[Bio] / FigKernelScripts / find_protein_homologs.pl Repository:
ViewVC logotype

View of /FigKernelScripts/find_protein_homologs.pl

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1.1 - (download) (as text) (annotate)
Sat Nov 24 01:11:24 2007 UTC (12 years, 4 months ago) by golsen
Branch: MAIN
CVS Tags: mgrast_dev_08112011, rast_rel_2009_05_18, mgrast_dev_08022011, rast_rel_2014_0912, rast_rel_2008_06_18, myrast_rel40, rast_rel_2008_06_16, mgrast_dev_05262011, rast_rel_2008_12_18, mgrast_dev_04082011, rast_rel_2008_07_21, rast_rel_2010_0928, rast_2008_0924, mgrast_version_3_2, mgrast_dev_12152011, rast_rel_2008_04_23, mgrast_dev_06072011, rast_rel_2008_09_30, rast_rel_2009_0925, rast_rel_2010_0526, rast_rel_2014_0729, mgrast_dev_02212011, rast_rel_2010_1206, mgrast_release_3_0, mgrast_dev_03252011, rast_rel_2010_0118, mgrast_rel_2008_0924, mgrast_rel_2008_1110_v2, rast_rel_2009_02_05, rast_rel_2011_0119, mgrast_rel_2008_0625, mgrast_release_3_0_4, mgrast_release_3_0_2, mgrast_release_3_0_3, mgrast_release_3_0_1, mgrast_dev_03312011, mgrast_release_3_1_2, mgrast_release_3_1_1, mgrast_release_3_1_0, mgrast_dev_04132011, rast_rel_2008_10_09, mgrast_dev_04012011, rast_release_2008_09_29, mgrast_rel_2008_0806, mgrast_rel_2008_0923, mgrast_rel_2008_0919, rast_rel_2009_07_09, rast_rel_2010_0827, mgrast_rel_2008_1110, myrast_33, rast_rel_2011_0928, rast_rel_2008_09_29, mgrast_rel_2008_0917, rast_rel_2008_10_29, mgrast_dev_04052011, mgrast_dev_02222011, rast_rel_2009_03_26, mgrast_dev_10262011, rast_rel_2008_11_24, rast_rel_2008_08_07
Debugging and adding support of all the NCBI genetic codes.

# -*- perl -*-
########################################################################
# Copyright (c) 2003-2007 University of Chicago and Fellowship
# for Interpretations of Genomes. All Rights Reserved.
#
# This file is part of the SEED Toolkit.
# 
# The SEED Toolkit is free software. You can redistribute
# it and/or modify it under the terms of the SEED Toolkit
# Public License. 
#
# You should have received a copy of the SEED Toolkit Public License
# along with this program; if not write to the University of Chicago
# at info@ci.uchicago.edu or the Fellowship for Interpretation of
# Genomes at veronika@thefig.info or download a copy from
# http://www.theseed.org/LICENSE.TXT.
########################################################################

use strict;
use gjoseqlib;
use find_special_proteins;

my $usage = <<'End_of_Usage';

Usage: find_protein_homologs [opts]          < contigs > loc_tab_seq_tab_note
       find_protein_homologs [opts] refprots < contigs > loc_tab_seq_tab_note

The reference proteins file must be included either as an option, or a command
parameter.  The option takes precidence.

Options:

   -d directory    Directory for temp files. If it exists, files are retained.
   -e expect       Max expectation value of blast match to reference (D = 1e-10).
   -g gen_code     NCBI genetic code to use (D = 1).
   -i init_codons  Comma separated list of (primary) initiator codons.
   -r refprotfile  Read reference sequences from specified file.
   -s alt_inits    Comma separated list of alternate (secondary) init codons.
   -t tmp          Location for a temporary directory of temporary files.
   -x term_codons  Comma separated list of terminator codons

End_of_Usage

my $expect = 1e-10;
my $gen_code;
my $is_init;
my $is_alt;
my $is_term;
my $reffile;
my $tmpdir;
my $tmp;

while ( $ARGV[0] =~ s/^-// )
{
    $_ = shift;

    #  These options consume the flag and a value

    if ( s/^d// ) { $tmpdir   = $_ || shift; next }
    if ( s/^e// ) { $expect   = $_ || shift; next }
    if ( s/^g// ) { $gen_code = $_ || shift; next }
    if ( s/^i// ) { $is_init  = $_ || shift; next }
    if ( s/^r// ) { $reffile  = $_ || shift; next }
    if ( s/^s// ) { $is_alt   = $_ || shift; next }
    if ( s/^t// ) { $tmp      = $_ || shift; next }
    if ( s/^x// ) { $is_term  = $_ || shift; next }

    #  These options just consume the flag letter, and can be combined

    #  So nothing should remain if we get here

    if ( $_ )
    {
        print STDERR "Bad flag '$_'.\n", $usage;
        exit;
    }
}

my @refs;
$reffile ||= shift;  #  If we did not get it as an option, it is paramter.
if ( $reffile )
{
    -f $reffile
        or print STDERR "Bad reference sequence file '$reffile'.\n", $usage
        and exit;

    @refs = gjoseqlib::read_fasta( $reffile );
    @refs or print STDERR "No reference sequences in '$reffile'.\n", $usage
          and exit;
}
else
{
    print STDERR "Program requires a reference sequence file.\n", $usage;
    exit;
}

my @contigs = gjoseqlib::read_fasta();
@contigs
    or print STDERR "No contig sequences read.\n", $usage
    and exit;

my $params = { contigs    => \@contigs,
               references => \@refs
             };
$params->{ expect  } = $expect  if $expect;
$params->{ code    } = $gen_code if $gen_code;
$params->{ is_init } = [ split /[^A-Za-z]+/, $is_init ] if $is_init;
$params->{ is_alt  } = [ split /[^A-Za-z]+/, $is_alt  ] if $is_alt;
$params->{ is_term } = [ split /[^A-Za-z]+/, $is_term ] if $is_term;
$params->{ tmp     } =  $tmp    if $tmp;
$params->{ tmpdir  } =  $tmpdir if $tmpdir;


my @prots = find_special_proteins::find_protein_homologs( $params );

my $entry;
foreach $entry ( @prots )
{
    print join( "\t", map { $entry->{$_} } qw( location sequence comment ) ), "\n"
}

exit;

MCS Webmaster
ViewVC Help
Powered by ViewVC 1.0.3