Bring hadge to nf-core Thank you for the review, @Nico Trummer and @Lukas Heumos! I have incorporated your feedback, and all tests are still passing. So what's next?

From my perspective: 1. Merge the PR 2. Set default branch to

dev

3. Send a message to #C09H6NYHR9T for moving the repo to the nf-core organization 4. Check if the docs publish correctly 5. Change the CI configuration to use nf-core's larger runners, open a test PR to check if everything works (including the things we previously ignored due to 'disk space' issues) 6. Clean up the branch structure 7. Solve any open issues that would be classified as 'bugs', for feature issues decide what should be done before the 1.0 release and what can be done after. Create milestones and assign accordingly, to keep track 8. I will look at the entire pipeline again and check concordance with latest nf-core guidelines - there have been some recent developments in relation to the nextflow

strict

mode, see #C04QR0T3G3H 9. 1.0 release

All right, thank you. I'll try to go through it step by step. If I need help, I'll get in touch with you.

You're now admin

Thank you!

Oh sorry, I somehow didn't receive this ping

Crazy stuff

i fixed the link

Thanks! For your reference, the

master

branch is currently the proper nf-core release branch, the

main

branch was the default branch before we re-built everything for nf-core

Pretty smooth so far. Good job @Luis Heinzlmeier and @Nico Trummer!

okay, that branch duality might confuse some of the tooling. it is totally up to you if you want to use

main

or

master

, but there can only be one (you can just add a

_old

prefix to one or something similar)

Let's go with

main

then. More modern for better or worse

be sure to also set that in the

nextflow.config

I added nf-core's larger runners, but now the CI is skipping

Run nf-test

. Have I forgotten anything?

No, it's only running nf-test if changes have been made to relevant files

Thank you, Nico! And we want to trigger running the test again with the new CI configuration, don't we?

Yes, also be aware that you can configure the disk volume like this

Ouhh, I see! Thank you for your help with this Nico. 🙏

I created a 1.0 release review issue on github which is a bit more detailed than the list on the beginning of this thread

Great. I just thought through the task of cleaning up the branch structure and came across a warning. The template version is

nf_core_version: 3.3.2

but my

nf-core, version 3.5.1

. Should I then run

pip install --upgrade nf-core==3.3.2

or do we first want to update

TEMPLATE

? In the tutorial, they say

master

is the default. In our case, I keep

dev

as the default, don't I? So, can I then also follow the tutorial when

dev

is the default branch?

dev

remains the default until the 1.0 release (as it is the only branch that actually contains the pipeline logic until the 1.0 release PR is merged) I think this tutorial does actually mostly not apply to us, it looks like it is targeted on pipelines that already have been released. For our use case, it should be sufficient to rename the branches and change the

nextflow.config

@Lukas Heumos • Do we want people who use the pipeline to cite it via Zenodo, bioRxiv, or Genome Biology? (The nf-core template suggests to cite with Zenodo, so I did both for now: Zenodo and Genome Biology) • There was a recommendation to add many keywords to the repo in the "new pipeline proposal thread" in November 2023. Concering hadge, we only have a few keywords now. Do you have any ideas for more keywords? How do you like my keyword ideas? ◦ hashing-based-deconvolution ◦ genotype-based-deconvolution ◦ donor-deconvolution ◦ cell-hashing ◦ genetic-demultiplexing ◦ donor-assignment ◦ hto ◦ hashtag-oligonucleotides ◦ lipid-tagging ◦ Maybe the integrated tools? vireo, demuxlet, freemuxlet, souporcell, scsplit, htodemux, multiseq, hasheddrops, demuxem, hashsolo, gmm-demux

Do we want people who use the pipeline to cite it via Zenodo, bioRxiv, or Genome Biology? (The nf-core template suggests to cite with Zenodo, so I did both for now: Zenodo and Genome Biology)

Only Genome biology

◦ hashing-based-deconvolution

◦ genotype-based-deconvolution

They're a bit complex because of the

-based

, While not wrong, I'm not sure whether it helps with search. Given it's keywords, we might just scrap the

-based

. The rest look good!

And I just saw that I don't have the permission to rename the

main

branch. 😂 So @Lukas Heumos could you maybe rename the

main

branch to

legacy-main

or to whatever name you like :)

i can do that

Also, I cannot rename

master

😅

It's not could test in the cloud but actually having these tests and running them in the cloud is part of the standard release process I am myself no expert in setting up and running the AWS tests, I know there have been some issues with running the tests in the last half year, but I think/hope they are fixed by now. To provide your data for upload, I think you need to use #C09H6NYHR9T. Then the people there can probably also assist you with the remaining steps. You are lucky, as currently the nf-core hackathon is ongoing, so it is quite likely that this will be resolved quite quickly if you initiate it today/tomorrow