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

View of /FigKernelScripts/salvage_subsystem_rows.pl

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1.6 - (download) (as text) (annotate)
Sat Jan 7 23:11:10 2006 UTC (14 years, 6 months ago) by overbeek
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 1.5: +16 -6 lines
fixes to salvage code

########################################################################
#
# 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;
my $fig = new FIG;

my $usage = "usage: salvage_subsystem_rows From To Maps [ToData]";  

# Foreach Subsystem in To (a directory of subsystems), if Subsystem is in From,
# then each row representing an organism in Mapping will be "salvaged" and added
# to Subsystem in To.  If ToData is included, then any subsystems in From, but not in To
# will be copied to To (deleting rows for genomes not in ToData)
#

(
 ($from    = shift @ARGV) &&
 ($to      = shift @ARGV) &&
 ($mapping = shift @ARGV)
)
    || die $usage;

my $tmpF = "$FIG_Config::temp/extract_subsystems.$$";
if (@ARGV == 1)
{
    open(TMP,">$tmpF") 
	|| die "could not open $tmpF";
    while (defined($_ = <DATA>))
    {
	print TMP $_;
    }
    close(TMP);
    &copy_new($from,$to,$ARGV[0],$tmpF);
}

foreach $_ (`cat $mapping/*`)
{
    if ($_ =~ /^(fig\|(\d+\.\d+)\.peg\.\d+)\t(fig\|(\d+\.\d+)\.peg\.(\d+))$/)
    {
	$old{$2} = $4;
	$map{$1} = $5;
    }
}

opendir(SUBS,$to) || die "could not open $to";
@subs = grep { $_ !~ /^\./ } readdir(SUBS);
closedir(SUBS);

foreach $sub (@subs)
{
    next if (! -s "$from/$sub/spreadsheet");
    print STDERR "updating $sub\n";
    undef %salvaged;
    open(IN,"<$from/$sub/spreadsheet") || die "$from/$sub/spreadsheet";
    while (defined($_ = <IN>))
    {
	if (($_ =~ /^(\d+\.\d+)\t(\S+)\t(.*)$/) && ($new = $old{$1}))
	{
	    my $genomeOld = $1;
	    $var = $2;
	    @old_pegs = split(/\t/,$3);
	    @new_pegs = ();
	    foreach my $set (@old_pegs)
	    {
		my @new_set = ();
		foreach $peg (map { "fig|$genomeOld.peg." . $_ } split(/,/,$set))
		{
		    if ($peg && ($pegN = $map{$peg}))
		    {
			push(@new_set,$pegN);
		    }
		}

		if (@new_set > 0)
		{
		    push(@new_pegs,join(",",@new_set));
		}
		else
		{
		    push(@new_pegs,"");
		}
	    }
	    $salvaged{$new} = join("\t",($new,$var,@new_pegs));
	}
    }
    close(IN);

    open(IN,"<$to/$sub/spreadsheet") || die "$to/$sub/spreadsheet";
    $/ = "\n//\n";
    @spreadsheet = <IN>;
    close(IN);

    open(OUT,">$to/$sub/spreadsheet") || die "$to/$sub/spreadsheet";
    for ($i=0; ($i < @spreadsheet); $i++)
    {
	if ($i != 2)
	{
	    print OUT $spreadsheet[$i];
	}
	else
	{
	    $_ = $spreadsheet[$i];
	    chomp;
	    @lines = split(/\n/,$_);
	    %in_already = map { $_ =~ /^(\d+\.\d+)/; $1 => 1 } @lines;
	    foreach $new (keys(%salvaged))
	    {
		if (($new =~ /^(\d+\.\d+)/) && (! $in_already{$1}))
		{
		    push(@lines,$salvaged{$new});
		}
	    }
	    print OUT join("\n",@lines),"\n";
	}
    }
    close(OUT);
    $/ = "\n";
}
if (@ARGV == 1) { unlink($tmpF); }

sub copy_new {
    my($from,$to,$to_data,$tmpF) = @_;

    opendir(ORG,"$to_data/Organisms") || die "$to/Organisms does not exist";
    my @orgs = grep { $_ =~ /^\d+\.\d+$/ } readdir(ORG);
    closedir(ORG);

    opendir(FROM,$from) || die "could not open $from";
    my @subsystems = grep { $_ !~ /^\./ } readdir(FROM);
    closedir(FROM);

    foreach my $subsystem (@subsystems)
    {
	if (! -d "$to/$subsystem")
	{
	    print STDERR "copying $subsystem\n";
	    &filter_genomes("$from/$subsystem","$to/$subsystem",\@orgs,$subsystem,$tmpF);
	}
    }
}

sub filter_genomes {
    my($from,$to,$orgs,$subsystem,$tmpF) = @_;

    my $genomesF = "$FIG_Config::temp/genomes.$$";
    open(TMP,">$genomesF") || die "could not open $genomesF";
    foreach $org (@$orgs)
    {
	print TMP "$org\n";
    }
    close(TMP);

    my $rc = system("extract_genomes",$genomesF,$from,$to,$tmpF);
    if ($rc != 0)
    {
	print STDERR "failed to copy $from $to\n";
    }
    unlink($genomesF);
}

__DATA__
/^\S.*\S$/
	Alignments,erase
	Backup,erase
	/^[CEV].*$/,copy
	curation.log,copy
	notes,copy
	reactions,copy
	spreadsheet,filter_rows
	SubsystemDiagrams,copyR
	diagrams,copyR
	/^assignments*/,erase
	/^rowss*/,erase
	constructs,copyR
	rows,erase
	/\.log$/,copyR
	MAP_SUPPORT,copyR
	/^.*~$/,erase

MCS Webmaster
ViewVC Help
Powered by ViewVC 1.0.3