search

geneontology / noctua-models

This is the data repository for the models created and edited with the Noctua tool stack for GO.
http://noctua.geneontology.org/
Creative Commons Attribution 4.0 International
10 stars 3 forks source link

Now with evidence on GP to anonymous MF edges. #43

Closed dosumis closed 7 years ago

kltm commented 7 years ago

@dosumis I had thought that yesterday's was the final for a bit. I'll be on and off for a couple of weeks from Friday, so if there is anything that you'd like in before mid-September, we should coordinate.

kltm commented 7 years ago

@dosumis Are you unable to provide ORCIDs for your users? I believe that's what we want to go with for the GO annotations; I also recall we had this discussion before, but I can't find it. As well, weren't the titles a bit more informative in a previous iteration?

dosumis commented 7 years ago

I had thought that yesterday's was the final for a bit. I'll be on and off for a couple of weeks from Friday, so if there is anything that you'd like in before mid-September, we should coordinate.

Kind of inevitable that I'd notice more issues once we loaded. It would be good to get the latest batch loaded soon as we'd like to test GPAD conversion. I was too conservative with which edges I put evidence on for conversion to work reliably with the current load. We will likely want at least one more batch loaded before the annual SynGO project meeting on the 23rd September, but these will not be ready this week so this can be done when you get back.

@dosumis Are you unable to provide ORCIDs for your users? I believe that's what we want to go with for the GO annotations; I also recall we had this discussion before, but I can't find it.

Not yet. The various expert contributors need to be asked for ORCIDs and these need to end up in the SynGO JSON. We've asked the SynGO guys to request ORCIDs from their experts, but don't know how much progress has been made. @ftwkoopmans should be able to fill us in on progress. For now, we've agreed to use 'SynGO:' for the contributors field. This is no different from GO-CAM models that just have GOC: for contributor (don't know how common this is now, but I've certainly seen some).

As well, weren't the titles a bit more informative in a previous iteration?

I think the only difference is that they no longer have 'test' in the title. Longer, descriptive titles with 'syngo' in come from Barbara's work.

kltm commented 7 years ago

Fair enough. I'll try and get this batch up tonight (which, looking at the mod flag on the Noctua landing page, may revert some unsaved changes from @BarbaraCzub and @LiNiMGI ). I'll be unavailable for the next week, then available sporadically the week after that.

ftwkoopmans commented 7 years ago

The ORCHIDs are coming, for each annotation we will provide the ORCHID of each contributing annotator, please bear with us as we sort this out throughout the consortium.

kltm commented 7 years ago

@ftwkoopmans No problem at all. I'm going to put out what is currently in this PR tonight and can do another one when I am back online in a week or so.

kltm commented 7 years ago

@dosumis Merged and now in production.

kltm commented 7 years ago

@cmungall As an aside, it takes about 10m to rebuild the journal. I'm pretty happy about that for now.