Toolbar

Summary

Bio::Network::IO::psi10

Package variables

Included modules

Inherit

Bio::Network::IO Bio::Root::Root

Synopsis

Do not use this module directly, use Bio::Network::IO:

  my $io = Bio::Network::IO->new(-format => 'psi10',
                                 -file   => 'data.xml');

  my $network = $io->next_network;

Description

PSI MI (Protein Standards Initiative Molecular Interaction) XML is a format
to describe protein-protein interactions and interaction networks.
This module parses version 1.0 of PSI MI. The following databases provide their data as PSI MI XML:
   DIP http://dip.doe-mbi.ucla.edu/
   HPRD http://www.hprd.org
   IntAct http://www.ebi.ac.uk/intact
   MINT http://cbm.bio.uniroma2.it/mint/
Each of these databases will call PSI format by some different name.
for example, PSI MI from DIP comes in files with the suffix "mif"
whereas PSI MI from IntAct or MINT has the "xml" suffix.
Documentation for PSI XML can be found at http://www.psidev.info. This module supports a subset of the fields described in PSI MI version 1.0
(http://www.psidev.info/index.php?q=node/88). The NODE DATA section below
describes which fields are currently parsed into ProteinNet networks. See the Bio::Network::IO::psi_xml page in the Bioperl Wiki
(http://bioperl.open-bio.org/wiki/Bio::Network::IO::psi_xml)
for notes on PSI XML from various databases.
When using this parser recall that some PSI MI fields, or classes,
are populated by values taken from an ontology created for the PSI MI
format. This ontology is an OBO ontology and can be browsed at
http://www.ebi.ac.uk/ontology-lookup/browse.do?ontName=MI.

Methods

Methods description

 Name       : next_network
 Purpose    : Constructs a protein interaction graph from PSI XML data
 Usage      : my $net = $io->next_network()
 Arguments  :
 Returns    : A Bio::Network::ProteinNet object

 Name      : _proteinInteractor
 Purpose   : Parses protein information into Bio::Seq::RichSeq objects
 Returns   :
 Usage     : Internally called by next_network()
 Arguments : None

 Name     : _addInteraction
 Purpose  : Adds a new Interaction to a graph
 Usage    : Do not call, called internally by next_network()
 Returns  :
 Notes    : The PSI MI 2.5 standard calls for a field titled interactorRef

Methods code

BEGIN {

	$fac = Bio::Seq::SeqFactory->new(-type => 'Bio::Seq::RichSeq');

}

sub next_network {

	my $self = shift;
	$net = Bio::Network::ProteinNet->new(refvertexed => 1);

	my $t = XML::Twig->new(TwigHandlers => {
								  proteinInteractor =>\& _proteinInteractor,
								  interaction       =>\& _addInteraction
														});
	$t->parsefile($self->file);
	$net;

}

sub _proteinInteractor {

	my ($twig, $pi) = @_;

	my ($acc, $sp, $desc, $prim_id);

	my $org = $pi->first_child('organism');
	my $taxid = $org->att('ncbiTaxId');

	# Make new species object if doesn't already exist
	if ( !exists($species{$taxid}) ) {
		my $common = $org->first_child('names')->first_child('shortLabel')->text;
		my $full;
		# some PSI MI files have entries with species lacking "fullName"
		eval {
			$full = $org->first_child('names')->first_child('fullName')->text;
		};
		$full = $common if $@;

		my $sp_obj = Bio::Species->new(-ncbi_taxid  => $taxid,
												 -name        => $full,
												 -common_name => $common
												);
		$species{$taxid} = $sp_obj;
	}

	# Extract sequence and ontology identifiers
	my @ids          = $pi->first_child('xref')->children();
	my %ids          = map {$_->att('db'), $_->att('id')} @ids;
	$ids{'psixml'}   = $pi->att('id');

	$prim_id = defined ($ids{'GI'}) ?  $ids{'GI'} : '';
	# needs to be done by reference to an actual ontology:
	$acc = $ids{'RefSeq'} || 
	       $ids{'SWP'} ||               # DIP's name for Swissprot
			 $ids{'Swiss-Prot'} ||        # db name from HPRD
			 $ids{'Ref-Seq'} ||           # db name from HPRD
          $ids{'uniprotkb'} ||         # db name from MINT
			 $ids{'GI'} || 
			 $ids{'PIR'} ||
			 $ids{'intact'} ||            # db name from IntAct
			 $ids{'psi-mi'} ||            # db name from IntAct
			 $ids{'DIP'} ||               # DIP node name
          $ids{'ensembl'} ||           # db name from MINT
          $ids{'flybase'} ||           # db name from MINT
          $ids{'wormbase'} ||          # db name from MINT
          $ids{'sgd'} ||               # db name from MINT
          $ids{'ddbj/embl/genbank'} || # db name from MINT
          $ids{'mint'};                # db name from MINT

	# Get description line - certain files, like PSI XML from HPRD, have
	# "shortLabel" but no "fullName"
	eval {
		$desc = $pi->first_child('names')->first_child('fullName')->text; 
	};
	if ($@) {
		warn("No fullName, use shortLabel for description instead");
		$desc = $pi->first_child('names')->first_child('shortLabel')->text;
	}
	
