Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Re-run RF maker prediction for *all* nodes in 202002013 #55

Closed
dosumis opened this issue Mar 10, 2021 · 10 comments
Closed

Re-run RF maker prediction for *all* nodes in 202002013 #55

dosumis opened this issue Mar 10, 2021 · 10 comments
Assignees
@BAevermann

Comments

@dosumis
Copy link
Contributor

dosumis commented Mar 10, 2021

Currently we only have markers for nodes names in an earlier version of the dendrogram.
@dosumis
Copy link
Contributor Author

dosumis commented Mar 31, 2021

Hi @BAevermann - any progress on this?

@BAevermann
Copy link
Collaborator

I am working on it. Taking the longer road. Instead of manually generating the necessary labels, I am writing code so that we can more easily pipeline it in the near future.

One issue is that we should request that the Allen release a cell to leaf node accession # mapping. As far as I know, the cells are still mapped to the preferred labels (however they update their website regularly...)

@dosumis
Copy link
Contributor Author

dosumis commented Mar 31, 2021

I am working on it. Taking the longer road. Instead of manually generating the necessary labels, I am writing code so that we can more easily pipeline it in the near future.

Not sure I follow. We have label generation code for the ontology and the dendrograms and your lookup for gene IDs -> labels.

One issue is that we should request that the Allen release a cell to leaf node accession # mapping. As far as I know, the cells are still mapped to the preferred labels (however they update their website regularly...)

Did you ask @jeremymiller - bit shocking we're not running everything on accessions already. I really think labels are unsafe.

I think this is part of a more general problem with standardising pipelines - where data comes from and in what form. Would be good to have a dedicated call/hackathon on this - maybe towards the end of next month?

@jeremymiller
Copy link
Collaborator

We (by which I mean the technology and project teams at the Allen Institute) are working on implementing a taxonomy service which would allow scientists to run a clustering on a specific set of cells and tag this specific instance of the cells in the database (e.g., versioned data set) to specific instance of a taxonomy so it can be tracked and reproduced. At best it will be a few months before something like this is available (plus convincing scientists to use it) and so for now we just need to be careful with tracking these kinds of things. I do like the idea of a dedicated call/hackathon for how to deal with versions in the near term. All this being said, I'm not sure I understand the specific issue right now. Maybe we can discuss briefly at the ontology meeting Monday?

@dosumis
Copy link
Contributor Author

dosumis commented Mar 31, 2021

Specific, urgent issue: @BAevermann would like cell_set_accession to cell mapping for his analysis as he is relying on mappings to potentially unstable/outdated node labels (cell_set_preferred_alias?) right now. Would you be able to provide these to him for mouse MOP (also needed for other primary motor cortex dendrograms)? Presumably we know what data each of these corresponds to.

Re general call - Let's aim for end April (Monday is a holiday here so I may cancel call).

@jeremymiller
Copy link
Collaborator

This file should have what you need, but maybe not in the format you want: cell_to_cell_set_assignments_CCN202002013.zip. Alternatively, this group of files contains most of the input information for our taxonomy service for this taxonomy, which might have what you want with some conversions: mouse_m1_miniatlas_v5.zip. Let me know if you need something else.

@dosumis
Copy link
Contributor Author

dosumis commented Apr 7, 2021

@BAevermann - any progress? Do the files @jeremymiller provided give you what you need?

@BAevermann
Copy link
Collaborator

BAevermann commented Apr 7, 2021 via email

@BAevermann
Copy link
Collaborator

Added file to the markers folder containing almost all of the nodes (probably all the necessary nodes)

@dosumis
Copy link
Contributor Author

dosumis commented Apr 16, 2021

Added file to the markers folder containing almost all of the nodes (probably all the necessary nodes)

Awesome. Please check QC report (if you haven't already) @hkir-dev can help point you to it & discuss any issues

@hkir-dev - if no major issues can you try running a build?

@dosumis dosumis closed this as completed Jun 23, 2021
hkir-dev added a commit that referenced this issue Apr 27, 2024
@hkir-dev
fix issue #55
6b43814
hkir-dev added a commit that referenced this issue Jul 11, 2024
@hkir-dev
release of fix #55 descriptions
d7043df
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@BAevermann BAevermann

Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@dosumis @jeremymiller @BAevermann