Summary

Bio::Tools::Analysis::Protein::Domcut - a wrapper around Domcut server

Package variables

Included modules

Inherit

Bio::Tools::Analysis::SimpleAnalysisBase

Synopsis

  use   Bio::Tools::Analysis::Protein::Domcut;
  #get a  Bio::PrimarySeq
  use Bio::PrimarySeq;
  my $seq = new Bio::PrimarySeq
     (-seq=>'IKLCVNLAILAKAHLIELALAL',
     -primary_id=>'test'); # a Bio::PrimarySeqI object

  my $domcut = Bio::Tools::Analysis::Protein::Domcut->new (-seq=>$seq);
  $domcut->run;
  print $domcut->result;# #raw text to standard out

Description

A module to remotely retrieve predictions of protein domain
boundaries. Each residue in the protein receives a score, those
better than the significance threshold and at a local minimum receive
a rank - i.e., the best minimum is rank 1, the second best minimum is
rank2 etc. These correspond to domain boundaries. e.g.,

  my $analysis_object = Bio::Tools::Analysis::Protein::Domcut->new
     (-seq => $seq);

creates a new object. The sequence supplied must be a Bio::PrimarySeq and not
a Bio::Seq object.

  $analysis_object->run;

submits the query to the server and obtains raw text output
Given an amino acid sequence the results can be obtained in 4 formats,
determined by the argument to the result method
    The raw text of the program output

  my $rawdata = $analysis_object->result;

    An reference to an array of hashes of scores for each state and the
assigned state. Each element in the array is a residue (indexed from 0).

  my $data_ref = $analysis_object->result('parsed');
  print "score for helix at residue 2 is $data_ref->[1]{'helix'}\n";
  print "predicted struc  at residue 2 is $data_ref->[1]{'struc}\n";

    An array of Bio::SeqFeature::Generic objects where each feature is a
predicted unit of secondary structure. Only stretches of helix/sheet
predictions for longer than 4 residues are defined as helices.
So, in order to add features to an existing Bio::Seq object;

  # get a Bio::Seq object
  my $seqobj;
  my $tool = Bio::Tools::Analysis::Protein::Domcut->new
      ( -seq => $seqobj->primary_seq);
  $tool->run;

  my @fts = $tool->result(Bio::SeqFeatureI);

  $seqobj->add_SeqFeature(@fts);

  # if you want  meta sequences as well :
  my $meta = $tool->result('meta');
  $seqobj->primary_seq($meta);

  # can access meta data in a Bio::Seq object via a 
  # call to primary_seq:

  print $seq4->primary_seq->named_submeta_text('Domcut', 1,2), "\n";

    A Bio::Seq::Meta::Array implementing sequence.
    This is a Bio::Seq object that can also hold data about each residue
in the sequence In this case, the sequence can be associated with a
single array of Domcut prediction scores. e.g.,

  my $meta_sequence = $analysis_object->result('meta');
  print "scores from residues 10 -20 are ",
      $meta_sequence->submeta_text(10,20), "\n";

    Many methods common to all analyses are inherited from
Bio::Tools::Analysis::SimpleAnalysisBase.

Methods

Methods description

 Name    : result
 Purpose : To retrieve results of analysis in one of several formats.
 Usage   : $job->result (...)
 Returns : a result created by running an analysis
 Args    : various - see keysin $RESULT_SPEC. 

Methods code

sub result {

    my ($self,$value) = @_;
    my @scores;
    my @fts;

    if ($value ) {
        # parse raw text if not already done so
        if (!exists($self->{'_parsed'})) {
            my $result = IO::String->new($self->{'_result'});
            while (my $line = <$result>) {
                next if $line =~/#/;
                $line =~/(\-?\d\.\d+)\s+(\d+)?/;
                push @scores, {score => $1,
                               rank  => ($2)?$2:'' ,
                              };
            }
            #hold parsed results in object, saves having to reparse each time
            $self->{'_parsed'} =\@ scores;
        }
        #make aarray of Bio::SeqFeature::Generic objects
        if ($value eq 'Bio::SeqFeatureI') {
            my $i = 0;          #array index (= aa num -1)
			my $in_trough = 0;
			my ($st, $end, $rank, $min_score, $min_locus) = (0,0,0,0,0);
			my $seqlen = $self->seq->length();
            for my $score (@{$self->{'_parsed'}}) {

