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

View of /FigKernelScripts/map_sims_for_bigfc.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, 3 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.

# Map pegs to numbers for the BigFC computation.

use strict;

use FIG;
my $fig = new FIG;
my $dbf = $fig->db_handle();

my $usage = "map_sims_for_bigfc sims_dir new_sims_dir close_tab new_close_tab";

my($sims, $new_sims, $close, $new_close) = @ARGV;

-d $sims or die "Sims dir $sims not present";
-d $new_sims and die "New sims dir $new_sims already present";
-f $close or die "Close table $close not present";
-f $new_close and die "New close table already present";

$dbf->drop_table(tbl => "genome_mapping");
$dbf->create_table(tbl => "genome_mapping",
		   flds => "genome varchar(32) primary key, gnum smallint unique");
my $next_genome = 1;
my %cached_map;

open(MAPFH, ">genome.mapping") or die "cannnot write genome.mapping: $!";

opendir(SIMS, $sims) || die "$sims does not exist";
my @files = grep { $_ !~ /^\./ } readdir(SIMS);

mkdir($new_sims) or die "mkdir $new_sims failed: $!";

foreach my $file (@files)
    if (-s "$sims/$file")
	open(S, "<$sims/$file") or die "Cannot open sims $sims/$file: $!";
	open(N, ">$new_sims/$file") or die "Cannot write new sims file $new_sims/$file: $!";
	while (<S>)
	    my($g1, $g2);

	    if (/^fig\|(\d+\.\d+)\.peg\.(\d+)\tfig\|(\d+\.\d+)\.peg\.(\d+)\t(\S+)\t(\S+)\t(\S+)$/)
		$g1 = map_genome($1);
		$g2 = map_genome($3);

		print N join("\t", $g1, $2, $g2, $4, $5, $6, $7), "\n"o;

open(C, "<$close") or die "Cannot open $close: $!";
open(NC, ">$new_close") or die "Cannot open $new_close: $!";

while (<C>)
    if (/^fig\|(\d+\.\d+)\.peg\.(\d+)\tfig\|(\d+\.\d+)\.peg\.(\d+)$/)
	my $g1 = map_genome($1);
	my $g2 = map_genome($3);
	print NC join("\t", $g1, $2, $g2, $4), "\n";


$dbf->load_table(tbl => "genome_mapping",
		 file => "genome.mapping");

sub map_genome
    my($genome) = @_;
    my $mapped;
    if (defined($mapped = $cached_map{$genome}))
	return $mapped;

    $mapped = $next_genome++;
    $cached_map{$genome} = $mapped;
    print MAPFH "$genome\t$mapped\n";
    return $mapped;

MCS Webmaster
ViewVC Help
Powered by ViewVC 1.0.3