A community effort to collect a curated set of analysis pipelines built using Nextflow.

nf-core

<@UE6D8290F> let me know that rnaseq was costing a lot at release time.

I think this is mostly a product of the approval trigger: <https://github.com/nf-core/tools/blob/b6c5737fde996c824b87d19d1c51a29b4810db1d/nf_core/pipeline-template/.github/workflows/awsfulltest.yml#L17>

I can turn that off in rnaseq, but should we do it in general? Seems like the release trigger should be enough, with ad-hoc manual triggers as needed

I think the motivation was that people should not release their pipelines until the full size run actually completes so we have fewer released pipelines that only work on the home infrastructure and/or not on bigger data. In reality this is working semi well....

Was this triggered multiple times every time someone gave an approval or why did it cost so much?

Yes, it gets triggered per review, then at release time

(plus any ad-hoc runs I choose to do to check up on myself before asking for review)

Plus it runs for both STAR and RSEM each time (from before I was maintainer).