[Bioperl-l] search2gff

[Jason Stajich]

On Fri, 2 Jul 2004, Neil Saunders wrote:

> I have a couple of queries regarding the search2gff script.
>
> 1) I'm using the latest CVS (1.8) which states "search2gff does the
> right thing now - now strand is done right properly".  Am I right in
> thinking that now target start is always less than target end?
>
exactly.

> 2) Should search2gff work with any BLAST flavour?
>
> I ask because I'm running it on tblastx files.  Without the -m switch
> it seems to run fine, but if I add -m, I get thrown exceptions looking
> like e.g. :
>
> ------------- EXCEPTION: Bio::Root::Exception -------------
> MSG: Undefined sub-sequence (2827710,2827711). Valid range = 2827710 -
> 2827889
>
> The undefined subsequence start/ends always differ by 1.
>

Can you send the whole stack trace and/or better an example blast file.  I
guess ideally this should be submitted as a bug.  I don't see any calls to
matches in the code so I am not sure why this is getting triggered.
I changed the script again to be a little simpler so might pull from
latest CVS again.


> thanks,
> Neil
>

--
Jason Stajich
Duke University
jason at cgt.mc.duke.edu