RPS-BLAST

ただの備忘録

blast+ 2.2.28でrps-blast用のpssmデータベースの作成

% wget ftp://ftp.ncbi.nih.gov/pub/mmdb/cdd/cdd.tar.gz

% tar zxfv cdd.tar.gz

% ls | grep "KOG" | grep "smp" > KOG.pn

 KOG*.smpからtitleだけパース

<code>

#!/usr/env/bin perl

use strict;

use warnings;

my $foo = "";

my $hoge = 1;

open my $fh, "<", shift;

while( <$fh> ) {

  if( $_=~ /descr/ && $hoge == 1) {

    $hoge=0;

  }

  elsif( $_ =~ /\}\,$/ && $hoge == 0) {

    $foo .= $_;

    $hoge = 1;

    $foo =~ s/\n//g;

    $foo =~ s/title//;

    $foo =~ s/[\"\}\,]//g;

    $foo =~ s/^\s+//;

    print $foo."\n";

    $foo = "";

  }

  elsif ( $hoge == 0 ) {

    $foo .= $_;

  }

}

</code>

# % makeprofiledb -in foo.pn

% rpsblast -query /path/to/query.fa -db foo.pn -out test_rps_blast.txt -outfmt '6 qseqid stitle pident length mismatch gapopen qstart qend sstart send evalue bitscore'