[Biocorba-l] SeqFeature -> get_Primary_Seq

Matthew Pocock mrp@sanger.ac.uk
Wed, 07 Feb 2001 11:56:11 +0000


Hi.

I think you should be able to go from a feature to the entire rich 
sequence that contains it and also to some simple 'stringy' sequence 
that is contained within its location (no ID, name etc.). This should be 
possible to implement for all cases. For things like CDSs, they should 
extend the IDL to provide a 3rd method to retrieve the translation 
(which is the simple 'stringy' sequence, but may be castable to the 
fully annotated sequenced). Does that sound workable?

Matthew

Alan Robinson wrote:

> Ewan wrote:
> 
> 
>>> On Tue, Feb 06, 2001 at 10:39:52AM -0500, Jason Stajich wrote:
>>> 
>>>> In the IDL for the object SeqFeature, the method get_Primary_seq I have
>>>> the follow question.
>>>> 
>>>> Is the intention to return the PrimarySeq that is bounded by the
>>>> start/end of the SeqFeature (as we do in bioperl) or the entire PrimarySeq
>>>> that the Feature lies on?
>>> 
>> if we have to go one way on this, I would go for the entire sequence
>> coming back.
>> 
>> As it is a PrimarySeq, it does not have seqfeatures. This does mean that
>> we don't have to haul around the entire seqfeatures set of a sequence as a
>> consequence of one seqfeature being asked for. Which is good.
>> 
>> 
>> BTW - I would like the
>> 
>>    interface SeqFeature {
>> 	AnonymousSeq just_the_region_seq;
>> 	PrimarySeq   the_entire_seq;
>> 	};
>> 
>> type route. This is a good use of an "identifier-less" sequence...
> 
> 
> 
> Hmmmm... Initially, I'm not keen on this suggestion for the PrimarySeq
> behaviour.
> 
> If I'm getting a sequence from a 'SeqFeature' object, I expect it to
> relate to the SeqFeature in question, not the original sequence (I'll use
> the methods of the 'Seq' interface for that).
> 
> However, I agree there may be a case for adding a 'get_AnonymousSeq()'
> method to the 'SeqFeature' interface.
> 
> 
> Alan.
> 
> _______________________________________________
> Biocorba-l mailing list
> Biocorba-l@biocorba.org
> http://www.biocorba.org/mailman/listinfo/biocorba-l