<@U039DEJ1RJS> I seem to remember you are a fan of...
# team-maintainers
m
@Matthias De Smet I seem to remember you are a fan of massive modules update 😆. I am looking at https://github.com/nf-core/modules/pull/12983/ which adds gzip has a dependencies on some modules, there was a way via wave to change the base layer, right and have gzip there by default?
While we're at it, this PR also add find-utils as a dependency. I think it's a sign we need to replace this
Copy code
INDEX=`find -L ./ -name "*.amb" | sed 's/.amb//'`
by something smarter
m
Yes! Containers need to be created through the seqera wave API or you need the use a slightly modified version of #CE5LG7WMB (PR pending approval) github.com/nf-core/tools/pull/4449
s
I don't understand why the base wave image doesn't have gzip
m
While we're at it. I see that here, this PR add strandness as an input for modules/nf-core/sentieon/rsemcalculateexpression/main.nf I'd rather we keep that in args, similarly to how we did it in https://github.com/nf-core/rnaseq/pull/1913/ cc @Jonathan Manning what is your take on this ? (as I saw you approved and merged said PR) And maybe we need to document this properly, or add strandeness as an official field
@Simon Pearce I have a feeling you have an opinion about this
m
that’s the thing, the docker image does, but the singularity image doesn’t. I wanted to pin it to the
conda-forge
version to avoid any drift, but Paolo wasn’t a fan 🤷🏻
s
My opinion is that strandedness should be an official field
👍 1
As I've said several times
😆 1
m
Can we maybe split these three topics in a separate thread so answers don’t get buried? 😅
🤣 1
m
sorry
j
this PR add strandness as an input for modules/nf-core/sentieon/rsemcalculateexpression/main.nf
Yeah, that's clearly wrong, it's not mandatory I think
My opinion is that strandedness should be an official field
I was happy enough when that was the situation too, don't really understand why we had to strip it, been dealing with some fallout of that this week