[Bioperl-l] Putative bug in Bio/SearchIO/blast.pm

[Jason Stajich]

I'm pretty sure this is already fixed with the code in CVS.  Can you  
try it with that code -- the regexp was already corrected in  
connection to very large bitscores (Scott Markel's bug) - it is done  
slightly differently from your suggested fix as we need to handle sci- 
value in that field as well.

Also, please post bugs and diffs to http://bugzilla.open-bio.org so  
that they can be tracked - otherwise folks have to follow a mailing  
list thread to find out the resolution for a bug.

-jason
On Jun 9, 2005, at 8:58 AM, Jérémy JUST wrote:

>
>     Hello,
>
>   I think I've found a little bug in the Blast parser.
>   On that Blast result (BLASTN 2.2.9 [May-01-2004]):
>
> <<<<<<
>                                                                   
> Score    E
> Sequences producing significant alignments:                       
> (bits) Value
>
> gi|42592260|ref|NC_003070.5| Arabidopsis thaliana chromosome  
> 1, ...  75.8   8e-13
>
>
>> gi|42592260|ref|NC_003070.5| Arabidopsis thaliana chromosome 1,  
>> complete
>>
>                 sequence
>           Length = 30432563
>
>  Score = 75.8 bits (38), Expect = 8e-13
>  Identities = 104/126 (82%)
>  Strand = Plus / Minus
>
>>>>>>>
>>>>>>>
>
> the score is read as « 8 » instead of « 75.8 ».
>
>
>
>   I attach a tiny patch against Bioperl-1.5.0.
> ($Id: blast.pm,v 1.84 2004/10/28 20:40:12 jason Exp $)
>
>
>   Cheers.
>
>
> -- 
> Jérémy JUST  <jeremy_just at netcourrier.com>
>
> <blast_score_format.diff>
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l at portal.open-bio.org
> http://portal.open-bio.org/mailman/listinfo/bioperl-l

--
Jason Stajich
Duke University
http://www.duke.edu/~jes12/