Bio::SeqIO tinyseq
SummaryIncluded librariesPackage variablesSynopsisDescriptionGeneral documentationMethods
Bio::SeqIO::tinyseq - reading/writing sequences in NCBI TinySeq format
Package variables
No package variables defined.
Included modules
Do not use this module directly; use the SeqIO handler system:
  $stream = Bio::SeqIO->new( -file => $filename, -format => 'tinyseq' );
while ( my $seq = $stream->next_seq ) { .... }
This object reads and writes Bio::Seq objects to and from TinySeq XML
format. A TinySeq is a lightweight XML file of sequence information,
analgous to FASTA format.
See for the DTD.
No description
No description
No description
Methods description
next_seqcode    nextTop
  Title		: next_seq
Usage : $seq = $stream->next_seq()
Function : returns the next sequence in the stream
Returns : Bio::Seq object
Args : NONE
  Title		: write_seq
Usage : $seq = $stream->write_seq(@sequence_objects); undef $stream
Function : outputs one or more sequence objects as TinySeq XML
Returns : 1 on success
Args : one or more sequence objects as TinySeq XML
Because the TSeq dtd includes closing tags after all sets are written,
the output will not be complete until the program terminates or the
object is forced out of scope (see close_writer()). May not perfectly
reproduce TSeq_sid element for all sequences
  Title		: _get_seqs
Usage : Internal function - use next_seq() instead
Function : parses the XML and creates Bio::Seq objects
Returns : 1 on success
Args : NONE
Currently stores all sequence objects into memory. I will work on do
more of a stream-based approach
  Title		: _get_species
Usage : Internal function
Function : gets a Bio::Species object from cache or creates as needed
Returns : a Bio::Species object on success, undef on failure
Args : a classification string (eg 'Homo sapiens') and
a NCBI taxon id (optional)
Objects are cached for parsing multiple sequence files.
  Title		: _create_species
Usage : Internal function
Function : creates a Bio::Species object
Returns : a Bio::Species object on success, undef on failure
Args : a classification string (eg 'Homo sapiens') and
a NCBI taxon id (optional)
  Title		: _assign_identifier
Usage : Internal function
Function : looks for sequence accession
Returns : 1 on success
Args : NONE
NCBI puts refseq accessions in TSeq_sid, others in TSeq_accver.
  Title		: _convert_seqtype
Usage : Internal function
Function : maps Bio::Seq::alphabet() values [dna/rna/protein] onto
TSeq_seqtype values [protein/nucleotide]
  Title		: _get_idstring
Usage : Internal function
Function : parse accession and version info from TSeq_accver
or TSeq_sid
  Title		: _get_writer
Usage : Internal function
Function : instantiate XML::Writer object if needed,
output initial XML
  Title		: close_writer
Usage : $self->close_writer()
Function : terminate XML output
Args : NONE
Returns : 1 on success
Called automatically by DESTROY when object goes out of scope
Methods code
sub _initialize {
    my ($self, @args) = @_;


    unless (defined $self->sequence_factory) {

    $self->{'_species_objects'} = {};
    $self->{_parsed} = 0;
sub next_seq {
    my ($self) = @_;

    $self->_get_seqs() unless ($self->{_parsed});

    return shift @{$self->{_seqlist}};
sub write_seq {
    my ($self, @seqobjs) = @_;

    $self->throw('write_seq must be called with at least one  Bio::SeqI or Bio::PrimarySeqI compliant object')
	unless (@seqobjs and ( $seqobjs[0]->isa('Bio::SeqI') || $seqobjs[0]->isa('Bio::PrimarySeqI')));

    my $writer = $self->_get_writer;

    foreach my $seqobj (@seqobjs) {
	my ($id_element, $id_value) = $self->_get_idstring($seqobj);
	$writer->emptyTag('TSeq_seqtype', value => $self->_convert_seqtype($seqobj));
	$writer->dataElement('TSeq_gi', $seqobj->primary_id || '');
	$writer->dataElement($id_element, $id_value);
	#$writer->dataElement('TSeq_orgname', $seqobj->taxid) if ($seqobj->can('taxid'); # just a placeholder
$writer->dataElement('TSeq_defline', $seqobj->desc); $writer->dataElement('TSeq_length', $seqobj->length); $writer->dataElement('TSeq_sequence', $seqobj->seq); if ($seqobj->can('species') && $seqobj->species) { $self->_write_species($writer, $seqobj->species); } $writer->endTag('TSeq'); } 1;
sub _get_seqs {
    my ($self) = @_;

    my $fh = $self->_fh;

    my $handler = Bio::SeqIO::tinyseq::tinyseqHandler->new();
    my $parser = XML::Parser::PerlSAX->new( Handler => $handler );

    my @seqatts = $parser->parse( Source => { ByteStream => $fh });

    my $factory = $self->sequence_factory;

