[Bioperl-guts-l] bioperl-live/Bio/SearchIO hmmer_pull.pm,NONE,1.1

Senduran Balasubramaniam sendu at dev.open-bio.org
Sun Aug 20 17:52:49 EDT 2006

Update of /home/repository/bioperl/bioperl-live/Bio/SearchIO
In directory dev.open-bio.org:/tmp/cvs-serv10511/Bio/SearchIO

Added Files:
Log Message:
New PullParserI and hmmer parser

--- NEW FILE: hmmer_pull.pm ---
# $Id: hmmer_pull.pm,v 1.1 2006/08/20 21:52:47 sendu Exp $
# BioPerl module for Bio::SearchIO::hmmer_pull
# Cared for by Sendu Bala <bix at sendu.me.uk>
# Copyright Sendu Bala
# You may distribute this module under the same terms as perl itself

# POD documentation - main docs before the code

=head1 NAME

Bio::SearchIO::hmmer_pull - A parser for HMMER output


    # do not use this class directly it is available through Bio::SearchIO
    use Bio::SearchIO;
    my $in = new Bio::SearchIO(-format => 'hmmer_pull',
                               -file   => 't/data/hmmpfam.bigout');
    while (my $result = $in->next_result) {
        # this is a Bio::Search::Result::HmmpfamResult object
        print $result->query_name(), " for HMM ", $result->hmm_name(), "\n";
        while (my $hit = $result->next_hit) {
            print $hit->name(), "\n";
            while (my $hsp = $hit->next_hsp) {
                print "length is ", $hsp->length(), "\n";


This object implements a pull-parser for HMMER output. It is fast since it
only does work on request (hence 'pull').


=head2 Mailing Lists

User feedback is an integral part of the evolution of this and other
Bioperl modules. Send your comments and suggestions preferably to
the Bioperl mailing list.  Your participation is much appreciated.

  bioperl-l at bioperl.org                  - General discussion
  http://bioperl.org/wiki/Mailing_lists  - About the mailing lists

=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


=head1 AUTHOR - Sendu Bala

Email bix at sendu.me.uk


Additional contributors names and emails here


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


# Let the code begin...

package Bio::SearchIO::hmmer_pull;

use vars qw(@ISA);
use strict;

use Bio::SearchIO;
use Bio::PullParserI;

@ISA = qw(Bio::SearchIO Bio::PullParserI);

=head2 new

 Title   : new
 Usage   : my $obj = new Bio::SearchIO::hmmer_pull();
 Function: Builds a new Bio::SearchIO::hmmer_pull object 
 Returns : Bio::SearchIO::hmmer_pull
 Args    : -fh/-file => HMMER output filename
           -format   => 'hmmer_pull'
           -evalue   => float or scientific notation number to be used
                        as an evalue cutoff for hits
           -score    => integer or scientific notation number to be used
                        as a score value cutoff for hits
           -hsps     => integer minimum number of hsps (domains) a hit must have
           -piped_behaviour => 'temp_file'|'memory'|'sequential_read'
           -piped_behaviour defines what the parser should do if the input is
            an unseekable filehandle (eg. piped input), see
            Bio::PullParserI::chunk for details. Default is 'memory'.


sub _initialize {
    my ($self, @args) = @_;
    # don't do normal SearchIO initialization
    my ($writer, $file, $fh, $piped_behaviour, $evalue, $score, $hsps) =
                                                  FILE FH
                                                  HSPS)], @args);
    $self->writer($writer) if $writer;
    $self->_fields( { ( header => undef,
                        algorithm => undef,
                        algorithm_version => undef,
                        algorithm_reference => '',
                        hmm_file => undef,
                        hmm_name => undef,
                        sequence_file => undef,
                        sequence_database => undef,
                        database_name => undef,
                        database_letters => undef,
                        database_entries => undef,
                        next_result => undef,
                        evalue_cutoff => '[unset]',
                        score_cutoff => '[unset]',
                        hsps_cutoff => '[unset]' ) } );
    $self->_fields->{evalue_cutoff} = $evalue if $evalue;
    $self->_fields->{score_cutoff} = $score if $score;
    $self->_fields->{hsps_cutoff} = $hsps if $hsps;
    $self->_dependencies( { ( algorithm => 'header',
                              algorithm_version => 'header',
                              hmm_file => 'header',
                              hmm_name => 'header',
                              sequence_file => 'header',
                              sequence_database => 'header' ) } );
    $self->chunk($file || $fh || $self->throw("-file or -fh must be supplied"),
                 -piped_behaviour => $piped_behaviour || 'memory');

