@Nico Trummer I have several questions regarding cell cycle implementation, wrote them in the comments https://github.com/nf-core/scdownstream/issues/170#issuecomment-4045845950

If you like the #C077PUX8DBP feel free to ⭐ on github (would mean a lot to me)

@Nico Trummer Hi! I have a question about sample sex prediction, leaved a comment with question in the issue: https://github.com/nf-core/scdownstream/issues/171

@Nico Trummer Hi! I'm working with online team, specifically with @Stepan Akinshin on sex prediction. Nice to meet you! Also added my question in the issues. Thank you!

https://github.com/nf-core/scdownstream/pull/261 scDblFinder PR open for review @Nico Trummer

A lot of questions right now, but I have a meeting for the next hour🥲

Don't stress yourself bud, do it when you do it 🙂

@Kurayi great presentation and much work was done! The PR meme made me laugh a lotkek

Thanks, I'm glad you liked it 😆

This one yeah? 😂

Sorry guys xD

I'm home from work now, so the hacking starts for mecattyping

@Nico Trummer I have some ideas regarding the testing data, reflected them in the comment! Still not urget for now! https://github.com/nf-core/scdownstream/issues/170#issuecomment-4048794248

So the problem why the CI did not pass for most PRs was that

hugounifier

(explanation in the thread) relies on data that comes from the HUGO database. Apparently they changed the database multiple times in the last weeks. In nf-core, tests are mostly based on hash snapshots of the output files of processes. So if the data from the API changes, the snapshots mismatch. If this would occur once every few months it would not really be a problem, but recently it has happened every few days. So I changed the tests that have this dependency to use less strict testing. You can see the changes here. With this merged, the CI for the SingleR PR from @Kim Fehl started working. Then I took the freedom to merge this changes into the other PRs (@Kurayi @Ze Yu @Miguel Rosell Hidalgo) so hopefully these will pass too now and can soon be merged

I just wrote a long paragraph about about modifying anndata files vs saving "what's new" throughout the pipeline. This is a key concept of the existing implementation and probably relevant for many of your tasks. Feel free to have a look here With this I call it a day, thank you for all of your effort, I'm quite impressed by how quickly you all got a hang of it. Looking forward to merging some more PRs tomorrow!

Good work man! Got home from work and got to work😂Hopefully you didn't tire yourself out too much ✨

@Nico Trummer by the way, did you change the doublet tools to use

mix

as part of the update you did yesterday? I could also get aorund to doing that soon

Hey y'all I'm done for the day, gotta travel home unfortunately 😞. It was lovely working with you all, I will keep working post hackathon, although at a slower pace. This year was a huge improvement imo, I should have an easier time contributing from now on

@Nico Trummer the PR for Cell Cycling is ready! https://github.com/nf-core/scdownstream/pull/266

We might join in metaverse?

Online and onsite team

Me and @Aditya Singh were discussing how to fix snapshotting for the failing modules on GHA with some floating point number rounding or somesuch, but I saw that there is some activity on https://github.com/nictru/nft-anndata. Are you already working on implementing this for the pipeline, @Nico Trummer, or do you want some help?

Another shout-out to @Nico Trummer, who finished PR #271 I started as a simple container optimization… which somehow (with all the versions updates) turned into a full test suite refactor 😅 Thanks a lot for all the time and effort you put into this. I believe it’ll pay off as a solid long-term improvement for the project!

Awesome! Am I to understand that this PR being merged means that we should now have cross-platform stability of testing, and I can update my own PRs with new snapshot runs after merging this in?

@Nico Trummer I made changes to this earlier this week, in case you didn't get review request notification from github https://github.com/nf-core/scdownstream/pull/261#pullrequestreview-3943432147

Any suggestions for ways to collaborate with @Kevin RUE ?

Hi, I'm trying to figure out a couple of things from the scdownstream pipeline, but I'm not sure how it works. I want to analyze a dataset that has been collected in a single batch, single lab. I believe I do not need any kind of integration method, however I don't see "integration_method: none" documented. Am I expected to use an empty string there?

I'd like to get both the LogNormalized counts and the z-scaled counts out of the pipeline. The metro map of the pipeline shows the algorithms used on each step but it is not clear to me what count matrix is used on each step, so it's hard for me to understand what I should expect adata.X to contain on each step. I've seen the celltypist template calculating log normalized counts, but I believe they are not reused afterwards. Or is this step happening in all cell type annotation methods for some reason? Sorry for asking multiple questions. I'm happy to provide pull requests improving the documentation once I know the answer if you believe documenting this can be helpful to others

Hello, The bioinformatics core at MDIBL is incredibly interested in using

scdownstream

for the cell type detection and annotation prediction modules. At MDIBL, we work primarily with non-mammalian animal models (zebrafish, axolotl, drosophila, etc.), and although the custom gene list parameter is available for cell cycle scoring, the cell type prediction algorithms are entirely human/mouse centric models trained on human/mouse datasets and often the orthology mapping from human to say, axolotl, is not complete. My question is, are there any people in the community who are working on scaling this pipeline’s capabilities for more species models to be trained or included? Or is this a task that has already been solved? If not, I would like to work with the community to add native organism annotation prediction modules for non-mammalian animal models using several publicly available single cell expression atlases as training reference datasets. I am a complete Nextflow beginner and would like to work together with anyone who may be interested in this project on developing such modules.

Hi! I’m really interested in using the scdownstream pipeline for my analyses, as it looks incredibly comprehensive. However, I noticed it’s still in development. Do you have any idea when v1 might be released, or how much work is left before then? Thanks!