added Juan Tena's split_wig.pl script, which is useful for uploading …

…WIG files from genomes with lots of scaffolds.

Changes

@@ -7,6 +7,8 @@
    optionally pass font names like "Helvetica-12:Italic" to the track "font" option). 
  * Clicking on the "Go" button when an annotation plugin is selected now turns on the corresponding track.
  * Clicking on the "Go" button when a filter plugin is selected opens up the configuration dialog.
+ * Added Juan Tena's split_wig.pl script, which is useful for uploading WIG files from genomes with lots
+   of scaffolds.
 
 2.54
  * Version 2.53 introduced a bad bug into track configuration such

MANIFEST

@@ -26,6 +26,7 @@ bin/process_ncbi_human.pl
 bin/process_sgd.pl
 bin/report_missing_language_tags.pl
 bin/scan_gbrowse.pl
+bin/split_wig.pl
 bin/ucsc_genes2gff.pl
 bin/wiggle2gff3.pl
 Build.PL

bin/split_wig.pl

@@ -0,0 +1,74 @@
+#!/usr/bin/perl -w
+use strict;
+use warnings;
+use File::Temp qw/tempdir/;
+use Getopt::Long;
+
+##
+##  Splits a wig file (variable or fixed step format) in different wig files with a maximum of 900 scaffolds/each
+##  and runs the wiggel2gff.pl script to upload these files to GBrowse2.
+##  Usage:  split_wig.pl -w FILE.wig -p DATABASE_PATH
+##  The whole path is needed, since the gff files will point to their respective wib files.
+##  After running this script, you can run it again for a different wig file, and all gff files will be pooled
+##  together in the same folder. To upload the data to GBrowse2, the MySQL Backend is recommended:
+##  bp_seqfeature_load.pl -f -a DBI::mysql -d DATABASE gff3_files/*.gff3
+##  The data track should be configured in your DATABASE.conf file, setting the 'feature' field with the name of
+##  your original wig file (without extension).
+##
+##  Juan J. Tena, CABD 2013
+##  jjtenagu@upo.es
+##
+
+my ($wig,$path)=('','');
+GetOptions
+(
+    "w=s" => \$wig,
+    "p=s" => \$path,
+);
+
+
+if (!$wig || !$path) {die "Usage: split_wig.pl -w FILE.wig -p DATABASE_PATH\n";}
+
+mkdir "$path/wib_files";
+mkdir "$path/gff3_files";
+
+my $count=0;
+my $chr_old='';
+my $dir=tempdir(CLEANUP => 1);
+my $out=File::Temp->new(DIR => $dir, UNLINK => 0, SUFFIX => '.dat');
+my $header=`head -n 1 $wig`;
+open IN, $wig or die "Cannot open $wig: $!\n";
+while (<IN>) {
+    my $line=$_;
+    chomp $line;
+    if ($line=~/chrom/) {
+        if ($count>=900) {
+            $out=File::Temp->new(DIR => $dir, UNLINK => 0, SUFFIX => '.dat');
+            print $out $header;
+            $count=0;
+        }
+        my @fields=split /\s/,$line;
+        my $chr=$fields[1];
+        $chr=~s/chrom=//;
+        if ($chr ne $chr_old) {
+            $count++;            
+        }
+        $chr_old=$chr;
+    }
+    print $out "$line\n";
+}
+close IN;
+
+my @files=<$dir/*.dat>;
+my @filepath=split /\//,$wig;
+my @filename=split /\./,$filepath[-1];
+my $suf=1;
+foreach (@files) {
+    my $tmpout=File::Temp->new();
+    my $outfile="$path/gff3_files/$filename[0]_$suf.gff3";
+    system("wiggle2gff3.pl --path=$path/wib_files $_ > $tmpout");
+    system ("sed 's/microarray_oligo/$filename[0]/' $tmpout > $outfile");
+    $suf++;
+}
+
+exit;

bin/wiggle2gff3.pl

@@ -155,6 +155,14 @@ =head2 Example WIG File
  chr19	example	example	59304701	59308020	.	.	.	Name=variableStep;wigfile=/var/gbrowse/db/track002.chr19.1199828298.wig
  chr19	example	example	59307401	59310400	.	.	.	Name=fixedStep;wigfile=/var/gbrowse/db/track003.chr19.1199828298.wig
 
+=head1 PROBLEMS
+
+This script has trouble with wig files from very fragmented genomes
+(>100K scaffolds). In this case, you may wish to run split_wig.pl,
+which splits the original wig file into a series of smaller files with
+a maximum of 900 scaffolds each. It then runs wiggle2gff3.pl for each
+subfile and stores the results in separate folders.
+
 =head1 SEE ALSO
 
 L<Bio::DB::GFF>, L<bp_bulk_load_gff.pl>, L<bp_fast_load_gff.pl>,