view release on metacpan or  search on metacpan

lib/Bio/SearchIO/XML/BlastHandler.pm  view on Meta::CPAN

                'Parameters_gap-open'   => { 'RESULT-parameters' => 'gapopen'},
                'Parameters_gap-extend' => { 'RESULT-parameters' => 'gapext'},
                'Parameters_filter'     => {'RESULT-parameters' => 'filter'},
                'Statistics_db-num'     => 'RESULT-database_entries',
                'Statistics_db-len'     => 'RESULT-database_letters',
                'Statistics_hsp-len'    => { 'RESULT-statistics' => 'hsplength'},
                'Statistics_eff-space'  => { 'RESULT-statistics' => 'effectivespace'},
                'Statistics_kappa'      => { 'RESULT-statistics' => 'kappa' },
                'Statistics_lambda'     => { 'RESULT-statistics' => 'lambda' },
                'Statistics_entropy'    => { 'RESULT-statistics' => 'entropy'},

                # HSP specific fields
                'Hsp_bit-score'  => 'HSP-bits',
                'Hsp_score'      => 'HSP-score',
                'Hsp_evalue'     => 'HSP-evalue',
                'Hsp_query-from' => 'HSP-query_start',
                'Hsp_query-to'   => 'HSP-query_end',
                'Hsp_hit-from'   => 'HSP-hit_start',
                'Hsp_hit-to'     => 'HSP-hit_end',
                'Hsp_positive'   => 'HSP-conserved',
                'Hsp_identity'   => 'HSP-identical',
                'Hsp_gaps'       => 'HSP-gaps',
                'Hsp_hitgaps'    => 'HSP-hit_gaps',
                'Hsp_querygaps'  => 'HSP-query_gaps',
                'Hsp_qseq'       => 'HSP-query_seq',
                'Hsp_hseq'       => 'HSP-hit_seq',
                'Hsp_midline'    => 'HSP-homology_seq',
                'Hsp_align-len'  => 'HSP-hsp_length',
                'Hsp_query-frame'=> 'HSP-query_frame',
                'Hsp_hit-frame'  => 'HSP-hit_frame',

                # Hit specific fields
                'Hit_id'               => 'HIT-name',
                'Hit_len'              => 'HIT-length',
                'Hit_accession'        => 'HIT-accession',
                'Hit_def'              => 'HIT-description',
                'Hit_num'              => 'HIT-order',
                'Iteration_iter-num'   => 'HIT-iteration',
                'Iteration_stat'       => 'HIT-iteration_statistic',

                # if these tags are present, they will overwrite the
                # above with more current data (i.e. multiquery hits)
                'Iteration_query-def'   => 'RESULT-query_description',
                'Iteration_query-len'   => 'RESULT-query_length',
                'Iteration_query-ID'    => 'runid',
               );

# these XML tags are ignored for now
my %IGNOREDTAGS = (
                'Hsp_num'              => 1,#'HSP-order',
                'Hsp_pattern-from'     => 1,#'patternend',
                'Hsp_pattern-to'       => 1,#'patternstart',
                'Hsp_density'          => 1,#'hspdensity',
                'Iteration_message'    => 1,
                'Hit_hsps'             => 1,
                'BlastOutput_param'    => 1,
                'Iteration_hits'       => 1,
                'Statistics'           => 1,
                'Parameters'           => 1,
                'BlastOutput'          => 1,
                'BlastOutput_iterations' => 1,
                   );



sub start_document{
    my ($self) = @_;
    $self->{'_lasttype'} = '';
    $self->{'_values'} = {};
    $self->{'_result'}= [];
}


sub end_document{
   my ($self,@args) = @_;

   # reset data carried throughout parse
   $self->{'_resultdata'} = undef;

   # pass back ref to results queue; caller must reset handler results queue
   return $self->{'_result'};
}


sub start_element{
    my ($self,$data) = @_;
    # we currently don't care about attributes
    my $nm = $data->{'Name'};

    if( my $type = $MODEMAP{$nm} ) {
        if( $self->eventHandler->will_handle($type) ) {
            my $func = sprintf("start_%s",lc $type);
            $self->eventHandler->$func($data->{'Attributes'});
        }
    }
}


sub end_element{
    my ($self,$data) = @_;

    my $nm = $data->{'Name'};
    my $rc;
    if($nm eq 'BlastOutput_program' &&
       $self->{'_last_data'} =~ /(t?blast[npx])/i ) {
        $self->{'_type'} = uc $1;
    }
    if ($nm eq 'Iteration') {
        map {
            $self->{'_values'}->{$_} = $self->{'_resultdata'}->{$_};
            } keys %{ $self->{'_resultdata'} };
    }
    if( my $type = $MODEMAP{$nm} ) {
        if( $self->eventHandler->will_handle($type) ) {
            my $func = sprintf("end_%s",lc $type);
            $rc = $self->eventHandler->$func($self->{'_type'},
                                              $self->{'_values'});
        }
    }
    elsif( exists $MAPPING{$nm} ) {
        if ( ref($MAPPING{$nm}) =~ /hash/i ) {