[Bioperl-l] question -- Bio::SeqFeature::Gene::Transcript

Hilmar Lapp hlapp at gmx.net
Sun Nov 4 12:42:13 EST 2007

Hi Stefanie,

sorry for taking so long to respond - your email got buried in a pile  
while I was away on travel. The Bio::SeqFeature::Gene::* modules were  
written mostly with the motivation to have a model that can represent  
the results of gene predictors.

GenBank AFAIK doesn't annotate introns explicitly, though they should  
be implicit from cDNA (or mRNA? or gene, as you say) features on  
genomic sequence. The Bioperl SeqIO parsers won't transform those  
into a Bio::SeqFeature::Gene-based model, but instead will yield just  
plain Bio::SeqFeatureI objects in a flat array. It's up to subsequent  
processing to build these into more hierarchical models.

I'm not sure whether someone's done this already for GenBank-type  
feature tables. There is a Unflattener that at least attempts to  
build a feature hierarchy from the flat array that's compliant with  
the Sequence Ontology (or so I recall).

I'm copying the list in case others have additional suggestions.


On Oct 25, 2007, at 3:40 AM, Stefanie Hartmann wrote:

> Hello Hilmar,
> I have a question about your bioperl module  
> Bio::SeqFeature::Gene::Transcript:
> I can't figure out how to generate the $gene object for use in this  
> line:
> @introns = $gene->introns();
> The data I'm working with is a local file in genbank format, and  
> I'm interested in extracting intron sequences (and maybe flanking  
> exons) for certain genes. I have been trying to get the introns via  
> the sequence features ('CDS' or 'gene'), but this has not been  
> working. Which approach will I have to take?
> I'd be very grateful if you could point me into the right direction!
> Hope things are going well in Durham! And thank you in advance!
> Stefanie

: Hilmar Lapp  -:-  Durham, NC  -:-  hlapp at gmx dot net :

More information about the Bioperl-l mailing list