[Bioperl-l] Bio::Annotation::SimpleValue, keywords()

[Hilmar Lapp]

On Tuesday, September 24, 2002, at 09:11 AM, Ewan Birney wrote:

> On Mon, 23 Sep 2002, Hilmar Lapp wrote:
>
>> Two ideas/proposals in one:
>>
>> 1) RichSeqI::keywords is currently implemented as keeping the string
>> as one, no splitting into individual keywords. In a database (yeah,
>> here's biosql) however, one would rather want to store them
>> separately I think. In a sense, keywords are an ontology. Also they
>> are annotation. I propose to stick keywords onto a sequence's
>> AnnotationCollection, using e.g. tag 'keyword' and class SimpleValue
>> (but see below).
>>
>> 2) Bio::Annotation::SimpleValue actually resembles an ontology term,
>> doesn't it? Well, not always probably (thinking descriptions and
>> alike). If SimpleValue is not an ontology term itself, then what
>> about adding Bio::Annotation::OntologyTerm, which would just adapt
>> Bio::AnnotationI to Bio::Ontology::TermI?
>>
>
> I like this idea - I would vape SimpleValue and claim that it 
> should be an
> ontology term. Or perhaps it *could be* an ontology term. Hmmm. (values
> being numbers?)

The conclusion I reached last night was that you may have annotation 
that is just a scalar but definitely not a controlled or standard 
vocabulary term. So, I decided to leave SimpleValue untouched and 
wrote up a Bio::Annotation::OntologyTerm that implements 
Bio::AnnotationI and Bio::Ontology::TermI (the latter by 
composition).

I stopped short of changing the swissprot parser to populate 
Bio::Annotation::OntologyTerm instances with gene names because I 
was too tired, and this morning I was outvoted to leave gene names 
as qualvals in biosql instead of making them OT associations, hence 
SimpleValue would work just fine ... How useful does it appear to 
make this change in the swissprot parser?

	-hilmar


>
>>
>> I'm thinking about how to best store gene names in biosql ... they
>> presently come as Bio::Annotation::SimpleValue, and I think we want
>> them end up in an ontology.
>>
>> 	-hilmar
>> --
>> -------------------------------------------------------------
>> Hilmar Lapp                            email: lapp at gnf.org
>> GNF, San Diego, Ca. 92121              phone: +1-858-812-1757
>> -------------------------------------------------------------
>>
>> _______________________________________________
>> Bioperl-l mailing list
>> Bioperl-l@bioperl.org
>> http://bioperl.org/mailman/listinfo/bioperl-l
>>
>
> -----------------------------------------------------------------
> Ewan Birney. Mobile: +44 (0)7970 151230, Work: +44 1223 494420
> <birney@ebi.ac.uk>.
> -----------------------------------------------------------------
>
>
--
-------------------------------------------------------------
Hilmar Lapp                            email: lapp at gnf.org
GNF, San Diego, Ca. 92121              phone: +1-858-812-1757
-------------------------------------------------------------