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

View of /FigKernelScripts/export_genome_as_gff.pl

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1.6 - (download) (as text) (annotate)
Mon Dec 5 18:56:37 2005 UTC (13 years, 11 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 1.5: +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.
#

use FIG;
use strict;
use Data::Dumper;
use URI::Escape;
use POSIX;

#
# usage: export_genome_as_gff genome-id
#

sub iso_time
{
    my($t) = @_;

    my ($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst) = localtime($t);

    return POSIX::strftime("%Y-%m-%dT%H:%M:%S",
			  $sec, $min, $hour, $mday, $mon, $year);
}

if (@ARGV != 1)
{
    die "Usage: $0 genome-id\n";
}

my $genome = shift @ARGV;

my $fig = new FIG;

my $outfh = \*STDOUT;

my $w = $fig->get_gff_writer(outputtype => ['cds'],
			     linelength => 78,
			     with_assignments => 0,
			     with_aliases => 1);

print $outfh "#gff-version\t3\n";

#
# Bookkeeping.
#

my(%contig_names, @clist, @flist);


#
# First retrieve all annotations for this genome.
#

my $annos = $fig->annotations_made_fast([$genome], 0, time, 'all', 1);

#
# Snag the annos for this genome.
#

my @j = map { $_->[1] } grep { $_->[0] eq $genome } @$annos;
@j == 1 or die "$0: annotations_made_fast returned an unexpected result";

$annos = $j[0];

#
# Now walk all the features, generating gff chunks as we go.
#

my $features = $fig->all_features_detailed($genome);

my $count = 0;

for my $feat (@$features)
{
    my($fid, $loc_str, $alias_str, $type) = @$feat;

    print STDERR "." if $count++ % 100 == 0;

    my @aliases = split(/,/, $alias_str);
    my @loc = split(/,/, $loc_str);

    my $feature_annos = $annos->{$fid};

    my $group = $FIG_Config::group;
    $group = 'FIG' unless $group;

    #
    # Create the notes text for the annotation.
    #
    my $ntxt = '';

    if (!$feature_annos and $type eq 'peg')
    {
	#
	# We had no annotation for this feature. If there is an assignment,
	# make up an annotation with this date.
	#

	my $assign = $fig->function_of($fid);
	if ($assign)
	{
	    push(@$feature_annos, new Annotation($fid, time, 'export_process',
						 "Set $group function to", $assign));
	}
    }

    if (ref($feature_annos))
    {
	my @notes;
	
	for my $ent (@$feature_annos)
	{
	    my $nbody = join(":", $ent->anno_time(), $ent->made_by(), $ent->anno_text());
	    push(@notes, uri_escape($nbody));
	}

	$ntxt = "anno=" . join(",", @notes);
    }

    #
    # Create the GFF for this feature.
    my($contigs, $fasta) = $w->gff3_for_feature($fid, undef, $ntxt, \@aliases, \@loc);

    map { $contig_names{$_}++ } keys(%$contigs);

    push(@clist, $contigs);
    push(@flist, $fasta);
}

my $gdir = "$FIG_Config::organisms/$genome";

my $project = &FIG::file_head("$gdir/PROJECT", 1);
chomp($project);

print $outfh "#seed\tgenome\t$genome\t" . $fig->genus_species($genome) . "\n";
print $outfh "#seed\tgenome_md5\t$genome\t" . $fig->genome_md5sum($genome) . "\n";
print $outfh "#seed\ttaxonomy\t" . $fig->taxonomy_of($genome) . "\n";
print $outfh "#seed\tproject\t$project\n";

$w->write_gff3($outfh, $genome, \@clist, \@flist);

close(F);

MCS Webmaster
ViewVC Help
Powered by ViewVC 1.0.3