[Biocorba-l] BSANE and bioCORBA

[Ewan Birney]

On Sun, 3 Jun 2001, Alan Robinson wrote:

> 
> Btw, since I (and others) are on the Biocorba list, there's really no need
> to include specific email addresses when replying.
> 
> My thoughts on Seq's and Features:
> 
>  1) We need a method to navigate from feature to sequence. But what should
>     this return? [And should we allow for mutable sequences?]
> 
> 	a) The ID of the parent (but which ID?: display, primary, or
> 	   accession)
> 	b) An object reference to the parent Seq object
> 
> 	I prefer 1a using primary id (since normally primary_id ==
> 	accession_number, and it maybe allows for some degree of sequence
> 	mutability).


I am firmly in the (b) camp. Often the subroutine which gets a sequence
feature has no way of getting a database object in which to use (a). I
vote with all my fingers and all my toes for objects coming back.

> 
> 
>  2) A convenience method in a feature to return the sequence would be
>     'convenient'. But what should this return?
> 
> 	a) The sequence as a string.
> 	b) An 'AnonymousSeq' or even 'PrimarySeq' object.
> 
> 	I initially prefer 2a with semantics as outlined by Ewan and also 
> 	used by the start() and end() functions in SeqFeature; but I could
> 	be swung to 2b - it might be more work for the server writer, but
> 	would it be more useful for the client? (Then again, perhaps it's
> 	better to keep SeqFeature simple and have a factory method defined 
> 	elsewhere that may be handed a Seq and Location objects and
> 	returns a new PrimarySeq?)
> 

Ambivelent about this. Marginal leanings towards (a) as I find (b) - which
is what we have in Bioperl - just gets in my way ;)



> 
> Alan.
> 
> 
> On Sun, 3 Jun 2001, Ewan Birney wrote:
> 
> > Juha Muilu wrote:
> > 
> > > What kind of use cases there are for having the method and also the
> > > seq_primary_id? I have assumed only navigation  from Sequence to feature 
> > > is necessary.
> > 
> > Going from feature back to the sequence is very common and in general the
> > function which is doing this only has the feature and does not have the
> > original sequence.
> > 
> > definite mistake to leave it out  :)
> > 
> > 
> > I am cool with alan's interpretation of primary_seq. I'd like the
> > convience function as well but this does lead to the bugbear of how the
> > convience function deals with composites and fuzzies ... probably best to
> > have a rigorous definition (composites spliced and fuzziesgiven back as
> > maximal hard numbers?
> > 
> > <aside>I HATE fuzzies . useless things. useless... useless</aside>
> 
> --
> ============================================================
> Alan J. Robinson, D.Phil.             Tel:+44-(0)1223 494444
> European Bioinformatics Institute     Fax:+44-(0)1223 494468
> EMBL Outstation - Hinxton             Email:  alan@ebi.ac.uk
> Wellcome Trust Genome Campus
> Hinxton, Cambridge
> CB10 1SD, UK                http://industry.ebi.ac.uk/~alan/
> ============================================================
> 
> _______________________________________________
> Biocorba-l mailing list
> Biocorba-l@biocorba.org
> http://www.biocorba.org/mailman/listinfo/biocorba-l
> 

-----------------------------------------------------------------
Ewan Birney. Mobile: +44 (0)7970 151230, Work: +44 1223 494420
<birney@ebi.ac.uk>. 
-----------------------------------------------------------------