Does the 1.0 release support submitting co-assemblies or is that still to come?

not yet, there is PR github.com/nf-core/seqsubmit/pull/66 you can potentially try from that branch maybe @Sofiia Ochkalova knows better

I still need to finish doing that - it got stalled as the ENA upload server went down and I went on paternity leave

I think we never tried it.. maybe you will just get a new ERR accession. I would recommend contact ENA and ask them to give you all data from you webin account. It is very easy for them. Then you can see what went through and what not

I will try and get the uploading started again this week - it needs finishing as we just re-submitted the paper after review. It will take quite a while - we did a back-of-the-envelope estimate and we have nominally got ~17 terabases of sequencing data to upload

do you remove host before upload?

no

is host different? not human, right?

Oak trees (Quercus robur) for all samples. But different trees

I'm merging co-assembly branch with the latest dev. I'll check if it works and will come back to you. You can try this branch directly, it calculates coverage incorrectly at the moment, but you don't need this feature anyway AFAIR One thing to consider: For each co-assembly all included source samples should have the same metagenome field name in ENA, e.g. "tree metagenome". Otherwise pipeline will fail. It's possible to set a value via parameter, but it will be applied to all assemblies in the samplesheet

Amazing! I will finish getting the reads uploaded and then let you know how it goes

Hi Jim I've tested the branch, seems to be fine You can try to upload once you're done with reads 😃

Perchance do you know what happens if you try to upload reads that you’ve already uploaded? Not exactly sure where the thing stalled to re-start the upload!

FYI: Using Webin-CLI, uploading a previously-uploaded manifest file just exits cleanly! Might require the output files to be in place already, though.

Yes, there is actually a inconsistent behaviour bug in this tool 😅 If the data is uploaded and output files are there - it exists cleanly, if the data is uploaded but output files are deleted - it fails with error I'm glad it worked for you 👍

Have you ever seen an error about an invalid username/password? I am getting intermittent failures that seem to disappear on a retry with an identical command...

Yes... It happens when you overload their submission system

I have 200+ manifest files, each which describes 2 read FASTQs (fwd/rev)

Well, it's not many, should be fine thinker

Currently traversing them with:

ls *.manifest | xargs Webin-Cli

Never used xargs for this, maybe it's too fast? 🤔 Or ENA server just not feeling well today

It wasn't having the same problem a couple months ago when I was last doing this :( So maybe it's just the server

I would try again tomorrow