[Bioperl-l] Fwd: blast result not matching.

[Dave Messina]

As I think has been said earlier in this thread, it's almost certainly a
discrepancy in the BLAST parameters between what the blast_sequence function
in the Bio::Perl module is sending, and what the BLAST website is doing.


In this case, you have a very short sequence. If you look in the "Algorithm
parameters" section of the BLAST web form, you'll see that there is an
option that is checked by default, "Automatically adjust parameters for
short input sequences".


If I uncheck that option, I get the same results as you did when you
submitted your BLAST through BioPerl (see http://cl.ly/9ynq).


So to get the same results from a BioPerl-submitted BLAST and a BLAST on
NCBI's website, you need to have the same parameters. You can set the
parameters from BioPerl as described in the documentation:

http://search.cpan.org/~cjfields/BioPerl-1.6.901/Bio/Tools/Run/RemoteBlast.pm


As Jason said earlier, the blast_sequence function in Bio::Perl is intended
as a simple demonstration and uses the default BLAST parameters. That
function is a wrapper around the RemoteBlast module. Since you want to do
something a little different, I believe you'll need to use the RemoteBlast
module directly.




Dave




On Thu, Sep 8, 2011 at 08:11, Manju Rawat <manju.rawat2 at gmail.com> wrote:

> Toady i installed the latest version of bioperl in my system via CPAN..
> But this still not sowing the complete result..
>
>
> I just want to do nucleotide blast using bioperl..but while i am doing
> blast
> with my sequence it shwowing very samll result..
> I dnt know whether it is wrong or right...but while i am blasting the same
> sequence in NCBI it showing a diffrent result..
>
> and i have also tried to use the orignl module..but it also dnt work..
>
> Pl see reult of the balst in  attached file of this mail..
> #!usr/bin/perl -w
> use Bio::Perl;
> use Bio::SearchIO;
> $blast_report =blast_sequence('acggctgctgtagatctgatgct');
> write_blast(">resl.blast",$blast_report);
>
> Thanks.
> Manju Rawat
>
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/bioperl-l
>