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

View of /FigKernelScripts/gff_statistics.pl

Parent Directory Parent Directory | Revision Log Revision Log

Revision 1.1 - (download) (as text) (annotate)
Wed Dec 6 16:01:55 2006 UTC (13 years, 2 months ago) by redwards
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
releasing gff statistics



=head1 gff_statistics

Generate some statistics about the GFF3 files. We'll parse each directory and count everything we can.

=head1 author

Rob Edwards


use strict;
use FIG;
my $fig=new FIG;
use File::Find;

my $dir=shift || die "$0 <dir of gff files>";

my @files;
find(sub {push @files, $File::Find::name if (m/gff$/i || m/gff3$/i)}, $dir);

my $type;
my $count; my $total; my $all;
foreach my $file ( @files )
	open(IN, $file) || die "Can't open $file";
	while (<IN>)
		my @line=split /\t/;
		next unless ($line[8]);
		next if ($line[2] eq "contig" || $line[2] eq "gene");
		my $fid=$1;
		unless ($fid) {print STDERR "Couldn't get a FIG ID from $_\n"; next}
		$fid =~ m/fig\|\d+\.\d+\.(\w+)\.\d+/;
		my @pieces=split /\;/, $line[8];
		foreach my $part ( @pieces )
			my ($id, $vars)=split /\=/, $part;
			unless ($id && $vars)
				print STDERR "Couldn't get any variables from $part in $_\n";
			$count->{$id}->{$fid}++; # how many pegs have this feature
			my @separate=split /\,/, $vars;
			$total->{$id}+=scalar(@separate); # how many pieces in total
			map {$all->{$id}->{$_}=1} @separate; # how many different ones we have
	close IN;

print "Totals\n", map {"$_\t".$type->{$_}."\n"} keys %$type;
print "\n\nFeature\t#pegs with feature\tTotal occurences\tUnique occurences\n";
foreach my $id (keys %$count)
	print join("\t", $id, scalar(keys %{$count->{$id}}), $total->{$id}, scalar(keys %{$all->{$id}})), "\n";

MCS Webmaster
ViewVC Help
Powered by ViewVC 1.0.3