				##start a potential trough
				if ($in_trough == 0 && $score->{'score'} < -0.09) {
					$in_trough = 1;
					$st        = $i+1;
					}

				## in a trough, is it ranked?
				elsif ( $in_trough == 1 && $score->{'score'} < -0.09 && $i +1 < $seqlen){
					if ($score->{'rank'} ) {
						$rank      = $score->{'rank'};
						$min_score = $score->{'score'};
					    $min_locus = $i + 1;
						}
				}
							
				## end of trough or end of sequence, make into feature
                ## if possible
				elsif ($in_trough == 1 && ($score->{'score'} > -0.09 ||
						 $i +1  == $seqlen) ){
					if ($rank != 0) {
                    	push @fts, Bio::SeqFeature::Generic->new (
                         	-start   => $st,
                            -end     => $i +1, #current position
                         	-primary => 'Linker',
							-source  => 'Domcut',
                         	-tag => {
                                  score   => $min_score,
                                  rank    => $rank,
								  residue => $min_locus,
                                 },
                        );
					}
					##and reset parameters ##
					($st, $in_trough, $min_locus, $min_score, $rank) = (0,0,0,0,0);
                }
                $i++;
            }
            return @fts;
        }
        ## convert parsed data into a meta array format
        elsif ($value eq 'meta') {

			## only need to bless  once
			if (! $self->seq->isa("Bio::Seq::MetaI")){
				bless ($self->seq, "Bio::Seq::Meta::Array");
				}
            $self->seq->isa("Bio::Seq::MetaI")
                || $self->throw("$self is not a Bio::Seq::MetaI");
            my $meta_name = "Domcut";

            #test that sequence does not have already a meta seq with same name
            if (grep{$_ eq $meta_name}$self->seq->meta_names ) {
                $self->warn ("$meta_name already exists , not overwriting!");
                next;
            }

            ### or should be an instance variable?? ##
            $Bio::Seq::Meta::Array::DEFAULT_NAME = 'Domcut';
            my @meta = map{$_->{'score'}} @{$self->{'_parsed'}};
            $self->seq->named_meta($meta_name,\@meta );

            # return  seq array object implementing meta sequence #
            return $self->seq;

        }
        #       return ref to array of predictions;
        elsif ($value eq 'parsed') {
            return $self->{'_parsed'};
        }
    }
    #else if no arguments return raw text
    return $self->{'_result'};

}

sub _init {

    my $self = shift;
    $self->url($URL);
    $self->{'_ANALYSIS_SPEC'} = $ANALYSIS_SPEC;
    $self->{'_INPUT_SPEC'}    = $INPUT_SPEC;
    $self->{'_RESULT_SPEC'}   = $RESULT_SPEC;
    $self->{'_ANALYSIS_NAME'} = $ANALYSIS_NAME;
    return $self;

}

sub _run {

    my $self  = shift;
    my $seq_fasta = $self->seq->seq;
    $self->delay(1);
    # delay repeated calls by default by 3 sec, set delay() to change
    $self->sleep;
    $self->status('TERMINATED_BY_ERROR');
    my $rqst = GET $self->url . "&seqnam=". "&sequence=".
                      $seq_fasta. "&outform=dat";

    my $content = $self->request($rqst);
    my $text = $content->content; #1st reponse
    $self->{'_result'} = $text;
    $self->status('COMPLETED') if $text ne '';

}

General documentation

Bio::SimpleAnalysisI,
Bio::Tools::Analysis::SimpleAnalysisBase,
Bio::Seq::Meta::Array,
Bio::WebAgent

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

  bioperl-l@bioperl.org                       - General discussion
  http://bio.perl.org/MailList.html           - About the mailing lists

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

  bioperl-bugs@bio.perl.org
  http://bugzilla.bioperl.org/

Richard Adams, Richard.Adams@ed.ac.uk,

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