#27 What is the prefix for the Orphanet disease ontology? I'm getting a 404

[Chris Mungall]

This query
http://www.ebi.ac.uk/miriam/main/search?query=orphanet

takes me here
http://www.ebi.ac.uk/miriam/main/collections/MIR:00000532;jsessionid=31C12755E07E667800C7D615B47FB45C

Which is currently a 404.

[Chris Mungall]

OK, easily fixed by removing the tail of the URL, this works
http://www.ebi.ac.uk/miriam/main/collections/MIR:00000532

So I guess you need to replace ; with ? in your URL construction.

[Chris Mungall]

Also, I'm confused about this pattern:

^Orphanet(_\d+|:C\d+)$

This means

1. Orphanet:C023 - valid
2. Orphanet77790 - valid
3. Orphanet:77790 - invalid

However, 2 is not accepted by OLS whereas 3 is.

I propose:

^Orphanet:(_\d+|C\d+)$

[Nick Juty]

Hi Chris

With regard to the query error you had, we are looking into it. Its a little bit sporadic in the sense that I am not getting any errors right now using it.

For Orphanet disease ontology, as far as I could see, identifiers for this collection are either of type 1 or 3 (of the examples you gave), ie Orphanet:C\d+ or Orphanet:\d+
(In fact the 'C' identifiers could be further specified as C\d{3})
The underscore is handled by OLS, which accepts : or _ interchangeably as far as I know, so we have included that. You are right though, the regex is a little off and I will modify it.

I just wanted to pick up on one thing. You state 'Orphanet77790 - valid' (your 2nd example).
Its not valid for the current regex we store, ^Orphanet(_\d+|:C\d+)$, since it has no underscore or colon. Also, its not accepted by OLS, and doesn't work through identifiers.org (http://identifiers.org/orphanet.ordo/Orphanet77790). Do you mean its actually a valid identifier? And if so can you send me a working link to it?

I'll change the regex to: ^Orphanet(_|:)C?\d+$
I think that covers everything, except maybe example 2 that you gave, if it wasn't an error....

Thanks for taking the time to point out our error!

cheers

Nick

[Chris Mungall]

My mistake, I somehow mentally skipped the underscore.

The new regex is better; it accepts what I believe to be all valid forms. Regarding the underscore - I believe this is confusion regarding OBO style purl construction and the construction of URI fragments given a standard OBO library base directive. This is bigger than this ticket...