[Bio] / FigKernelScripts / remove_clustered_pchs2.pl Repository:
ViewVC logotype

View of /FigKernelScripts/remove_clustered_pchs2.pl

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1.1 - (download) (as text) (annotate)
Thu May 31 19:24:43 2007 UTC (12 years, 8 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, 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, HEAD
Missing scripts.

#
# 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.
#

#
# Version of remove_clustered_pchs that does not use the genomeid/pegid form.
#
###########################################
#
# RAO: I have changed this to keep all PCHs, but adding a code (0 means "do not count in score")
#

use strict;

use DB_File;
use Data::Dumper;
use FIG;
use FIGV;
my $fig;

use DBrtns;

my $usage = "usage: remove_clustered_pchs2 [-orgdir orgdir]  CutOff < all_pchs > filtered_pchs";

my $orgdir;
while ((@ARGV > 0) && ($ARGV[0] =~ /^-/))
{
    my $arg = shift @ARGV;
    if ($arg =~ /^-orgdir/)
    {
	$orgdir = shift @ARGV;
    }
    else
    {
	die "Unknown option $arg\n";
    }
}

my($cutoff);

@ARGV == 1 or die $usage;

(
 ($cutoff   = shift @ARGV) 
)
    || die $usage;

if ($orgdir)
{
    $fig = new FIGV($orgdir);
}
else
{
    $fig = new FIG;
}

my($keep, $discard);

open(SORT, "sort -k 1,2 -k 5,5n |");
#my $pch = <STDIN>;
my $pch = <SORT>;
while ($pch && ($pch =~ /^(\S+\t\S+)/))
{
    my $curr = $1;
    my @set = ();

    #
    # The input is four "columns" of pegs, followed by two iden columsn and two paranN columns.
    # Each peg column is really a pair of numbers.
    #

    while ($pch and
	   $pch =~ /^((\S+)\t(\S+))\t   # first two. $1 is the pair.
	             (\S+)\t		    # third peg col, $4.
	             (\S+)\t		    # fourth peg col, $5
	   	     (\S+)\t(\S+)\t		    # and the iden  columns. $6 and $7
	   	     (\S+)\t(\S+)  	    	    # and the paran cols $8 and $9
	            /x and
	   ($1 eq $curr))
    {
	# warn "got 1='$1' 4='$4' 5='$5' 6='$6' 7='$7'\n";

	#
	# We remap the line for output.
	#

	my $remap = join("\t",
			 $2, $3, $4, $5,
			 $6, $7, $8, $9);
	my $max_iden = ($6 < $7) ? $7 : $6;
	push(@set,[$4,$5,$remap,$max_iden]);
	$pch = <SORT>;
    }
    &process_set(\@set);
}
warn "keep=$keep discard=$discard\n";


# You have a list of PCHs.  Basically, we sort them based on max % identity
# between the corresponding genes in each PCH.
#
# We are trying to build a set of PCHs (marked by 1s) such that
#
#    1. The % identity is below the cutoff for each PCH and
#
#    2. The % identity between the "other" pairs of the PCHs is
#       always below the cutoff.
#
# 

sub process_set {
    my($set) = @_;
    my($pch,$i);

#    warn "Process " . Dumper($set);

    my @unprocessed = sort { $a->[3] <=> $b->[3] } @$set;

    #
    # Speculatively grab all the sims for this block.
    #
    # Didn't work, too many sims calls
    #
#    my @sims = $fig->sims([map { ($_->[0], $_->[1]) } @unprocessed],
#			  5000, 1.0e-5, 'fig', undef, { min_iden => $cutoff });

    if (@unprocessed > 0)
    {
	while ($pch = shift @unprocessed)
	{
	    #
	    # $pch here is a triple (peg3, peg4, whole line from file)
	    #
	    if ($pch->[3] <= $cutoff)
	    {
		print $pch->[2] . "\t1\n";

		my @sims = $fig->sims([$pch->[0], $pch->[1]], 5000, 1.0e-5, 'fig', undef, { min_iden => $cutoff });
		my $too_close1 = &get_too_close($pch->[0], \@sims);
		my $too_close2 = &get_too_close($pch->[1], \@sims);
		for ($i = $#unprocessed; ($i >= 0); $i--)
		{
		    if ($too_close1->{$unprocessed[$i]->[0]} ||
			$too_close2->{$unprocessed[$i]->[1]})
		    {
			print $unprocessed[$i]->[2] . "\t0\n";
			splice(@unprocessed,$i,1);
		    }
		}
	    }
	    else
	    {
		print $pch->[2] . "\t0\n";
	    }
	}
    }
}
    
sub get_too_close {
    my($peg, $sims) = @_;

    my $too_close = {};

#    my @sims = $fig->sims($peg, 5000, 1.0e-5, 'fig');
#    my @sims = grep { $_->id1 eq $peg } @$sims;

    # print "too_close: ". Dumper($peg, \@sims);
    foreach my $sim (grep { $_->id1 eq $peg } @$sims)
    {
	if ($sim->iden >= $cutoff)
	{
	    $too_close->{$sim->id2} = 1;
	    $keep++;
	}
	else
	{
	    $discard++;
	}
    }
    # print Dumper($peg, $too_close);
    return $too_close;
}


MCS Webmaster
ViewVC Help
Powered by ViewVC 1.0.3