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

View of /FigKernelScripts/build_nr_sources.pl

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1.14 - (download) (as text) (annotate)
Fri May 11 20:18:06 2007 UTC (12 years, 6 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
Changes since 1.13: +1 -1 lines
Rollup of various rast-related changes.

# -*- perl -*-
#
# 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 strict;
use Getopt::Long;
use FIG;

my $job_base_dir = "/vol/48-hour/Jobs";

# usage: build_nr_sources [DataDir]

sub usage
{
    print STDERR "Usage: $0 [--48hr-job-file file] [--new-ext-data ext-data-dir] [data-dir]\n";
    print STDERR "   --48hr-job-file file: file containing job ids in the 48hr server which should be included in new NR\n";
    print STDERR "   --new-ext-data dir: directory of new external data to incorporate\n";
    exit(1);
}

my $ext_data_new;
my $data_dir;
my $jobfile;

my $rc = GetOptions("new-ext-data=s" => \$ext_data_new,
		    "48hr-job-file=s" => \$jobfile);

if (!$rc)
{
    &usage;
}

if (@ARGV == 0)
{
    $data_dir = $FIG_Config::data;
}
elsif (@ARGV == 1)
{
    $data_dir = shift @ARGV;
}
else
{
    &usage;
}

use FIG;
my $fig = new FIG;

#
# Find all NR files in the Data/NR directory. If we've got an $ext_data_new setting,
# directories there override those in the Data/NR directory.
#

my @dirs;

my %seen;

opendir(TMP,"$data_dir/NR") || die "could not open $data_dir/NR";
for my $dir (grep { $_ !~ /^\./ } readdir(TMP))
{
    my $path = "$data_dir/NR/$dir";
    next unless -d $path;
    
    if ($ext_data_new and -d "$ext_data_new/$dir")
    {
	$path = "$ext_data_new/$dir";
    }
    print STDERR "Use sources from $path\n";
    push(@dirs, $path);
    $seen{$dir}++;
}
closedir(TMP);

if ($ext_data_new and opendir(ED, $ext_data_new))
{
    for my $dir (grep { $_ !~ /^\./ } readdir(ED))
    {
	next if $seen{$dir};
	my $path = "$ext_data_new/$dir";
	next unless -d $path;

	print STDERR "Use sources from $path\n";
	push(@dirs, $path);
	$seen{$dir}++;
    }
    closedir(ED);
}

my $dir;
foreach $dir (@dirs)
{
    opendir(TMP,$dir) || die "could not open $dir";
    my @files = grep { $_ !~ /^\./ } readdir(TMP);
    closedir(TMP);

    if (@files == 1)
    {
	print "$dir/$files[0]\n";
    }
    else
    {
	my @tmp = grep { $_ =~ /fasta/ } @files;
	foreach $_ (@tmp)
	{
	    print "$dir/$_\n";
	}
    }
}

#
# Find the fasta files in the 48hr server based on the jobs file
#

if ($jobfile)
{
    open(J, "<$jobfile") or die "Cannot open jobfile $jobfile: $!\n";
    print STDERR "Use sources from 48hr job file $jobfile\n";
    while (<J>)
    {
	chomp;
	if (! /^\s*(\d+)\s*$/)
	{
	    die "Invalid job at line $. in jobfile $jobfile\n";
	}
	
	my $job = $1;
	my $jobdir = "$job_base_dir/$job";
	
	my $genome = &FIG::file_head("$jobdir/GENOME_ID", 1);
	if (!$genome )
	{
	    die "Job dir $jobdir not valid\n";
	}
	
	chomp $genome;
	
	my $fasta = "$jobdir/rp/$genome/Features/peg/fasta";
	if (-f $fasta)
	{
	    print "$fasta\n";
	}
	else
	{
	    warn "48hr fasta $fasta does not exist\n";
	}
    }
}    
#
# Find all fasta files in the Organisms directory
#

opendir(TMP, "$data_dir/Organisms");
while ($_ = readdir(TMP))
{
    next if /^\./;
    #next if /^9999999.\d+$/;

    #
    # Strip environmental sequences.
    # 
    # c.f. seed-tech mail thread of 2/8/2007 for discusson on the rationale of the following
    # logic.
    #
    # next if $fig->is_environmental($_);
    next if /^4{7}/ or /^9{7}/;
    
    next if (-e "$data_dir/Organisms/$_/DELETED");
    
    my $dir = "$data_dir/Organisms/$_/Features/peg/fasta";
    if (-f $dir)
    {
	print "$dir\n";
    }
}
closedir(TMP);


MCS Webmaster
ViewVC Help
Powered by ViewVC 1.0.3