Search Mailing List Archives
[MISC] RE: [GO] Re: [MOBY] Re: GO Xref Abbs now served by MOBY
suzi at fruitfly.org
Thu Feb 26 12:09:04 PST 2004
I agree that the LSID would be the way to go (no sense knocking
heads if we can avoid it). But I think many people on the list may
not be all that familiar with what the ACRONYM "LSID" means
in concrete terms. Would you send out a lay-person's description.
I think it really is a simple thing that looks gussied and complex
because it has an acronym.
Mark Wilkinson wrote:
>This actually brings up another point - a point that "smells" the same
>as the point I made last week w.r.t. evidence codes.
>At the moment, the Xref Abbs document has, e.g. "taxon" as a
>cross-reference abbreviation for NCBI's taxonomy identifiers. This then
>precludes anyone else from using "taxon" as their namespace, even if
>their database contains taxon identifiers (that may or may not be
>identical to those of NCBI). Sure, I agree that we should probably all
>use the "taxon" identifiers as they are defined by NCBI, since theirs is
>the ~canonical list... but what if someone suddenly registers "GeneID"
>as a Xref Abb, with the intention that it means an identifier in the
>potatoe genome database, or as in the message below!?! From now on,
>nobody will be allowed to use 'GeneID' as their namespace/Xref
>abbreviation, so they will have to prefix it with e.g. TAIR_GeneID.
>That's fine and dandy, and in fact, at the computational level we don't
>much care so long as there are no collisions... but it seems that the
>Xref Abbs list was made without consideration of the scale that it might
>grow to. It would be nice to agree on a format for declaring these
>abbreviations that ensures that we ENSURE that we avoid collisions. The
>most *obvious* and interoperable way to do this is to adopt the LSID
>naming system, but that seems to be meeting with some resistance from
>the other list members. An alternative would be to either insist that
>the Xref Abbs are prefixed with some vendor-specific tag (e.g.
>NCBI_GeneID), or with their domain (ncbi.nlm.nih.gov:GeneID). This
>would allow multiple groups who have genes in their databases (i.e. most
>of us!) to use the very generic "GeneID" namespace without stepping on
>each others toes.
>...just a thought...
>On Thu, 2004-02-26 at 10:58, Maglott, Donna (NIH/NLM/NCBI) wrote:
>>Although a date has not been set, NCBI plans to replace LocusLink with
>>The LocusID/URL is still valid:
>>database: NCBI LocusLink ID.
>>But this also works and will apply to more genomes:
>>database: NCBI Entrez Gene.
>>NOTE: The value of the LocusID is conserved as the value of the GeneID.
>>From: Mark Wilkinson [mailto:markw at illuminae.com]
>>Sent: Thursday, February 26, 2004 11:19 AM
>>To: Michael Ashburner (Genetics)
>>Subject: Re: [GO] Re: [MOBY] Re: GO Xref Abbs now served by MOBY
>>just click the link above to instantiate the cat... whether it is smiling or
>>not will depend on the parse-test :-)
>>Attached is a saved version of the output from the script above as of this
>>On Thu, 2004-02-26 at 07:57, Michael Ashburner (Genetics) wrote:
>>>Then can you instantiate the cat & send it (|her|him) to me. Keep the
>>>smile on its (|hers|his) face please. M
>>>This message is from the GOFriends moderated mailing list. A list of
>>>public announcements and discussion of the Gene Ontology (GO) project.
>>>Problems with the list? E-mail: owner-gofriends at geneontology.org
>>>Subscribing send "subscribe" to gofriends-request at geneontology.org
>>>Unsubscribing send "unsubscribe" to gofriends-request at geneontology.org
This message is from the GOFriends moderated mailing list. A list of public
announcements and discussion of the Gene Ontology (GO) project.
Problems with the list? E-mail: owner-gofriends at geneontology.org
Subscribing send "subscribe" to gofriends-request at geneontology.org
Unsubscribing send "unsubscribe" to gofriends-request at geneontology.org
More information about the go-friends