Skip to content

Latest commit

 

History

History
187 lines (148 loc) · 4.97 KB

semweb.md

File metadata and controls

187 lines (148 loc) · 4.97 KB
Raw

Semantic Web

Semantic Web technologies are the new glue to link different data sources. Bioclipse provides support for several of the technologies, including general Resource Description Format (RDF) support, the SPARQL query language, and the Web Ontology Langauge (OWL). This chapter outlines how these technologies can be used in Bioclipse.

The rdf manager

The rdf manager can be used to handle Resource Description Framework (RDF) data [1]. The basic unit of information in RDF is a triple and these triples are stored in a triple store. A file based store can be created with this code:

Script code/CreateStore.groovy

base = rdf.createStore("/tmp/chebiowl")

Here, the path is a full operating system-style path, and this one works only on Unix/Linux/BSD systems.

As soon as you have a store, you can start adding triples to it. For example, when they come from a file, you may want to do something like this:

knowledgebase = "/WikiPathways/chebi.owl";
kbFormat = "RDF/XML";
base = rdf.createStore("/tmp/chebiowl")
rdf.importFile(base, knowledgebase, kbFormat);

If the amount of data is limited, you can also create an in-memory model:

knowledgebase = rdf.createInMemoryStore();

This can be used to create triples from data you are processing:

rdf.addObjectProperty(knowledgebase,
  "http://linkedchemistry.info/chembl/molecule/m443",
  "http://www.w3.org/2000/01/rdf-schema#subClassOf",
  "http://semanticscience.org/resource/CHEMINF_000000"
)
rdf.addDataProperty(knowledgebase,
  "http://linkedchemistry.info/chembl/molecule/m443",
  "http://www.w3.org/2000/01/rdf-schema#label",
  "CHEMBL268854"
)

You can get the number of triples held in the store with the size method:

rdf.size(knowledgebase);

And the collected set of triples can be serialized as Turtle, which is one of the easier to read serialization formats of RDF triples. For this, you can use:

rdf.asTurtle(knowledgebase);

The owlapi manager

The owlapi manager exposes functionality of the OWLAPI library, to deal with Web Ontology Languages (OWL) ontologies (doi:10.3233/SW-2011-0025). The manager is oriented at ontologies in the Bioclipse workspace. Thus, loading an ontology works like:

ontology = owlapi.load(
  "/eNanoMapper/enanomapper.owl", null
);

The second parameter is a mapper, which can be used to indicate where imported ontologies can be locally found. For example:

mapper = null; // initially no mapper
mapper = owlapi.addMapping(mapper,
  "http://purl.bioontology.org/ontology/npo",
  "/eNanoMapper/npo-asserted.owl"
);
mapper = owlapi.addMapping(mapper,
  "http://www.enanomapper.net/ontologies/" + 
  "external/ontology-metadata-slim.owl",
  "/eNanoMapper/ontology-metadata-slim.owl"
)
ontology = owlapi.load(
  "/eNanoMapper/enanomapper.owl", mapper
);

You can always see what IRIs are mapped with:

owlapi.listMappings(mapper)

Once you have loaded an ontology, you can list all the imported ontologies:

imported = owlapi.getImportedOntologies(ontology)
for (var i = 0; i < imported.size(); i++) {
  js.say(
    imported.get(i).getOntologyID().getOntologyIRI()
  )
}

Similarly, you can list all the classes defined by the ontology or imported ontologies:

imported = owlapi.getImportedOntologies(ontology)
for (var i = 0; i < imported.size(); i++) {
  js.say(
    owlapi.showClasses(imported.get(i))
  )
}

Finally, the owlapi manager also provide functionality to check for profile violations:

owlapi.checkVioloations(ontology)

The ldf manager

Linked Data Fragments (LDF) are a novel technology allowing distributed querying. The LDF software used in Bioclipse provides the functionality by exposing a fragment as a RDF store. That is very fortunate, because it means we can automatically integrate it with the existing RDF functionality, for example provided by the rdf manager. For example, we can query for and validate CAS registry numbers in WikiData (as Groovy script):

wikidataldf = ldf.createStore(
  "http://data.wikidataldf.com/wikidata"
)

identifier = "P231"

sparql = """
PREFIX wd: <http://www.wikidata.org/entity/>

SELECT ?compound ?id WHERE {
  ?compound wd:${identifier}s [ wd:${identifier}v ?id ] .
}
"""
mappings = rdf.sparql(wikidataldf, sparql)

// safe to a file
for (i=1; i<=mappings.rowCount; i++) {
  casNo = mappings.get(i, "id")
  compound = mappings.get(i, "compound")
  if (!cdk.isValidCAS(casNo)) {
    js.say(
      "Compound " + compound + " has an " +
      " invalid CAS registry number: " +
      casNo
    );
  }
}

References

  1. Samwald M, Jentzsch A, Bouton C, Kallesøe CS, Willighagen E, Hajagos J, et al. Linked open drug data for pharmaceutical research and development. J Cheminform. 2011;3(1):19. doi:10.1186/1758-2946-3-19 (Scholia)