol.py#L206

This would greatly facilitate installation through EasyBuild - at the moment preparation of EasyBuild package is stalled due to this independency.

boegel commented 1 month ago

cctbx is massive pain to install from, so avoiding that dependency would be a significant help, not only for installing AlphaPulldown using EasyBuild, but also to get AlphaPulldown included in EESSI

gchojnowski commented 1 month ago

I replaced all refs to cctbx in this file with biopython. it's slower, but cctbx is not there.

jkosinski commented 1 month ago

Thanks! You mean you replaced in the af2plots repo? Do you think you could replace it also here:

https://github.com/KosinskiLab/AlphaPulldown/blob/9d75f236e20a83a9dfd8396bc75a1516c4e190d6/alphapulldown/analysis_pipeline/af2_3dmol.py#L206 and submit a PR for AP?

gchojnowski commented 1 month ago

there is a separate branch wit a fix 343-..

jkosinski commented 1 month ago

Ah, ok, can you submit PR with Geoffrey as a reviewer? Thank you for changing this even if it slower!

boegel commented 1 month ago

Out of curiosity: how much slower when using Biopython?

To be clear: I'm very much in favor of getting rid of cctbx, since it's a pretty painful dependency from what I've seen (in terms of getting it installed)...

gchojnowski commented 1 month ago

boost is an issue, I guess, but that's what make cctbx faster. anyway, not a big deal here; simple code and reltively small structures.