[Biopython-dev] [Bug 2677] BioSQL seqfeature enhancements

[bugzilla-daemon at portal.open-bio.org]

http://bugzilla.open-bio.org/show_bug.cgi?id=2677





------- Comment #5 from biopython-bugzilla at maubp.freeserve.co.uk  2008-11-21 05:52 EST -------
(In reply to comment #4)
> (In reply to comment #3)
> > (In reply to comment #0)
> > > Ive used the "Sequence Keys" ontology for the location operator and stored
> > > loc op in the location_qualifier_value table - not sure this is right...
> > >
> > 
> > I'm not sure off hand either, but would like us to check before committing
> > this.  In the short term, what ever BioPerl does is "right" as I'm treating
> > that as the BioSQL reference implementation.
> 
> I don't read Perl - but I grep'ed through the source and only found one ref to
> the location_qualifier_value, and that was in the docs. So maybe they don't
> store it there...
> 
> Sorry I can be of more help, C.
> 

I tried browsing and searching the BioPerl-db source, but couldn't find the
answer, so I tried the direct route and used their load_seqdatabase.pl script
to import a GenBank file (with at least one join location) and inspected the
tables.

The answer is that location.term_id is always left as NULL, so there is no
ontology to worry about.  Doing something sensible with ontologies (e.g.
support for existing strict ontologies like SO or SOFA) rather than the current
ad-hoc relaxed approach (adding new ontology terms on the fly) taken by BioPerl
and Biopython is a possible future enhancement.

I'm going to look at modifying you patch to leave location.term_id as NULL,
with the aim of committing that today and then doing the Biopython 1.49
release.


-- 
Configure bugmail: http://bugzilla.open-bio.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.