Consider AnnData & CITE-seq postprocessing about scpca-nf HOT 4 CLOSED

I don't think this affects processing for the portal at all as the extra file should simply be skipped and not added to the portal at this stage.

So basically, just don't copy over the hdf5 files into the portal is your view it sounds like?

I think this does probably make the most sense to minimize modifications to the workflow. I think all that would need to be done is when you copy things over from the nextflow-ccdl-results bucket to the portal inputs bucket is excluding any files with the .hdf5 extension.

outputting an extra file (noting that in the release notes).

To me, seems like this is not experimental if we get #355 in. I might favor holding off on merging #355 into development which makes this feature more in-progress, but I don't feel super strongly.

I think we could still merge it? The only thing is that it outputs two files for CITE-seq instead of just one for RNA. If we are including it in the release notes I'm not sure if it really matters either way.

from scpca-nf.

I guess it depends how anxious we are to get the next release out. If we have some time I might make the process optional, with a config variable.

But if we are anxious to get the release out, I don't see a problem with leaving it as is and outputting and extra file (noting that in the release notes). I don't think this affects processing for the portal at all as the extra file should simply be skipped and not added to the portal at this stage.

from scpca-nf.

But if we are anxious to get the release out

We have portal updates planned for a deadline of 7/28, and scpca-nf itself should therefore be roughly end of next sprint.

I don't think this affects processing for the portal at all as the extra file should simply be skipped and not added to the portal at this stage.

So basically, just don't copy over the hdf5 files into the portal is your view it sounds like?

outputting an extra file (noting that in the release notes).

To me, seems like this is not experimental if we get #355 in. I might favor holding off on merging #355 into development which makes this feature more in-progress, but I don't feel super strongly.

from scpca-nf.

Discussion has been discussed, so we can close this out now. We'll keep the code as is, and for portal releases do some selective copying.

from scpca-nf.