[Bioperl-l] Proposal: SemanticMapping and call for info on Gene Objects

Jason Stajich jason@cgt.mc.duke.edu
Tue, 14 May 2002 15:15:38 -0400 (EDT)


Name is certainly up for debate - FeatureAggregator seems fine to me.

I too think that it would be useful to have things chained together and
also not pushing all the logic into one object.  Whether this means a
"master"  aggregator holder needs to be created which has an ordered list
of handles to the specific aggregators or if we just do this within a
SeqIO object or else one will have to call each aggreator explicitly in
code that will work on a list of SeqFeatures?

For now I'm going to aggregate the bioperl -> generic object code within
the FeatureAggregator object because I think this may need to be done
in a specific form depending on the use of the objects or the output
format they will be delievered to.

-j

On Tue, 14 May 2002, Lincoln Stein wrote:

> I love the idea but am not happy about the "SemanticMapper" name.  In
> the Bio::DB::GFF module these things are called Aggregators, and it
> would be very nice if we could maintain consistency between the
> names.  How about FeatureAggregator?  This reflects that the job of
> the class is to aggregate individual Bio::SeqFeatureI objects into
> larger objects.  Or maybe FeatureBuilder would be better...
>
> Lincoln


-- 
Jason Stajich
Duke University
jason at cgt.mc.duke.edu