[Bioperl-l] Bio::SeqFeature::GeneStructure and Prediction

[Hilmar Lapp]

James Gilbert wrote:
> 
> Hilmar,
> 
> I'd like to see the Ensembl Gene structure in
> BioPerl, because I think it is more or less
> correct.  In Ensembl a Gene has one or many
> Transcripts, each of which can have a Translation,
> and each Transcript has one or many Exons.
> Transcripts share exons, and so you can get a
> non-redundant list of exons from a gene.  So I'd
> like to see a Transcript layer in there somewhere.
> 

I've already explained that to Ewan yesterday. A transcript is beyond the
scope of this class. The GeneStructure is merely a representation of a gene
structure or its elements as a feature with subfeatures, not more, not less.
In particular, it is _not_ a Gene. It represents data, but it doesn't
interprete them (mostly), and it is fairly independent from an underlying
sequence object. One could add transcript features as sets of exon features
though, but these will not be computed (how would you know?).

A gene with transcripts would actually go into its separate class, as I think
of it. So, there could be a Bio::Gene which, given a Sequence and a
GeneStructure, may provide transcripts and information about alternative
splicing.

Maybe this needs more thinking.

	Hilmar

-- 
-----------------------------------------------------------------------
Hilmar Lapp                                      email: hlapp@gmx.net
NFI Vienna, IFD/Bioinformatics                   phone: +43 1 86634 631
A-1235 Vienna                                      fax: +43 1 86634 727
-----------------------------------------------------------------------