    $self->{_seqlist} ||= [];
    foreach my $seqatt(@seqatts) {
	foreach my $subatt(@$seqatt) { # why are there two hashes?
my $seqobj = $factory->create(%$subatt); $self->_assign_identifier($seqobj, $subatt); if ($seqobj->can('species')) { # my $class = [reverse(split(/ /, $subatt->{'-organism'}))];
# my $species = Bio::Species->new( -classification => $class,
# -ncbi_taxid => $subatt->{'-taxid'} );
my $species = $self->_get_species($subatt->{'-organism'}, $subatt->{'-taxid'}); $seqobj->species($species) if ($species); } push(@{$self->{_seqlist}}, $seqobj); } } $self->{_parsed} = 1;
sub _get_species {
     my ($self, $orgname, $taxid) = @_;

     unless ($self->{'_species_objects'}->{$orgname}) {
	 my $species = $self->_create_species($orgname, $taxid);
	 $self->{'_species_objects'}->{$orgname} = $species;
     return $self->{'_species_objects'}->{$orgname};
sub _create_species {
    my ($self, $orgname, $taxid) = @_;
    return unless ($orgname); # not required in TinySeq dtd so don't throw an error
my %params; $params{'-classification'} = [reverse(split(/ /, $orgname))]; $params{'-ncbi_taxid'} = $taxid if ($taxid); my $species = Bio::Species->new(%params) or return; return $species;
sub _assign_identifier {
    my ($self, $seqobj, $atts) = @_;
    my ($accession, $version);

   if ($atts->{'-accver'}) {
	($accession, $version) = split(/\./, $atts->{'-accver'});;
    elsif ($atts->{'-sid'}) {
	my $sidstring =$atts->{'-sid'};
	$sidstring =~ s/^.+?\|//;
	$sidstring =~ s/\|[^\|]*//;
	($accession, $version) = split(/\./, $sidstring);;
    else {
	$self->throw('NO accession information found for this sequence');
    $seqobj->accession_number($accession) if ($seqobj->can('accession_number'));
    $seqobj->version($version) if ($seqobj->can('version'));
sub _convert_seqtype {
    my ($self, $seqobj) = @_;

    return 'protein' 	if ($seqobj->alphabet eq 'protein');
    return 'nucleotide' if ($seqobj->alphabet eq 'dna');
    return 'nucleotide' if ($seqobj->alphabet eq 'rna');

    # if we get here there's a problem!
$self->throw("Alphabet not defined, can't assign type for $seqobj");
sub _get_idstring {
    # NCBI puts refseq ids in TSeq_sid, others in TSeq_accver.  No idea why.
my ($self, $seqobj) = @_; my $accver = $seqobj->accession_number; $accver .= '.' . $seqobj->version if ($seqobj->can('version') and $seqobj->version); if ($accver =~ /^(NM_|NP_|XM_|XP_|NT_|NC_|NG_)/) { return ('TSeq_sid', join('|', 'ref', $accver, '')); } else { return ('TSeq_accver', $accver); }
sub _get_writer {
    # initialize writer, start doc so write_seq can work one at a time
my ($self) = @_; unless ($self->{_writer}) { my $fh = $self->_fh; my $writer = XML::Writer->new(OUTPUT => $fh, DATA_MODE => 1, DATA_INDENT => 2, NEWLINE => 1, ); $writer->doctype('TSeqSet', '-//NCBI//NCBI TSeq/EN', ''); $writer->comment("Generated by Bio::SeqIO::tinyseq VERSION ".$Bio::SeqIO::tinyseq::VERSION); $writer->startTag('TSeqSet'); $self->{_writer} = $writer; } return $self->{_writer};
sub close_writer {
    # close out any dangling writer
my ($self) = @_; if ($self->{_writer}) { my $writer = $self->{_writer}; $writer->endTag('TSeqSet'); $writer->end; undef $writer; } close($self->_fh) if ($self->_fh); 1;
sub _write_species {
    my ($self, $writer, $species) = @_;
    $writer->dataElement('TSeq_orgname', $species->binomial);
    $writer->dataElement('TSeq_taxid', $species->ncbi_taxid)
    # primarily to close out a writer!
my ($self) = @_; $self->close_writer; undef $self; } 1; __END__
General documentation
Mailing ListsTop
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.                  - General discussion - About the mailing lists
Support Top
Please direct usage questions or support issues to the mailing list:
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.
Reporting BugsTop
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:
Bio::SeqIO, Bio::Seq.
Donald Jackson, <>
Parts of this module and the test script were patterned after Sheldon
McKay's Bio::SeqIO::game. If it breaks, however, it's my fault not
his ;).
The rest of the documentation details each of the object methods.
Internal methods are usually preceded with a _