| Bio-Graphics documentation | Contained in the Bio-Graphics distribution. |
Bio::Graphics::Glyph::hybrid_plot - An xyplot plot drawing dual graph using data from two wiggle files per track
See <Bio::Graphics::Panel> <Bio::Graphics::Glyph> and <Bio::Graphics::Glyph::wiggle_xyplot>.
Note that for full functionality this glyph requires Bio::Graphics::Glyph::generic (generic glyph is used for drawing individual matches for small RNA alignments at a high zoom level, specified by semantic zooming in GBrowse conf file) Unlike the regular xyplot, this glyph draws two overlapping graphs using value data in Bio::Graphics::Wiggle file format:
track type=wiggle_0 name="Experiment" description="snRNA seq data" visibility=pack viewLimits=-2:2 color=255,0,0 altColor=0,0,255 windowingFunction=mean smoothingWindow=16
2L 400 500 0.5 2L 501 600 0.5 2L 601 700 0.4 2L 701 800 0.1 2L 800 900 0.1
##gff-version 3
2L Sample_rnaseq rnaseq_wiggle 41 3009 . . . ID=Samlpe_2L;Name=Sample;Note=YourNoteHere;wigfileA=/datadir/track_001.2L.wig;wigfileB=/datadir/track_002.2L.wig
The "wigfileA" and "wigfileB" attributes give a relative or absolute pathname to Bio::Graphics::Wiggle format files for two concurrent sets of data. Basically, these wigfiles contain the data on signal intensity (counts) for sequences aligned with genomic regions. In wigfileA these data are additive, so for each sequence region the signal is calculated as a sum of signals from overlapping matches (signal). In wigfileB the signal represents the maximum value among all sequences (signal quality) aligned with the current region so the user can see the difference between accumulated signal from overlapping multiple matches (which may likely be just a noise from products of degradation) and high-quality signal from unique sequences.
It is essential that wigfile entries in gff file do not have score, because score used to differentiate between data for dual graph and data for matches (individual features visible at higher magnification). After an update to wiggle_xyplot code colors for dual plot are now hard-coded (blue for signal and orange for signal quality). Alpha channel is also handled by wiggle_xyplot code now.
In addition to some of the wiggle_xyplot glyph options, the following options are recognized:
Name Value Description
---- ----- -----------
wigfileA path name Path to a Bio::Graphics::Wiggle file for accumulated vales in 10-base bins
wigfileB path name Path to a Bio::Graphics::Wiggle file for max values in 10-base bins
fasta path name Path to fasta file to enable BigWig drawing
u_method method name Use method of [method name] to identify individual features (like alignment matches)
to show at high zoom level. By default it is set to 'match'
Please report them.
Bio::Graphics::Panel, Bio::Graphics::Glyph, Bio::Graphics::Glyph::arrow, Bio::Graphics::Glyph::cds, Bio::Graphics::Glyph::crossbox, Bio::Graphics::Glyph::diamond, Bio::Graphics::Glyph::dna, Bio::Graphics::Glyph::dot, Bio::Graphics::Glyph::ellipse, Bio::Graphics::Glyph::extending_arrow, Bio::Graphics::Glyph::generic, Bio::Graphics::Glyph::graded_segments, Bio::Graphics::Glyph::heterogeneous_segments, Bio::Graphics::Glyph::line, Bio::Graphics::Glyph::pinsertion, Bio::Graphics::Glyph::primers, Bio::Graphics::Glyph::rndrect, Bio::Graphics::Glyph::segments, Bio::Graphics::Glyph::ruler_arrow, Bio::Graphics::Glyph::toomany, Bio::Graphics::Glyph::transcript, Bio::Graphics::Glyph::transcript2, Bio::Graphics::Glyph::translation, Bio::Graphics::Glyph::allele_tower, Bio::DB::GFF, Bio::SeqI, Bio::SeqFeatureI, Bio::Das, GD
Peter Ruzanov <pruzanov@oicr.on.ca>.
