WormBase prefix expansion

[balhoff]

Biolink has this expansion for Wormbase:

"WormBase": "https://www.wormbase.org/get?name=",

GO is using http://identifiers.org/wormbase/ in the NEO ontology (and so in GO-CAMs). It would be good to align these. @cmungall which is preferred?

[RichardBruskiewich]

@balhoff @cmungall...

Actually, the currently registered URI from identifiers.org for wormbase is: https://identifiers.org/wb/.

This resolves properly to the wormbase site, e.g. try https://identifiers.org/wb/WBGene00000001 whereas the GO version http://identifiers.org/wormbase/WBGene00000001 doesn't work (anymore.. perhaps it did years ago?).

The current version in the Biolink Model context.jsonld (https://www.wormbase.org/get?name=WBGene00000001) also appears to resolve but seems to originate from one of the default_curi_maps since it is not listed under the model yaml prefixes section.

[RichardBruskiewich]

I'm going to set the WormBase: prefix to the identifiers.org resolved URI, https://identifiers.org/wb/, OK?

We don't have much influence over what other external resources are using (e.g. GO). Somebody else will need to poke them to update their prefix to URI mapping. BTW, this includes the Monarch default_curi_map entry, which is the source of the Wormbase URI in Biolink... I issued a PR prefixcommons/biocontext#32 for this change.

[cmungall]

none of the URIs we use should ever have '?'s or query parameters in them!

…

On Wed, Nov 18, 2020 at 10:35 AM Jim Balhoff ***@***.***> wrote: Biolink has this expansion for Wormbase: "WormBase": "https://www.wormbase.org/get?name=", GO is using http://identifiers.org/wormbase/ in the NEO ontology (and so in GO-CAMs). It would be good to align these. @cmungall <https:/cmungall> which is preferred? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#530>, or unsubscribe <https:/notifications/unsubscribe-auth/AAAMMON6VBL4PJE4TO54AZDSQQHYBANCNFSM4T2K6WLA> .

[RichardBruskiewich]

none of the URIs we use should ever have '?'s or query parameters in them!

Yep. Totally agree @cmungall and @balhoff. However, since the Monarch PrefixCommons entry for WormBase had that deficiency, note that I've posted a PR request to fix the URI to the Identifiers.org version (see PR prefixcommons/biocontext#32).

[kshefchek]

My understanding is the default_curi_maps section of the biolink model works as a priority list, so the idot_context overrides the monarch context file, is this not the behavior we're seeing?

[deepakunni3]

@kshefchek Thanks for pointing that out. I wonder how the ordering is. Perhaps the idot_context gets overridden by monarch context.

Will have to look at how biolinkml parses the default_curi_maps directive

[cmungall]

order should be order of priority what sometimes happens is that we end up with 2 entries rather than a de-dedupe, e.g. if different case is used (this is a prefixcommons bug, clobbering should be case insensitive), or a different prefix. Ideally we'd have warnings (e.g. "prefixes include FB and FlyBase but these are the same resource")

…

On Wed, Nov 25, 2020 at 10:02 AM Deepak ***@***.***> wrote: @kshefchek <https:/kshefchek> Thanks for pointing that out. I wonder how the ordering is. Perhaps the idot_context gets overridden by monarch context. Will have to look at how biolinkml parses the default_curi_maps directive — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#530 (comment)>, or unsubscribe <https:/notifications/unsubscribe-auth/AAAMMOLTRKR3BCWZSI6BO7LSRVBEXANCNFSM4T2K6WLA> .

[RichardBruskiewich]

@cmungall, @kshefchek @deepakunni3

I certainly stirred a tempest in a teapot here!

I issued the correct PR at monarch-initiative/dipper#1000
I cancelled the PR at prefixcommons/biocontext#32
I removed the WormBase prefix under prefixes of the Biolink YAML in this master commit: a525c53