RTXteam / RTX-KG2

Build system for the RTX-KG2 biomedical knowledge graph, part of the ARAX reasoning system (https://github.com/RTXTeam/RTX)
MIT License
39 stars 8 forks source link

What version of Biolink are we planning to use in the next build? #301

Open ecwood opened 1 year ago

ecwood commented 1 year ago

@acevedol @saramsey

Spun off of https://github.com/RTXteam/RTX-KG2/issues/300#issuecomment-1608056600, I wanted to make an issue specifically for discussing what version of Biolink we will be using next build. It looks like we've used Biolink v3.1.2 for a while (https://github.com/RTXteam/RTX-KG2/blob/issue-263/kg2-versions.md). With the changes desired from #286, will we continue using this version of Biolink? If not, what version are we planning to use?

saramsey commented 1 year ago

Tagging @acevedol what are your thoughts on this?

My feeling is we should be careful not to (in our enthusiasm to closely track the evolving Biolink standard) select a Biolink version that would be so different from 3.1.2 that it would significantly slow down getting a KG2.8.4pre build. On the other hand, if the changes are minor in the newer version, it could be useful to update.

saramsey commented 1 year ago

These are the releases to choose from, evidently:

https://github.com/biolink/biolink-model/releases

edeutsch commented 1 year ago

Breaking news from this morning's Architecture call, we will be expected to be compliant with Biolink 3.5.0 by July 7, heh..

image

https://docs.google.com/spreadsheets/d/1LEV3FyXBt3xm4bC5TQkVbrZalrr6OV2LnPtuLj1RVB4/edit#gid=1034097652

ecwood commented 1 year ago

How will we be compliant with Biolink 3.5.0 by July 7 when it hasn't been released yet? Apparently, it is being released on June 30th.

Do we have an idea of what is changing in it?

We can try for the 7th, but I think that will be tough, especially since Tuesday is a holiday next week and builds take a bit.

edeutsch commented 1 year ago

My opinion: based on the experience of the last few years, it is Translator custom to set somewhat unrealistic deadlines, which we (ARAX/KG2) meet about 50% of the time and almost no one else does, and no one really enforces any deadlines. So I would just proceed with implementing 3.5.0 at best comfortable speed (considering that it isn't even released yet).