[Bioperl-l] using Bio::DB::GFF for aggregation

[Dan Kortschak]

Hi Chris (or others),

I've been looking at ways to do large assemblies (really rnaseq/readseq
comparisons for coverage) with maq/bowtie output and it's clear that for
the size of project that I'm working on the space complexity is too
nasty with Bio::DB::Sam. So I thought Bio::DB:GFF might be the way to
go.

I was thinking: B:T:R:Bowtie ~> B:SeqFeat:Generic -> B:T:GFF -> B:DB:GFF

This depends on the behaviour of B:DB:GFF->features(-merge=>1). I've
read through the docs, and it's not entirely clear (I'm hoping I've
interpreted it the right way), but does this result in the return of
features such that overlapping features are returned as a single feature
while non-overlapping features come back separately. If this is the
case, it would satisfy my requirements perfectly.

thanks for your time
Dan