Copyright (c) 2008 Ontario Institute for Cancer Research
This package and its accompanying libraries is free software; you can redistribute it and/or modify it under the terms of the GPL (either version 1, or at your option, any later version) or the Artistic License 2.0. Refer to LICENSE for the full license text. In addition, please see DISCLAIMER.txt for disclaimers of warranty.
| Bio-Graphics documentation | Contained in the Bio-Graphics distribution. |
package Bio::Graphics::Glyph::hybrid_plot; use strict; use base qw(Bio::Graphics::Glyph::wiggle_xyplot Bio::Graphics::Glyph::smoothing); use constant DEBUG=>0; use constant NEGCOL=>"orange"; use constant POSCOL=>"blue"; our $VERSION = '1.0'; sub my_options { { min_score => [ 'integer', undef, "Minimum value of the signal graph feature's \"score\" attribute."], max_score => [ 'integer', undef, "Maximum value of the signal graph feature's \"score\" attribute."], flip_sign => [ 'boolean', 0, "Optionally flip the signal for wigfileB (if the scores are positive but we wish to paint the signal along negative y-axis)"] }; } sub my_description { return <<END; This glyph draws signal graph (wiggle_xyplot) using wiggle or BigWig files requires a special load gff file that uses attributes 'wigfileA' and 'wigfileB' For BigWig support it is also required to have 'fasta' attribute set to point to a fasta file for your organism of interest Example: 2L test hybrid 5407 23011573 . . . Name=Experiment 1;wigfileA=SomeWigFile1.wigdb;wigfileB=SomeWigFile2.wigdb END } #Checking the method for individual features (RNA-Seq reads) sub _check_uni { return shift->option('u_method') || 'match'; } # Override height and pad functions (needed to correctly space features with different sources): sub height { my $self = shift; my $h = $self->SUPER::height; return $self->feature->method eq $self->_check_uni ? 3 : $h; } sub pad_top { my $self = shift; return $self->feature->method eq $self->_check_uni ? 0 : 4; } sub pad_bottom { my $self = shift; return $self->feature->method eq $self->_check_uni ? 0 : 4; } # we override the draw method so that it dynamically creates the parts needed # from the wig file rather than trying to fetch them from the database sub draw { my $self = shift; my ($gd,$dx,$dy) = @_; my ($left,$top,$right,$bottom) = $self->calculate_boundaries($dx,$dy); my $height = $bottom - $top; my $feature = $self->feature; my $set_flip = $self->option('flip_sign') | 0; #Draw individual features for reads (unlike wiggle features reads will have scores) my $t_id = $feature->method; if($t_id && $t_id eq $self->_check_uni){return Bio::Graphics::Glyph::generic::draw_component($self,@_);} #Draw multiple graph if we don't have a score my @wiggles = (); foreach ('A'..'Z') { my $filename = 'wigfile'.$_; my ($wiggle) = $feature->get_tag_values('wigfile'.$_); push (@wiggles, $wiggle); } my ($fasta) = $feature->get_tag_values('fasta'); my($scale,$y_origin,$min_score,$max_score); $self->panel->startGroup($gd); #Depending on what we have (wiggle or BigWig) pick the way to paint the signal graph for(my $w = 0; $w < @wiggles; $w++){ if ($w > 0) { $self->configure(-pos_color, NEGCOL); $self->configure(-no_grid, 1); } else { $self->configure(-pos_color, POSCOL); } if ($wiggles[$w] =~ /\.wi\w{1,3}$/) { $self->draw_wigfile($feature,$wiggles[$w],@_); } elsif ($wiggles[$w] =~ /\.bw$/ && $fasta) { my $flip = ($w > 0 && $set_flip) ? -1 : 1; use Bio::DB::BigWig 'binMean'; use Bio::DB::Sam; my $wig = Bio::DB::BigWig->new(-bigwig => "$wiggles[$w]", -fasta => Bio::DB::Sam::Fai->open("$fasta")); my ($summary) = $wig->features(-seq_id => $feature->segment->ref, -start => $self->panel->start, -end => $self->panel->end, -type => 'summary'); my $stats = $summary->statistical_summary($self->width); my @vals = map {$_->{validCount} ? $_->{sumData}/$_->{validCount}*$flip:0} @$stats; $self->draw_coverage($self,\@vals,@_); } } } 1; __END__