Open jeffkaufman opened 2 weeks ago
I do think of this as out-of-scope for the core workflow, and honestly an awkward fit for a pipeline-based approach in general. I think it should be the user's responsibility to get the data into a suitable state to run the pipeline (though I don't think that should necessarily mean it has to be in our S3 in a very specific structure).
That said, I'm not opposed to including some auxiliary scripts in the repo that users could use to do this in common use cases.
One place where we're still using the v1 pipeline is to import data into S3. This includes the hacky import-accession.sh as well as
[mgs-restricted](https://github.com/naobservatory/mgs-restricted/)/import-*.sh
for several of our partners. These are just linux scripts and don't use Nextflow.@willbradshaw are you thinking these are out of scope for this project and should stay in mgs-pipeline / mgs-restricted?