[Bioperl-l] need a devloper for RemoteBlast (was RemoteBlast: can't
get a TaxBlast report)
Jason Stajich
jason at cgt.duhs.duke.edu
Thu Oct 16 20:19:13 EDT 2003
I really have no idea how to parse this so I suspect SearchIO doesn't
either. This is not something we have ever tried to support because it
isn't in part of the standard output from stand alone blast reports.
Frankly this module started as a hack to provide and illustrate minimal
access just to running BLAST on a remote server such as NCBI when the cgi
page itsself was pretty minimal.
If people want to do more complicated stuff the module needs a dedicated
soul actively developing it who will adapt it to the new features NCBI is
providing. In case it isn't clear, I am not going to be that person any
time soon.
Currently there are also new features which allow sending multiple
sequences in and using discontinuous megablast which will allow, for
example, in parallel searching of multiple sequences reducing the total
search time to a fraction of the time versus compared to each query
in-serial. This would be a big time save against things like dbest.
(I bang this drum every few months so it isn't directed at you Bill.)
-jason
On Thu, 16 Oct 2003, Cannon, William R wrote:
> Hi, I tried running RemoteBlast using the synopsis script found on at
> http://doc.bioperl.org/releases/bioperl-1.2/Bio/Tools/Run/RemoteBlast.html.
> Everything worked fine, as is. However, when I try to get a Taxonomy
> report by adding the line:
> $Bio::Tools::Run::RemoteBlast::HEADER{'FORMAT_OBJECT'} = 'TaxBlast';
>
> I still get the alignment report back. Anybody know what else I should
> be doing?
>
> Thanks.
> Bill
>
> (Using :
> perl 5.6
> bioperl 1.2.3 on redhat linux 7.2.
>
> Here is the full script:
> #! /msrc/apps/perl-5.6/bin/perl
>
> #Remote-blast "factory object" creation and blast-parameter initialization
>
> use Bio::Tools::Run::RemoteBlast;
> use strict;
> my $prog = 'blastp';
> my $db = 'nr';
> my $e_val= '1.0';
>
> my @params = ( '-prog' => $prog,
> '-data' => $db,
> '-expect' => $e_val,
> '-readmethod' => 'SearchIO' );
>
> my $factory = Bio::Tools::Run::RemoteBlast->new(@params);
>
> #change a paramter
> $Bio::Tools::Run::RemoteBlast::HEADER{'ENTREZ_QUERY'} = 'Homo sapiens [ORGN]';
> $Bio::Tools::Run::RemoteBlast::HEADER{'FORMAT_OBJECT'} = 'TaxBlast';
>
> #remove a parameter
> delete $Bio::Tools::Run::RemoteBlast::RETRIEVALHEADER{'ALIGNMENT_VIEW'};
> delete $Bio::Tools::Run::RemoteBlast::RETRIEVALHEADER{'DESCRIPTIONS'};
> delete $Bio::Tools::Run::RemoteBlast::RETRIEVALHEADER{'ALIGNMENTS'};
>
> my $v = 1;
> #$v is just to turn on and off the messages
>
> # my $str = Bio::SeqIO->new(-file=>'amino.fa' , '-format' => 'fasta' );
> my $str = Bio::SeqIO->new(-file=>'roa1.fasta' , '-format' => 'fasta' );
>
> while (my $input = $str->next_seq()){
> #Blast a sequence against a database:
>
> #Alternatively, you could pass in a file with many
> #sequences rather than loop through sequence one at a time
> #Remove the loop starting 'while (my $input = $str->next_seq())'
> #and swap the two lines below for an example of that.
> print STDERR "%s\n", $Bio::Tools::Run::RemoteBlast::HEADER{'FORMAT_OBJECT'};
> my $r = $factory->submit_blast($input);
> #my $r = $factory->submit_blast('amino.fa');
>
> print STDERR "waiting..." if( $v > 0 );
> while ( my @rids = $factory->each_rid ) {
> foreach my $rid ( @rids ) {
> my $rc = $factory->retrieve_blast($rid);
> if( !ref($rc) ) {
> if( $rc < 0 ) {
> $factory->remove_rid($rid);
> }
> print STDERR "." if ( $v > 0 );
> sleep 5;
> } else {
> my $result = $rc->next_result();
> #save the output
> my $filename = $result->query_name()."\.out";
> $factory->save_output($filename);
> $factory->remove_rid($rid);
> print "\nQuery Name: ", $result->query_name(), "\n";
> while ( my $hit = $result->next_hit ) {
> next unless ( $v > 0);
> print "\thit name is ", $hit->name, "\n";
> while( my $hsp = $hit->next_hsp ) {
> print "\t\tscore is ", $hsp->score, "\n";
> }
> }
> }
> }
> }
> }
>
>
>
>
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l at portal.open-bio.org
> http://portal.open-bio.org/mailman/listinfo/bioperl-l
>
--
Jason Stajich
Duke University
jason at cgt.mc.duke.edu
More information about the Bioperl-l
mailing list