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

View of /FigKernelScripts/make_rates.pl

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1.2 - (download) (as text) (annotate)
Mon Dec 5 18:56:37 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.
#

# usage: make_rates Ali BigTree > weights

use tree_utilities;
use Carp;
use fastDNAml;

$usage = "usage: make_rates Ali BigTree > weights";

($ali = shift @ARGV) 
    || die "bad alignment: $usage";

(($tree_file = shift @ARGV) && (-s $tree_file))
    || die "missing tree: $usage";

##############  get the tree ##########
($tree = &parse_newick_tree(join("",`cat \"$tree_file\"`)))
    || die "could not parse the tree";

if (&size_tree($tree) > 100)
{
    $rep_tree = &representative_by_size($tree,80);
}
else
{
    $rep_tree = $tree;
}

$rep_tips = &tips_of_tree($rep_tree);

@in_tree = ();
foreach $x (@$rep_tips)
{
    $y = &label_to_printable($x);
    $in_tree{$y} = 1;
    push(@in_tree,$y);
}

############### load relevant sequecnes from alignment ###########

open(ALI,"<$ali") || die "could not open $ali";
$_ = <ALI>;
while ($_ && ($_ =~ /^>(\S+)/))
{
    $id = $1;
	
    @lines = ();
    while (defined($_ = <ALI>) && ($_ !~ /^>/))
    {
	chomp;
	push(@lines,$_);
    }

    if ($in_tree{$id})
    {
	$seq = join("",@lines);
	$seq =~ tr/a-z/A-Z/;
	$seq =~ tr/U/T/;
	$seq =~ s/[~.]/\?/g;
	$x   = [$seq];               # this is a hack, I want the assoc array
	$seq_of{$id} = \($x->[0]);   # to contain pointers to the seqs (not the seqs)
    }
}
close(ALI);

@id_seqs = ();
foreach $tip (@$rep_tips)
{
    $id  = &label_to_printable($tip);
    $seq = $seq_of{$id};
    defined($id) || die "bad id: $id";
    defined($seq) || die "bad seq: $id $seq";
    push(@id_seqs,[$id,$$seq]);
}


@translated = ();
foreach $pair (@id_seqs)
{
    $id = $pair->[0];
    if (index($id,"_") >= 0)
    {
	$id1 = $id;
	$id1 =~ s/_/./g;
	$pair->[0] = $id1;
	push(@translated,[$id,$id1]);
    }
}

$bad = 0;
foreach $id (@$rep_tips)
{
    $id = &label_to_printable($id);
    if (! $seq_of{$id})
    {
	$bad = 1;
	warn "The tree contains $id, but it is not in the alignment\n";
    }
}
(! $bad) || die "Could not insert - alignment is missing sequences";

$num_seqs = @id_seqs;
$len_seqs = length($id_seqs[0]->[1]);
mkdir("Tmp$$",0777);
open(TMP,">Tmp$$/tmp$$") || die "could not open Tmp$$/tmp$$";
print TMP "$num_seqs $len_seqs F C Y\nC 35\n";

foreach $pair (@id_seqs)
{
    ($id,$seq) = @$pair;
    $pad = " " x (13 - length($id));
    print TMP "$id$pad$seq\n";
}
print TMP "1\n";
$rep_tree = &uproot($rep_tree);
$ptree = &to_translated_newick($rep_tree);
print TMP "$ptree\n";
close(TMP);
@out = `cd Tmp$$; DNAml_rates < tmp$$ > /dev/null; cat weight_rate*; cd ..`;
system("/bin/rm -r Tmp$$");
print @out;

sub to_translated_newick {
    my($tree) = @_;
    my($x,$pair,$y);

    $x = &to_newick($tree);
    foreach $pair (@translated)
    {
	$y = quotemeta($pair->[0]);
	$x =~ s/$y/$pair->[1]/g;
    }
    return $x;
}

MCS Webmaster
ViewVC Help
Powered by ViewVC 1.0.3