[Bioperl-l] Unigene proposal and basic implementation

[Lincoln Stein]

If this is to be modeled after the Bio::Seq interface, then there should be 
one interface that models the static objects -- the clusters themselves and 
the various operations one can perform on clusters -- and another that is a 
factory of cluster objects, e.g. the Bio::ClusterIO class.

So the hierarchy I see would be:

	Bio::ClusterI
	Bio::Cluster
	Bio::Cluster::UniGene  # MIGHT NOT BE NECESSARY
	Bio::ClusterIO
	Bio::ClusterIO::unigene

Lincoln

On Tuesday 16 April 2002 05:54, Ewan Birney wrote:
> On Tue, 16 Apr 2002 andrew@anatomy.otago.ac.nz wrote:
> > Hi Ewan,
> >
> > >I certainly prefer this to the straightforward Unigene one. Let's do it
> > >this way.
> >
> > Sounds good.
> >
> > >What the difference between Bio::Cluster::UniGene and
> > >
> > >Bio::Cluster::UniGene::unigene.pm?
> >
> > I've modeled this on SeqIO and SeqIO/fasta.pm. So I''ve made the
> > Bio::Cluster::UniGene::unigene.pm the format module for
> > Bio::Cluster::UniGene.pm.
>
> Can we make the directory Bio::Cluster::ClusterIO therefore?
>
> > I thought this might be a good way to go because there are a few other
> > files that Unigene has on offer that I'll look into in due course, that
> > may require different format modules. This will  need me to tidy it up a
> > bit. I'll do the same sort of format guessing thing that I think SeqIO
> > does.
> >
> > - Andrew.
>
> -----------------------------------------------------------------
> Ewan Birney. Mobile: +44 (0)7970 151230, Work: +44 1223 494420
> <birney@ebi.ac.uk>.
> -----------------------------------------------------------------
>
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l@bioperl.org
> http://bioperl.org/mailman/listinfo/bioperl-l