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

View of /FigKernelScripts/verify_pegs_in_tbl.pl

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1.2 - (download) (as text) (annotate)
Mon Dec 5 18:56:38 2005 UTC (14 years, 2 months ago) by olson
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, caBIG-05Apr06-00, 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, caBIG-13Feb06-00, rast_rel_2009_03_26, mgrast_dev_10262011, rast_rel_2008_11_24, rast_rel_2008_08_07, HEAD
Changes since 1.1: +17 -0 lines
Add license words.

#
# Copyright (c) 2003-2006 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 FIG;


# usage: verify_pegs_in_tbl Contigs < tbl > report

($contigs = shift @ARGV)
    || die "usage: verify_pegs_in_tbl Contigs < tbl > report";

open(CONTIGS,$contigs) || die "could not open $contigs";
$/ = "\n>";
while (defined($_ = <CONTIGS>))
{
    chomp;
    if ($_ =~ /^>?(\S+)[^\n]*\n(.*)/s)
    {
	$id  =  $1;
	$seq =  $2;
	$seq =~ s/\s//g;
	$contigs{$id} = uc $seq;
    }
}
close(CONTIGS);
$/ = "\n";

while (defined($_ = <STDIN>))
{
    if (($_ =~ /^(fig\|\d+\.\d+\.peg\.\d+)\t(\S+)/) &&
	(($rid,$loc) = ($1,$2)) &&
	($loc =~ /^([^,]+)_(\d+)_\d+(,|$)/) &&
	(($contig,$beg) = ($1,$2)) &&
	($loc =~ /$contig\_\d+_(\d+)$/) &&
	($end = $1))
    {
	$seq = &FIG::extract_seq(\%contigs,$loc);
	if ($contigs{$contig} && ($beg < $end))
	{
	    $start = substr($contigs{$contig},$beg-1,3);
	    $stop  = substr($contigs{$contig},$end,3);
	    for ($i=0; ($i < (length($seq) - 3)) && (substr($seq,$i,3) !~ /(TAG|TGA|TAA)/); $i += 3) {}
	    if ($i < (length($seq)-3))
	    {
		print "$rid has an embedded stop codon\n";
	    }
	}
	elsif ($contigs{$contig})
	{
	    $start = &reverse_comp(substr($contigs{$contig},$beg-3,3));
	    $stop  = &reverse_comp(substr($contigs{$contig},$end-4,3));
	    for ($i=0; ($i < (length($seq) - 3)) && (substr($seq,$i,3) !~ /(TAG|TGA|TAA)/); $i += 3) {}
	    if ($i < (length($seq)-3))
	    {
		print "$rid has an embedded stop codon\n";
	    }
	}
	else
	{
	    print STDERR "invalid: $_";
	}

	if (($start !~ /^[ATG]TG/) && ($beg > 40) && ($beg < (length($contigs{$contig})-40)))
	{
	    print "bad start for $rid: $start\n";
	}
	if (($stop !~ /(TAG|TGA|TAA)/) && ($end > 40) && ($end < (length($contigs{$contig}) - 40)))
	{
	    print "bad stop for $rid: $stop\n";
	}
	$strand = ($beg < $end) ? "+" : "-";

	push(@{ $on{$contig} }, [&FIG::min($beg,$end),&FIG::max($beg,$end),$strand,$rid]);
    }
}

foreach $contig (sort keys(%on))
{
    $x = $on{$contig};
    @on = sort { ($a->[0] <=> $b->[0]) or ($a->[1] <=> $b->[1]) or ($a->[2] cmp $b->[2]) } @$x;

    for ($i=0; ($i < $#on); $i++)
    {
	for ($j=$i+1; ($j < @on) && ($on[$i]->[1] > ($on[$j]->[0]) + 10); $j++)
	{
	    $overlap = ($on[$j]->[1] > $on[$i]->[1]) ?
		           ($on[$i]->[1] - $on[$j]->[0]) + 1 :
			   ($on[$j]->[1] - $on[$j]->[0]) + 1;
	    push(@overlaps,[$overlap,"$overlap\t$on[$i]->[2],$on[$j]->[2]\t$on[$i]->[3] overlaps $on[$j]->[3]\n"]);
	}
    }
}

foreach $_ (sort { $b->[0] <=> $a->[0] } @overlaps)
{
    print $_->[1];
}
	
sub reverse_comp {
    my( $seq ) = @_;
    my( $rev  );

    $rev =  reverse( $seq );
    $rev =~ tr/a-z/A-Z/;
    $rev =~ tr/ACGTUMRWSYKBDHV/TGCAAKYWSRMVHDB/;
    return $rev;
}

MCS Webmaster
ViewVC Help
Powered by ViewVC 1.0.3