Quite crazy bug in version reporting fixed in

eggnog-mapper

module: github.com/nf-core/modules/pull/13016

shouldnt the snapshot also require an update?

But also just be adressed in eggnog mapper istelf? github.com/eggnogdb/eggnog-mapper

Ha, just left this comment at the Pr.. 😛

Regarding the snapshot: No — the snapshot already holds

2.1.13

and still matches. The tool is transitioning to v.3 and the error has been reported before, so I don't think it's any use yet. (Also, this is from Peer Bork's lab I believe, and since his death they have other things on their minds.) I think we can go ahead with a local fix. (I'm fine with waiting for an nf-core PR to release metatdenovo, but not upstream of that.)

go ahead

Do so. I think @Danilo Di Leo has an InterProScan PR for #C02FC6VFQG1 in the works. Will be prioritized for 2.1. Should he hold off a bit?

Not because of me, if I make it on time, sure. But I dont want to block you 😛

For us probably best if you do it first. We had a discussion a couple of months ago regarding this and then I understood you as this was a local workflow at EBI that was difficult to port to nf-core -- but you're doing it?

definitely no. that's its own non nf-core thing. we will have to stick to the latest old version with bioconda - but opened two issues through testing; one has to do with version mismatch of container + url for proteinannotator, and the other one is the path tha is fed to the module itself (something about /data/ path, and not having to bind as a separate volume due to some handicraft in the current version of the module) . Stay tuned