[Gmod-schema] Re: unique dbxref_id on cvterm

[Chris M. <cj...@fr...>]

On Mon, 21 Feb 2005, Scott Cain wrote:

> On Mon, 2005-02-21 at 19:49 -0800, Chris Mungall wrote:
> >
> > On Mon, 21 Feb 2005, Scott Cain wrote:
> >
> > > On Mon, 2005-02-21 at 19:18 -0800, Chris Mungall wrote:
> > > > the dbxref must be persistent and consistent across database
> > > > instances
> > >
> > > Why?  By there very nature, they are going to be the sort of thing that
> > > are locally used and not global.  If they were more generally used, they
> > > should be on OBO.
> >
> > Not necessarily. If you look at most of the non-obo ontologies out there
> > typically available in OWL format, most conflate the name and ID
> >
> > > > what about making dbxref = cv.name + cvterm.name?
> > >
> > > Seems ugly.  So an accession of "Ad Hoc Ontology:synonym" is what you
> > > are suggesting?
> >
> > Eh?
> >
> > A cv.name should never be "Ad hoc" because there will be collisions
> > between cvs
>
> Fair enough.  Would you prefer 'local'?

Sorry, I was being cryptic. What I meant was, even these "ad hoc"
ontologies must have some kind of name that communicates the nature of the
cvterms within them. It seemed like you were planning on lumping all
ad-hoc ontologies together, which has a high likelihood of producing
collisions on cvterm unique keys.

It sounds like "Ad hoc:synonym" is actually from a cv of property types
that can be attached using featureprop et al. This is very definitely not
an ad-hoc ontology, it's crucial that featureprop types have their own cv
and are defined

> > > >
> > > > On Mon, 21 Feb 2005, Scott Cain wrote:
> > > >
> > > > > Hi Chris,
> > > > >
> > > > > I am dreading writing this email a little bit, given the discussion that
> > > > > occurred a few weeks ago involving dbxrefs and cvterms.  Anyway, I
> > > > > gather from the comments you added to cvterm that you really did intend
> > > > > to make a unique index on dbxref_id in cvterm, and I even understand why
> > > > > you did it given that much of the time, the cvs are coming from a real
> > > > > ontology and they have accessions.  Of course, some of the time, there
> > > > > will be "ad hoc" cvs that won't have accessions.  The solution I am
> > > > > suggesting is the creation of a db sequence and items that don't belong
> > > > > to a formal ontology will get the next available value from the
> > > > > sequence.  Does that sound OK to you?
> > > > >
> > > > > Thanks,
> > > > > Scott
> > > > >
> > > > >
> > > >
> > >
> >
>