	# Use ids other than accession_no or primary_id for DBLink annotations
	my $ac = Bio::Annotation::Collection->new();	
	for my $db (keys %ids) {
		next if $ids{$db} eq $acc;
		next if $ids{$db} eq $prim_id;
		my $an = Bio::Annotation::DBLink->new( -database   => $db,
															-primary_id => $ids{$db},
											);
		$ac->add_Annotation('dblink',$an);
	}

	# Make sequence object
	my $prot = $fac->create(
						-accession_number => $acc,
						-desc             => $desc,
						-display_id       => $acc,
						-primary_id       => $prim_id,
						-species          => $species{$taxid},
						-annotation       => $ac);

	# Add node to network
	my $node = Bio::Network::Node->new(-protein => [($prot)]);
	$net->add_node($node);

	# Add primary identifier and accession to internal id <-> node mapping hash
	$net->add_id_to_node($ids{'psixml'},$node);
	$net->add_id_to_node($prot->primary_id,$node);
	$net->add_id_to_node($prot->accession_number,$node);

	# Add secondary identifiers to internal id <-> node mapping hash
	$ac = $prot->annotation();
	for my $an ($ac->get_Annotations('dblink')) {
		$net->add_id_to_node($an->primary_id,$node);
	}

	$twig->purge();

}

sub _addInteraction {

	my ($twig, $i) = @_;
	my @ints = $i->first_child('participantList')->children;
	my @nodeids = map {$_->first_child('proteinInteractorRef')->att('ref')} @ints;

	my $interx_id = $i->first_child('xref')->first_child('primaryRef')->att('id');
	
	my $node1 = $net->get_nodes_by_id($nodeids[0]);
	my $node2 = $net->get_nodes_by_id($nodeids[1]);

	my $interx = Bio::Network::Interaction->new(-id => $interx_id);
	$net->add_interaction(-nodes => [($node1,$node2)],
							    -interaction => $interx );
	$net->add_id_to_interaction($interx_id,$interx);

	$twig->purge();
}

1;

__END__

}

General documentation

The Node (protein or protein complex) is roughly equivalent to the PSI MI
interactor (entrySet/entry/interactorList/interactor). The following are
subclasses of interactor whose values are accessible through the Node
object.
   interactor/names/shortLabel
Bio::Annotation::SimpleValue
   interactor/names/fullName
Bio::Annotation::SimpleValue
   interactor/xref/primaryRef
Bio::Annotation::DBLink
   interactor/xref/secondaryRef
Bio::Annotation::DBLink
Bio::Species object
   interactor/organism/names/alias
Bio::Species object
   interactor/organism/names/fullName
Bio::Species object
   interactor/organism/names/shortLabel
Bio::Species object

The following are subclasses of interactor whose values are currently not
accessible through the Node object.
   interactor/names/alias
Bio::Annotation::SimpleValue
   interactor/sequence
   interactor/interactorType/names
   Controlled vocabulary maintained by PSI MI
http://www.ebi.ac.uk/ontology-lookup/browse.do?ontName=MI.
Example: "protein".
Bio::Annotation::OntologyTerm
   interactor/interactorType/xref
Bio::Annotation::DBLink
   interactor/organism/cellType
Bio::Annotation::OntologyTerm
   interactor/organism/compartment
Bio::Annotation::OntologyTerm
   interactor/organism/tissue
Bio::Annotation::OntologyTerm

The Interaction object is roughly equivalent to the PSI MI interaction
(entrySet/entry/interactionList/interaction) and experimentDescription
(entrySet/entry/experimentList/experimentDescription). The following are
subclasses of interaction and experimentDescription whose values are
NOT yet accessible through the Interaction object.
   interaction/xref/primaryRef
Bio::Annotation::DBLink
   interaction/xref/secondaryRef
Bio::Annotation::DBLink
   interaction/organism/names/shortLabel
Bio::Species object
   interaction/organism/names/alias
Bio::Species object
   interaction/organism/names/fullName
Bio::Species object
   interaction/modelled
Bio::Annotation::SimpleValue
   interaction/intraMolecular
Bio::Annotation::SimpleValue
   interaction/negative
Bio::Annotation::SimpleValue
   interaction/interactionType
   Controlled vocabulary maintained by PSI MI
http://www.ebi.ac.uk/ontology-lookup/browse.do?ontName=MI.
Example: "phosphorylation reaction".
Bio::Annotation::OntologyTerm
   interaction/confidenceList
Bio::Annotation::SimpleValue
   experimentDescription/confidenceList
Bio::Annotation::SimpleValue
   experimentDescription/interactionDetectionMethod
   Controlled vocabulary maintained by PSI MI
http://www.ebi.ac.uk/ontology-lookup/browse.do?ontName=MI.
Example: "two hybrid array".
Bio::Annotation::OntologyTerm
   featureElementType/featureType
   Controlled vocabulary maintained by PSI MI
http://www.ebi.ac.uk/ontology-lookup/browse.do?ontName=MI.
The featureType includes data on post-translational modification.
Example: "phospho-histidine".
Bio::Annotation::OntologyTerm

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://bioperl.org/wiki/Mailing_lists  - About the mailing lists

Please direct usage questions or support issues to the mailing list:
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.

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

  http://bugzilla.open-bio.org/

Brian Osborne bosborne at alum.mit.edu
Richard Adams richard.adams@ed.ac.uk