[Bioperl-l] Unigene

[Ewan Birney]

On Wed, 9 May 2001, Elia Stupka wrote:

> dear all,
> 
> had very little time to get together a proper proposal, so I will post a
> vague one here. I am trying to parse Unigene entries. It seems to me that
> there is little point in making a Bio::Seq co,pliant object for unigene
> entries since they are a Boentry object which contains many Seq objects.
> The SeqContainerI or SeqClusterI would be hte interface. Unigene would
> inherit from this interface. If people think that there could be lots of
> other SeqCluster objects in other scenarios, we could write
> Bio::SeqCluster and make Bio::Unigene inherit from Bio::SeqCluster. If not
> for now I will write the interface and a Unigene object. 

Hmmmm....

This is probably the right route (I have thought and rejects Bio::DB::SeqI
type interface - sort of ok but implies too much and Bio::SimpleAlign is
definitely *wrong*)

I guess Bio::SeqContainerI would be my preferred route with three methods
defined:

	->all_Seq()  # returns an array of sequence objects
	->iterator() # returns a SeqIO compliant object
                     # ie, on which ->next_seq works
        ->number()   # gives back the number of sequences


Interesting question - 

should Bio::SimpleAlign inheriet from Bio::SeqContainerI?


> 
> Hte other thing is that SeqCluster will need its own IO system, which in
> teh case of unigene needs to take in two files, the unigene cluster
> information and the unigene sequences (which can be read  with SeqIO).. I
> have started playing with this idea, and probably on my next plane trip on
> Sunday I will implement the rest of it.  Commens very very welcome,
> because as you see this is not one of those nice mails with bullets and
> precise definitions ;)
> 

How many other things fit into the cluster definition which are not
databases, alignments or SeqIO streams?



> Elia
> 
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l@bioperl.org
> http://bioperl.org/mailman/listinfo/bioperl-l
> 

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