[Bioperl-l] search2gff.PLS

[Hilmar Lapp]

I was looking for an existing tool to convert a SearchIO report 
(multi-report BLASTN in my case) to GFF3 and found 
scripts/utilities/search2gff.PLS.

Is this the tool I'm looking for or did miss a better suited one 
elsewhere in either bioperl or gbrowse/gmod?

If this is the tool, why is it suitable only for protein matches? (The 
doc says so. I used it for BLASTN report and didn't find any 
nucleotide-related flaws.) Or is this old documentation that should be 
fixed?

Also, the container match feature comes out with start and end being 
equal. I'm not sure whether I was doing something wrong, but the 
following fixes this.

146,148c149,151
<           $max{$type} = $proxyfor->start unless defined $max{$type} 
&& $max{$type} > $proxyfor->end;
<           $min{$other} = $otherf->start unless defined $min{$type} && 
$min{$type} < $otherf->start;
<           $max{$other} = $otherf->start unless defined $max{$type} && 
$max{$type} > $otherf->end;
---
 >           $max{$type} = $proxyfor->end unless defined $max{$type} && 
$max{$type} > $proxyfor->end;
 >           $min{$other} = $otherf->start unless defined $min{$other} 
&& $min{$other} < $otherf->start;
 >           $max{$other} = $otherf->end unless defined $max{$other} && 
$max{$other} > $otherf->end;

Since I don't understand 100% what should happen for a correct GFF3 
match container, I just wanted to make sure that this is indeed a fix 
and not introducing a bug before I commit it. Was anybody using this 
tool with the -m option?

	-hilmar
-- 
-------------------------------------------------------------
Hilmar Lapp                            email: lapp at gnf.org
GNF, San Diego, Ca. 92121              phone: +1-858-812-1757
-------------------------------------------------------------