[Bioperl-l] gff -> gene, or exonerate ->gene?

[Charles Hauser]

Hi,

I am trying to see if exonerate faithfully predicts splice sites for the
organism we work with.  So, I'd like to construct genes (exons only)
from GFF files and compare the translations of these to 'known' gene
products.

I have not found a way yet to assemble exons usingBio::Tools::GFF? 
Are there better approaches?


Charles 

my $gffio = Bio::Tools::GFF->new(
				 -file => $infile,
				 -gff_version => 2);

my $feature;
while($feature = $gffio->next_feature()) {
** -> assemble exons
}
$gffio->close();


##gff-version 2
##source-version exonerate:est2genome 0.6.7
##date 2003-03-20
##type DNA
#
#
# seqname source feature start end score strand frame attributes
#
scaffold_17	exonerate:est2genome	gene	63923	66012	4300	+	.	gene_id	1	;	sequence	20021010.141.1.rev	;	gene_orientation	+	;
scaffold_17	exonerate:est2genome	exon	63923	64093	4300	+	.	insertions	0	;	deletions	0	;
scaffold_17	exonerate:est2genome	splice5	64093	64095	4300	+	.	intron_id	1	;	splice_site	"GT"	;
scaffold_17	exonerate:est2genome	intron	64157	64157	4300	+	.	intron_id	1	;
....
# --- END OF GFF DUMP ---