view release on metacpan or  search on metacpan

Bio/Search/HSP/PullHSPI.pm  view on Meta::CPAN

I<bioperl-l@bioperl.org>

rather than to the module maintainer directly. Many experienced and 
reponsive experts will be able look at the problem and quickly 
address it. Please include a thorough description of the problem 
with code and data examples if at all possible.

=head2 Reporting Bugs

Report bugs to the Bioperl bug tracking system to help us keep track
of the bugs and their resolution. Bug reports can be submitted via the
web:

  https://github.com/bioperl/bioperl-live/issues

=head1 AUTHOR - Sendu Bala

Email bix@sendu.me.uk

=head1 COPYRIGHT

Copyright (c) 2006 Sendu Bala. All Rights Reserved.

=head1 DISCLAIMER

This software is provided "as is" without warranty of any kind.

=head1 APPENDIX

The rest of the documentation details each of the object methods.
Internal methods are usually preceded with a _

=cut

# Let the code begin...

package Bio::Search::HSP::PullHSPI;


use strict;

use base qw(Bio::Search::HSP::HSPI Bio::PullParserI);

=head2 _setup

 Title   : _setup
 Usage   : $self->_setup(@args)
 Function: Implementers should call this to setup common fields and deal with
           common arguments to new().
 Returns : n/a
 Args    : @args received in new().

=cut

sub _setup {
    my ($self, @args) = @_;
	
	# fields most subclasses probably will want
	$self->_fields( { ( hsp_length => undef,
                        identical => undef,
                        percent_identity => undef,
                        conserved => undef,
                        hsp_gaps => undef,
                        query_gaps => undef,
                        hit_gaps => undef,
						evalue => undef,
						pvalue => undef,
						score => undef,
						query_start => undef,
						query_end => undef,
						query_string => undef,
						hit_start => undef,
						hit_end => undef,
						hit_string => undef,
						homology_string => undef,
						rank => undef,
                        seq_inds => undef,
                        hit_identical_inds => undef,
                        hit_conserved_inds => undef,
                        hit_nomatch_inds => undef,
                        hit_gap_inds => undef,
                        query_identical_inds => undef,
                        query_conserved_inds => undef,
                        query_nomatch_inds => undef,
                        query_gap_inds => undef ) } );
	
	my ($parent, $chunk, $hsp_data) = $self->_rearrange([qw(PARENT
														    CHUNK
															HSP_DATA)], @args);
	
    $self->throw("Need -parent or -chunk to be defined") unless defined $parent || $chunk;
    
	$self->parent($parent) if $parent;
    
    if ($chunk) {
        my ($io, $start, $end) = (undef, 0, undef);
        if (ref($chunk) eq 'ARRAY') {
            ($io, $start, $end) = @{$chunk};
        }
        else {
            $io = $chunk;
        }
        $self->chunk($io, -start => $start, -end => $end);
    }
    
	$self->_raw_hsp_data($hsp_data) if $hsp_data;
	
    return $self;
}

sub _raw_hsp_data {
	my $self = shift;
	if (@_) {
		$self->{_raw_hsp_data} = shift;
	}
	return $self->{_raw_hsp_data};
}

#
# Some of these methods are written explitely to avoid HSPI throwing not
# implemented or the wrong ancestor class being used to answer the method;

Bio/Search/HSP/PullHSPI.pm  view on Meta::CPAN

 Usage   : my $homo_string = $hsp->homology_string;
 Function: Retrieves the homology sequence for this HSP as a string.
         : The homology sequence is the string of symbols in between the 
         : query and hit sequences in the alignment indicating the degree
         : of conservation (e.g., identical, similar, not similar).
 Returns : string
 Args    : none

=cut

sub homology_string {
	return shift->get_field('homology_string');
}

=head2 length

 Title    : length
 Usage    : my $len = $hsp->length( ['query'|'hit'|'total'] );
 Function : Returns the length of the query or hit in the alignment (without gaps) 
            or the aggregate length of the HSP (including gaps;
            this may be greater than either hit or query )
 Returns  : integer
 Args     : 'query' = length of query seq (without gaps)
            'hit'   = length of hit seq (without gaps)
            'total' = length of alignment (with gaps)
            default = 'total' 
 Args    : none

=cut

sub length {
    my ($self, $type) = @_;
    $type = 'total' unless defined $type;
    $type = lc $type;

    if ($type =~ /^q/i) {
        return $self->query->length;
    }
	elsif ($type =~ /^(hit|subject|sbjct)/) {
        return $self->hit->length;
    }
	else { 
        return $self->hit->length + $self->gaps('hit');
	}
}

=head2 hsp_length

 Title   : hsp_length
 Usage   : my $len = $hsp->hsp_length()
 Function: shortcut  length('hsp')
 Returns : floating point between 0 and 100 
 Args    : none

=cut

sub hsp_length {
	return shift->length('total');
}

=head2 percent_identity

 Title   : percent_identity
 Usage   : my $percentid = $hsp->percent_identity()
 Function: Returns the calculated percent identity for an HSP
 Returns : floating point between 0 and 100 
 Args    : none

=cut

sub percent_identity{
	my ($self) = @_;
	return $self->frac_identical('hsp') * 100;   
}

=head2 get_aln

 Title   : get_aln
 Usage   : my $aln = $hsp->get_aln
 Function: Returns a Bio::SimpleAlign representing the HSP alignment
 Returns : Bio::SimpleAlign
 Args    : none

=cut

sub get_aln {
	my $self = shift;
	
    require Bio::LocatableSeq;
    require Bio::SimpleAlign;
    my $aln = Bio::SimpleAlign->new();
    my $hs = $self->seq('hit');
    my $qs = $self->seq('query');
	if ($hs && $qs) {
		$aln->add_seq($hs);
		$aln->add_seq($qs);
		return $aln;
	}
	return;
}

=head2 seq_inds

 Title   : seq_inds
 Purpose   : Get a list of residue positions (indices) for all identical 
           : or conserved residues in the query or sbjct sequence.
 Example   : @s_ind = $hsp->seq_inds('query', 'identical');
           : @h_ind = $hsp->seq_inds('hit', 'conserved');
           : @h_ind = $hsp->seq_inds('hit', 'conserved', 1);
 Returns   : List of integers 
           : May include ranges if collapse is true.
 Argument  : seq_type  = 'query' or 'hit' or 'sbjct'  (default = query)
              ('sbjct' is synonymous with 'hit') 
             class     = 'identical' or 'conserved' or 'nomatch' or 'gap'
                          (default = identical)
                          (can be shortened to 'id' or 'cons')
		                  Note that 'conserved' includes identical unless you
		                  use 'conserved-not-identical'

             collapse  = boolean, if true, consecutive positions are merged
                         using a range notation, e.g., "1 2 3 4 5 7 9 10 11" 
                         collapses to "1-5 7 9-11". This is useful for 
                         consolidating long lists. Default = no collapse.
 Throws    : n/a.
 Comments  : 

See Also   : L<Bio::Search::BlastUtils::collapse_nums()|Bio::Search::BlastUtils>, L<Bio::Search::Hit::HitI::seq_inds()|Bio::Search::Hit::HitI>

=cut

sub seq_inds {