biolink-model
biolink-model copied to clipboard
BioLink Model documentation does not give context to prefixes nor CURIEs
The documentation page for Gene (and other documentation pages) refer to prefixes and CURIEs without referencing what registry/prefix map these come from, allowing for ambiguity about what each one means.
Specifically about prefixes:
![Screenshot 2023-03-09 at 17 36 03 1](https://user-images.githubusercontent.com/5069736/224091319-d7ff368e-1445-4838-ac9d-42e57e1307f6.png)
Later in the document, with CURIEs
![Screenshot 2023-03-09 at 17 36 57](https://user-images.githubusercontent.com/5069736/224091405-d1a69997-2230-42fd-bac5-c3fe505d2fca.png)
While I personally know what a lot of these prefixes are, there are still several questions:
- What's the difference between
WB
andWormbase
? - What are
aspgd
anddcid
? These don't appear in the Bioregistry.
There should be external references to ensure there is zero ambiguity. I'd be happy to help add any missing prefixes to the Bioregistry to facilitate this.
The prefix expansions are provided with the model; they can be found here: https://github.com/biolink/biolink-model/blob/master/prefix-map/biolink-model-prefix-map.json
Thanks @balhoff. We're actually using exactly that document as a basis for alignment between BioLink and the Bioregistry. The remaining prefixes we haven't yet been able to contextualize or otherwise handle are listed in https://github.com/biopragmatics/bioregistry/blob/main/src/bioregistry/data/external/biolink/curation.tsv. Note that this contains a lot of references to hash IRIs in OBO PURL space