Large numbers of entities are not exporting to the API

geneontology / noctua

Graph-based modeling environment for biology, including prototype editor and services

http://noctua.geneontology.org/

BSD 3-Clause "New" or "Revised" License

38 stars 12 forks source link

Large numbers of entities are not exporting to the API #906

Closed ValWood closed 3 months ago

ValWood commented 4 months ago

@kimrutherford has made a query so that we can track our genes in GO-CAMs https://www.pombase.org/results/from/id/f99d8133-3206-4941-b44e-9314e7cae3d2

However, it seems that any part of the model that does not have evidence on any outputs/inputs are not exported?

i.e here:

**http://noctua.geneontology.org/workbench/noctua-visual-pathway-editor/?model_id=gomodel%3A665912ed00000459

dfg10 gpt2 alg1 are not exported (even though they do appear in the pathway widget).

I'm not sure why we need to add evidence to inputs and outputs that are recorded in the MF activities? the MF evidence should be sufficient?
It would be nice if there was a way to access a list of the entities in production models that are missing evidence (I have many which I am waiting to plug in the annotation when it comes through from PomBase. These aren't exported either.

thanks

VAL

@PCarme I think we have done a lot more than it appears!

vanaukenk commented 4 months ago

@ValWood

I think the original idea was that curators wouldn't release models to production until they were fully evidenced. That said, there may be things we can do to make it easier and faster to add evidence so models can be released more quickly.

I'd need to know more about your workflow to help with this.

vanaukenk commented 4 months ago

@ValWood - also one curation issue I noticed when looking at the model above was that there are some MFs annotated with 'occurs in' a protein-containing complex' child term. You'll want to update those CC terms to a non-protein-containing complex term. Let me know if you have any questions.

ValWood commented 4 months ago

Hi @vanaukenk 1) I will fix the CC annotations, I didn't realise this at first

~2) I think we can make all of the missing non-experimental annotations in Noctua rather than in PomBase (I'll talk to @PCarme about this, (although it is confusing because we do access these genes via the widget, just not from the API).~ ignore, we decided to continue with our current workflow

However, if there is anything that can be done to speed up our experimental annotations through to Noctua that would always be useful. 3) I still don't really want to add evidence to all of the inputs and outputs because these are already supported by the MF activity term? This should be automatic?

~Did you mention that there was a way to do edits without deleting and recreating the annoton and losing the connections? if so, can you point me to the docs for this?~ this actually works, it just happens that the model I tried to edit yesterday was one of the ones below