IC-OnSim

Extension of the similarity measure OnSim for any two entities in a knowledge graph.

[1] Ignacio Traverso-Ribón, Maria-Esther Vidal and Guillermo Palma. A Similarity Measure for Determining Relatedness Between Ontology Terms. 11th International Conference on Data Integration in the Life Sciences 2015 (DILS2015).

[2] Ignacio Traverso-Ribón, Maria-Esther Vidal. Exploiting Information Content and Semantics to Accurately Computing Similarity of GO-based Annotated Entities. 2015 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology

[3] Palma, G.; Vidal, M. E.; Haag, E.; Raschid, L.; Thor, A. Measuring Relatedness Between Scientific Entities in Annotation Datasets. ACM International Conference on Bioinformatics, Computational Biology, and Biomedical Informatics (BCB), 2013

Setting up AnnSim-OnSim

Once you have downloaded the code from GitHub you only need to modify the file src/main/java/test/DatasetTest.java to run AnnSim-OnSim.

IC-OnSim need as inputs:

• An OWL Ontology.
• A file describing the comparisons to be performed.
• A file for each entity containing their annotations.

Comparison file

The comparison file follows the following format:

ENTITY1 ENTITY2
ENTITY1 ENTITY3
ENTITY3 ENTITY4

Each line represents a pair of entities to be compared. These entities can be, e.g., proteins. In this case, the file will look like this:

Q9JKJ9	Q9NYL5
Q9JKJ9	Q60991
Q9JKJ9	Q64505
Q9JKJ9	P11511
Q9JKJ9	P27786

Annotation files

For each entity present in the comparison file, AnnSim-OnSim needs a file containing its annotations. The annotations are terms from the given ontology. In case our entities are proteins annotated with the Gene Ontology, the file would look like this:

3
GO:0045454
GO:0006467
GO:0055114

The first line contains the number of annotations included in the file and each annotations is written in a different line.

Main example

Map<String, String> ontPrefix = new HashMap<String,String>();
	ontPrefix.put("src/main/resources/dataset3/", "http://purl.org/obo/owl/GO#");
	String prefix = "src/main/resources/dataset3/";
	String ontFile = prefix + "goProtein/go.owl"; // File containing the OWL Ontology
	
	MyOWLOntology o = new MyOWLOntology(ontFile, ontPrefix.get(prefix));
	String comparisonFile = "src/main/resources/dataset3/proteinpairs.txt"; // File containing the entity pairs to be compared.

	List<ComparisonResult> comparisons = readComparisonFile(comparisonFile);
	int counter = 0, total = comparisons.size();

	String[] files = {"src/main/resources/dataset3/process_annt"}; //Folder containing the annotations files
	
	for (Iterator<ComparisonResult> i = comparisons.iterator(); i.hasNext();)
	{
		ComparisonResult comp = i.next();
		double sim = 0;
		for (String file:files)
		{
			Set<OWLConcept> a = getConceptAnnotations(comp.getConceptA(), file, o);
			Set<OWLConcept> b = getConceptAnnotations(comp.getConceptB(), file, o);
			AnnSim bpm = new AnnSim();
			double aux = bpm.matching(a, b, null, null); 
			String s = comp.getConceptA() + "\t" + comp.getConceptB() + "\t" + aux;
			sim += aux;
		}
		comp.setSimilarity(sim/files.length);
		System.out.println(comp + "\t" + counter++ + "/" + total);
	}

License

This work is licensed under GNU/GPL v2.