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

View of /FigKernelScripts/to_map_coords.pl

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1.3 - (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 1.2: +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.
#

# usage: to_map_coords < regions > map.coords

use FIG;
my $fig = new FIG;

$/ = "\n//\n";

while (defined($_ = <STDIN>))
{
    chomp;
    my @lines = split(/\n/,$_);
    my $oligo_line = shift @lines;
    my($org,$contig,$beg,$end,undef) = split(/\t/,$oligo_line);
    my @genes = map { [split(/\t/,$_)] } @lines;

    if ($beg > $end)
    {
	my $minC = &minC(\@genes);
	my $maxC = &maxC(\@genes);

	($beg,$end) = (&transform($minC,$maxC,$beg),&transform($minC,$maxC,$end));
	@genes = &flip(\@genes,$minC,$maxC);
    }

    foreach $x (@genes)
    {
	$x->[1] -= $beg;
	$x->[2] -= $beg;
    }

    unshift(@genes,[$org,$contig,0,abs($end-$beg)]);
    push(@maps,[@genes]);
}

$min = 100000000;
$max = 0;
foreach $genes (@maps)
{
    for ($i=1; ($i < @$genes); $i++)
    {
	$x = $genes->[$i];
	$min = &FIG::min($x->[1],$min); $min = &FIG::min($x->[2],$min);
	$max = &FIG::max($x->[1],$max); $max = &FIG::max($x->[2],$max);
    }
}

foreach $genes (@maps)
{
    $genes->[0]->[2] -= ($min-1);
    $genes->[0]->[3] -= ($min-1);

    for ($i=1; ($i < @$genes); $i++)
    {
	$x = $genes->[$i];
	$x->[1] -= ($min-1);
	$x->[2] -= ($min-1);
    }
}

$max = $max+1 - $min;

print "1\t$max\n";
foreach $genes (@maps)
{
    foreach $x (@$genes)
    {
	print join("\t",@$x),"\n";
    }
    print "//\n";
}

sub flip {
    my($genes,$minC,$maxC) = @_;
    my($x,$t);

    my @flipped;

    foreach $x (@$genes)
    {
	my $copy = [@$x];
	($copy->[1],$copy->[2]) = (&transform($minC,$maxC,$copy->[1]),&transform($minC,$maxC,$copy->[2]));
	push(@flipped,$copy);
    }
    return @flipped;
}

sub transform {
    my($minC,$maxC,$x) = @_;

    return $minC + ($maxC - $x);
}

sub minC {
    my($genes) = @_;

    my $minC = 100000000;
    foreach $gene (@$genes)
    {
	$minC = &FIG::min($minC,$gene->[1]);
	$minC = &FIG::min($minC,$gene->[2]);
    }
    return $minC;
}

sub maxC {
    my($genes) = @_;

    my $maxC = 100000000;
    foreach $gene (@$genes)
    {
	$maxC = &FIG::max($maxC,$gene->[1]);
	$maxC = &FIG::max($maxC,$gene->[2]);
    }
    return $maxC;
}


MCS Webmaster
ViewVC Help
Powered by ViewVC 1.0.3