sub _discover_header {
    my $self = shift;
    my $header = $self->_get_chunk_by_nol(8);
    $self->{_after_header} = $self->_chunk_tell;
    my ($algo) = $header =~ /^(hmm\S+) - search/m;
    $self->_fields->{algorithm} = uc $algo;
    ($self->_fields->{algorithm_version}) = $header =~ /^HMMER\s+?(\S+)/m;
    ($self->_fields->{hmm_file}) = $header =~ /^HMM file:\s.+?(\S+)$/m;
    $self->_fields->{hmm_name} = $self->_fields->{hmm_file};
    ($self->_fields->{sequence_file}) = $header =~ /^Sequence (?:file|database):\s.+?(\S+)$/m;
    $self->_fields->{sequence_database} = $self->_fields->{sequence_file};
    $self->_fields->{header} = 1;

sub _discover_database_name {
    my $self = shift;
    my $type = $self->get_field('algorithm');
    if ($type eq 'HMMPFAM') {
        $self->_fields->{database_name} = $self->get_field('hmm_file');
    elsif ($type eq 'HMMSEARCH') {
        $self->_fields->{database_name} = $self->get_field('sequence_file');

sub _discover_next_result {
    my $self = shift;
    my $type = $self->get_field('algorithm'); # also sets _after_header if not set
    if ($type eq 'HMMPFAM') {
        use Bio::Search::Result::HmmpfamResult;
        unless ($self->_sequential) {
            if (defined $self->{_end_of_previous_result}) {
            else {
            my ($start, $end) = $self->_find_chunk_by_end("//\n");
            return if $start == $end;
            $self->_fields->{next_result} = new Bio::Search::Result::HmmpfamResult(-chunk => [($self->chunk, $start, $end)],
                                                                                   -parent => $self);
            $self->{_end_of_previous_result} = $end;
        else {
            my $chunk = $self->_get_chunk_by_end("//\n");
            $chunk || return;
            $self->_fields->{next_result} = new Bio::Search::Result::HmmpfamResult(-chunk => [$chunk],
                                                                                   -parent => $self);
    elsif ($type eq 'HMMSEARCH') {
        $self->throw("Can't handle hmmsearch yet\n");
    else {
        $self->throw("Unknown report type");

=head2 next_result

 Title   : next_result
 Usage   : my $hit = $searchio->next_result;
 Function: Returns the next Result from a search
 Returns : Bio::Search::Result::ResultI object
 Args    : none


sub next_result {
    my $self = shift;
    my $result = $self->get_field('next_result') || return;
    undef $self->_fields->{next_result};
    return $result;

=head2 result_count

 Title   : result_count
 Usage   : my $count = $searchio->result_count
 Function: Returns the number of results we have processed.
 Returns : integer
 Args    : none


sub result_count {
    my $self = shift;
    return $self->{'_result_count'};

=head2 rewind

 Title   : rewind
 Usage   : $searchio->rewind;
 Function: Allow one to reset the Result iterator to the beginning, so that
           next_result() will subsequently return the first result and so on.
           NB: result objects are not cached, so you will get new result objects
           each time you rewind. Also, note that result_count() counts the
           number of times you have called next_result(), so will not be able
           tell you how many results there were in the file if you use rewind().
 Returns : n/a
 Args    : none


sub rewind {
	my $self = shift;
	delete $self->{_end_of_previous_result};


More information about the Bioperl-guts-l mailing list