lib/Bio/Tools/RepeatMasker.pm

   1 #
   2 # BioPerl module for Bio::Tools::RepeatMasker
   3 #
   4 # Please direct questions and support issues to <bioperl-l@bioperl.org>
   5 #
   6 # Cared for by Shawn Hoon <shawnh@fugu-sg.org>
   7 #
   8 # Copyright Shawn Hoon
   9 #
  10 # You may distribute this module under the same terms as perl itself
  11
  12 # POD documentation - main docs before the code
  13
  14 =head1 NAME
  15
  16 Bio::Tools::RepeatMasker - a parser for RepeatMasker output
  17
  18 =head1 SYNOPSIS
  19
  20     use Bio::Tools::RepeatMasker;
  21     my $parser = Bio::Tools::RepeatMasker->new(-file => 'seq.fa.out');
  22     while( my $result = $parser->next_result ) {
  23       # get some value
  24     }
  25
  26 =head1 DESCRIPTION
  27
  28 A parser for RepeatMasker output
  29
  30 =head1 FEEDBACK
  31
  32 =head2 Mailing Lists
  33
  34 User feedback is an integral part of the evolution of this and other
  35 Bioperl modules. Send your comments and suggestions preferably to
  36 the Bioperl mailing list.  Your participation is much appreciated.
  37
  38   bioperl-l@bioperl.org                  - General discussion
  39   http://bioperl.org/wiki/Mailing_lists  - About the mailing lists
  40
  41 =head2 Support
  42
  43 Please direct usage questions or support issues to the mailing list:
  44
  45 I<bioperl-l@bioperl.org>
  46
  47 rather than to the module maintainer directly. Many experienced and
  48 reponsive experts will be able look at the problem and quickly
  49 address it. Please include a thorough description of the problem
  50 with code and data examples if at all possible.
  51
  52 =head2 Reporting Bugs
  53
  54 Report bugs to the Bioperl bug tracking system to help us keep track
  55 of the bugs and their resolution. Bug reports can be submitted via
  56 the web:
  57
  58   https://github.com/bioperl/bioperl-live/issues
  59
  60 =head1 AUTHOR - Shawn Hoon
  61
  62 Email shawnh@fugu-sg.org
  63
  64 =head1 APPENDIX
  65
  66 The rest of the documentation details each of the object methods.
  67 Internal methods are usually preceded with a _
  68
  69 =cut
  70
  71
  72 # Let the code begin...
  73
  74
  75 package Bio::Tools::RepeatMasker;
  76 use strict;
  77
  78 use Bio::SeqFeature::FeaturePair;
  79
  80 use base qw(Bio::Root::Root Bio::Root::IO);
  81
  82 =head2 new
  83
  84  Title   : new
  85  Usage   : my $obj = Bio::Tools::RepeatMasker->new();
  86  Function: Builds a new Bio::Tools::RepeatMasker object
  87  Returns : Bio::Tools::RepeatMasker
  88  Args    : -fh/-file => $val, for initing input, see Bio::Root::IO
  89
  90 =cut
  91
  92 sub new {
  93   my($class,@args) = @_;
  94
  95   my $self = $class->SUPER::new(@args);
  96   $self->_initialize_io(@args);
  97
  98   return $self;
  99 }
 100
 101 =head2 next_result
 102
 103  Title   : next_result
 104  Usage   : my $r = $rpt_masker->next_result
 105  Function: Get the next result set from parser data
 106  Returns : Bio::SeqFeature::FeaturePair
 107            Feature1 is the Query coordinates and Feature2 is the Hit
 108  Args    : none
 109
 110 =cut
 111
 112 sub next_result {
 113     my ($self) = @_;
 114     local $_;
 115     while (defined($_=$self->_readline()) ) {
 116         if (/no repetitive sequences detected/) {
 117             $self->warn( "RepeatMasker didn't find any repetitive sequences\n");
 118             return ;
 119         }
 120         #ignore introductory lines
 121         if (/\d+/) {
 122             my @element = split;
 123             # ignore features with negatives
 124             next if ($element[11-13] =~ /-/);
 125             my (%feat1, %feat2);
 126             my @line = split;
 127             my ($score, $query_name, $query_start, $query_end, $strand,
 128                 $repeat_name, $repeat_class ) = @line[0, 4, 5, 6, 8, 9, 10];
 129
 130             my ($hit_start,$hit_end);
 131
 132             if ($strand eq '+') {
 133                 ($hit_start, $hit_end) = @line[11, 12];
 134                 $strand = 1;
 135             } elsif ($strand eq 'C') {
 136                 ($hit_end, $hit_start) = @line[12, 13];
 137                 $strand = -1;
 138             }
 139             my $rf = Bio::SeqFeature::Generic->new
 140                 (-seq_id      => $query_name,
 141                  -score       => $score,
 142                  -start       => $query_start,
 143                  -end         => $query_end,
 144                  -strand      => $strand,
 145                  -source_tag  => 'RepeatMasker',
 146                  -primary_tag => $repeat_class,
 147                  -tag => { 'Target'=> [$repeat_name, $hit_start, $hit_end]},
 148                 );
 149
 150             my $rf2 = Bio::SeqFeature::Generic->new
 151                 (-seq_id         => $repeat_name,
 152                  -score          => $score,
 153                  -start          => $hit_start,
 154                  -end            => $hit_end,
 155                  -strand         => $strand,
 156                  -source_tag     => "RepeatMasker",
 157                  -primary_tag    => $repeat_class,
 158                  -tag => { 'Target'=> [$query_name,$query_start,$query_end] },
 159                 );
 160
 161             my $fp = Bio::SeqFeature::FeaturePair->new(-feature1 => $rf,
 162                                                        -feature2 => $rf2);
 163             return $fp;
 164         }
 165     }
 166 }
 167
 168 1;