[Bioperl-l] Re: Bio::DB::GFF start/end coordinates

Lincoln Stein lstein at cshl.edu
Mon Aug 29 17:43:30 EDT 2005

Hi Jason,

You've got to set $db->absolute(1) to get true Bioperl-compliant coordinates. 
The reason for this is because of Bio::DB::GFF's (perhaps regrettable) use of 
relative coordinate addressing by default. This is explicitly mentioned in 
the documentation under a section named (something like) "BioPerl 


On Thursday 25 August 2005 01:00 pm, Jason Stajich wrote:
> Lincoln -
> One bug I'm still seeing in Bio::DB::GFF::Feature objects is start/
> end are still returning start > end when strand < 0.   I know this is
> different expectation for Bioperl / Gbrowse but this causes a little
> problems, especially when you get an aggregated feature out from
> Bio::DB:GFF and then write it to a genbank file.  The locations looks
> like this:
> complement(join(1031..975,676..501))
> My workaround is just to create new Location objects and features
> from the Bio::DB::GFF obtained objects  (some of these aren't
> allowing write-back to overwrite the values).
> Note on a slightly separate topic:
>   I have patched my Bio::Location::Split to_FTstring to simplify the
> string, current behavior would be to output the location like this:
> join(complement(1031..975),complement(676..501),))
> I'm seeing about how applying the patch, I'm not sure whether or not
> it perfectly works.
> -jason
> --
> Jason Stajich
> jason.stajich at duke.edu
> http://www.duke.edu/~jes12/

Lincoln D. Stein
Cold Spring Harbor Laboratory
1 Bungtown Road
Cold Spring Harbor, NY 11724
SANDRA MICHELSEN, AT michelse at cshl.edu

More information about the Bioperl-l mailing list