Open GoogleCodeExporter opened 9 years ago
Considering we are trying to get towards meaningless IDs, what is the interest
of using
http://purl.obolibrary.org/obo/PR_P04637
vs
http://purl.obolibrary.org/obo/PR_xxxxxxx and an annotation property?
Original comment by mcour...@gmail.com
on 16 May 2013 at 6:02
Long story. A short answer would be to call Cathy Wu and talk to her about it.
A longer answer would be better provided as discussion. I might be able to send
you the complete thread.
Original comment by alanruttenberg@gmail.com
on 16 May 2013 at 6:04
[deleted comment]
These are UniProt accessions
Original comment by cmung...@gmail.com
on 16 May 2013 at 6:07
If by PR_xxxxxxxx you mean IDs of the form currently used by PRO (nine-digit
numeric), then the interest is in not having to mint new IDs for every existing
identifier in the external database.
Original comment by d...@georgetown.edu
on 16 May 2013 at 6:08
Alan is sending out proposed update to obo-discuss policy with 2 weeks review,
CHris is checking on what may need to be updated on the apache conf side
Original comment by mcour...@gmail.com
on 7 Jun 2013 at 5:09
See
http://code.google.com/p/obo-foundry-operations-committee/issues/detail?id=88
Original comment by mcour...@gmail.com
on 7 Jun 2013 at 5:09
Assigning to Alan.
Original comment by mcour...@gmail.com
on 21 Jun 2013 at 11:00
update June 28: during call, Alan mentioned he'll take care of sending out the
proposal
Original comment by mcour...@gmail.com
on 28 Jun 2013 at 4:44
PRO wants to use isoform ids as well, which add a "-"\d+ at the end. We need to
see whether this is compatible with OBO ids before approving this. Chris?
Original comment by alanruttenberg@gmail.com
on 22 Jul 2013 at 3:22
In response to comment #10, Chris has responded that this causes no troubles on
his side.
Original comment by alanruttenberg@gmail.com
on 10 Sep 2013 at 4:13
I have made a proposed change to the google drive version of the document at
http://goo.gl/elqRql
Text below, but better to look at the document in google drive. Formatting may
have been lost.
Exceptions modifying the productions above
In addition to these OBO format ids, a decison was made to allow for cases
where a non-ontology resource, such as a database, has an ontology created to
adapt the resource into ontology form. Requests for exceptions to the standard
format should be made to the OBO Operations Technical Group by submitting a
ticket at http://code.google.com/p/obo-foundry-operations-committee. On
approval this document will be augmented to add the allowable formats.
Issue 81 proposed that a variant of UniProt Accessions be used for the Protein
Ontology in their adaptation of certain UniProtKB entries that correspond to
their “organism-gene” and isoform category. These accessions can be
specified as followed:
LOCALID_PR_UNIPROT ::= [A-Z][0-9]+("-"[0-9]+)?
PR_OBO_IDENTIFIER ::= "PR:" LOCALID_PR_UNIPROT
We amend the OBO_IDENTIFIER production to be
OBO_IDENTIFIER ::= (IDSPACE ":" LOCALID) | PR_OBO_IDENTIFIER
Format of Foundry-compliant URIs
FOUNDRY_OBO_URI ::= "http://purl.obolibrary.org/obo/"
((IDSPACE "_" LOCALID) |
PR_OBO_IDENTIFIER
)
Original comment by alanruttenberg@gmail.com
on 7 Oct 2013 at 4:13
eh.
should be
...
FOUNDRY_OBO_URI ::= "http://purl.obolibrary.org/obo/"
((IDSPACE "_" LOCALID) |
("PR_" LOCALID_PR_UNIPROT)
)
Original comment by alanruttenberg@gmail.com
on 7 Oct 2013 at 4:21
Per discussions with Darren, updated the regex to
LOCALID_PR_UNIPROT ::= [A-Z][A-Z0-9]{5,6}+("-"[0-9]+)?
Original comment by alanruttenberg@gmail.com
on 27 Oct 2013 at 5:33
Original issue reported on code.google.com by
alanruttenberg@gmail.com
on 16 May 2013 at 5:57