Polyphobius.pl

This Perl script automatically generate input data for Polyphobius from given query sequence and call Polyphobius for prediction of transmembrane helices.

Code

<source lang="perl">

1. !/usr/bin/perl -w
2. run polyphobius
3. 1, get all homologous sequence with blastget
4. 2, insert query sequence at the beginning of the sequence set
5. 3, run kalign to generate multiple sequence alignment
6. 4, run polyphobius at last
7. usage: polyphobius [-d db=uniprot_sprot] [-i index=uniprot_sprot.idx] -q query -o output_dir

use strict; use File::Basename; use constant DATABASE => '/mnt/project/pracstrucfunc13/data/swissprot/uniprot_sprot';

use constant INDEX => '/mnt/project/pracstrucfunc13/data/index_pp/uniprot_sprot.idx';

my $blastget = '/mnt/project/pracstrucfunc13/polyphobius/blastget'; my $kalign = 'kalign'; my $polyphobius = '/mnt/project/pracstrucfunc13/polyphobius/jphobius -poly ';

1. parse command-line arguments

my ($db, $q, $odir, $index); while ( (my $param = shift @ARGV) ) { if ($param eq '-d') { $db = shift @ARGV; }

if ($param eq '-i') { $index = shift @ARGV; }

if ($param eq '-q') { $q = shift @ARGV; }

if ($param eq '-o') { $odir = shift @ARGV; } }

if (!$db) { $db = DATABASE; $index = INDEX; }

if (!$index) { $index = INDEX; }

if (! $q or ! $odir) { die "Usage: polyphobius [-d db=uniprot_sprot] [-i index=uniprot_sprot.idx] \\ -q query -o output_dir\n"; }

1. prepare output directory

if (! -d $odir) { mkdir $odir }

print "Run blastget ...\n";

1. run blastget

my $basename = &basename($q); my $outfile = "$odir/$basename-blast.fasta"; system("cat $q > $outfile"); system("$blastget -d $db -i $index $q >> $outfile 2> /dev/null");

print "Run kalign ...\n";

1. run kalign

my $msafile="$odir/$basename-kalign.msa"; system("$kalign $outfile > $msafile 2> /dev/null");

print "Run polyphobius ...\n";

1. run polyphobius

my $phobiusfile = "$odir/$basename-phobius.txt"; system("$polyphobius $msafile > $phobiusfile 2> /dev/null");

print "Done!\n"; </source>