[Bioperl-l] Re: Bio::DasI

[Lincoln Stein]

There's this interesting and unecessary dichotomy between the entry-based 
view of Bio::SeqI, which began life as a way to capture EMBL entries, and the 
landmark-based view of DasI, which began life as a way to handle annotated 
genomes.  It might be that the right way to think about this is "virtual 
entries", which combines concepts from the DasI segment and the Bio::SeqI 
sequence.

Lincoln

On Tuesday 21 May 2002 11:34, Jason Stajich wrote:
> Does it make sense to think about a new Bio::SeqI implementation which
> implements the DasI interface by storing features in a Collection rather
> than as the current hash/array.
>
> All of the DAS interface methods seem useful to integrating into Sequence
> objects at some point, how about I get it working in an independent
> implementation and then we see if we should consider building a new Seq
> object that has all of this functionality.
>
> I'm not sure the Collection object by itself can handle these requests.
> I think that if the features were created on top of a sequence object then
> they will have references to the sequence object and can then pull in the
> underlying seq.
>
> -j
>
> On Tue, 21 May 2002, Lincoln Stein wrote:
> > Hi Jason,
> >
> > I've just checked in revisions to Bio::DasI and Bio::Das::SegmentI.
> > These provide the interface methods needed to run generic genome
> > browser on top of a feature collection.  I don't think the interface
> > is particularly onerous to implement and I'm really excited by the
> > prospect of being able to run gbrowse on top of your new feature
> > Collection classes.
> >
> > Despite the name, DasI is not quite at the point where it encapsulates
> > all functionality from the Bio::Das module.  What's missing is calls
> > to fetch DNA and stylesheets.
> >
> > Lincoln