1. Biomedical Informatics Introduction to Ontology Liqin Wang, MS SWE Workshop 2011 Aug 10 th, 2011

2. Biomedical Informatics Outline What is ontology Ontology Language How to create an ontology Ontology Reasoning Evaluation of Ontology Application of Ontology Institutes & People Processing Tools & Triple Store Resources Practice 8/10/20112

3. Biomedical Informatics What is ontology? Def. in Philosophy –A foundational discipline of philosophy, which has its origins in ancient Greece. –About “existence” 8/10/20113

4. Biomedical Informatics What is ontology? (Cont.) Def. in Information Science –A formal, explicit specification of a conceptualization. –A model for describing the world Concepts Properties Constraints Individuals –Also as a domain-specific knowledge base that is machine interpretable, reusable and sharable T. R. Gruber. A translation approach to portable ontologies. Knowledge Acquisition, 5(2):199-220, 1993. 8/10/20114

5. Biomedical Informatics With ontology, we are able… To share common understanding of the structure of information among people or software agents To enable reuse of domain knowledge To make domain assumptions explicit To analyze domain knowledge To separate domain knowledge from the operational knowledge Natalya F. Noy, Deborah McGuinness. Ontology Development 101: A Guide to Creating Your First Ontology 8/10/20115

6. Biomedical Informatics Ontology Language In order to facilitate the create, reading, writing, query and sharing of the ontology, it is a necessary to have ontology language(s). 8/10/20116 Semantic Web Stack by Tim Berners-Lee

7. Biomedical Informatics Ontology Language (Cont.) By syntax: –RDF/XML Interchange (can be written and read by all conformant OWL 2 software) –OWL/XML Easier to process using XML tools –Functional Syntax Easier to see the formal structure of ontologies –Manchester syntax Easier to read/write DL ontologies –Turtle Easier to read/write RDF triples 8/10/20117 Go to protégé and Notepad++

8. Biomedical Informatics OWL OWL (Web Ontology Language) –W3C standard –OWL is built on top of RDF, and written in XML –OWL has three sublanguages DL, Lite, Full –Hard to read by people –Interpretable by computers 8/10/20118

9. Biomedical Informatics Three Variants of OWL OWL Full –an extension of RDF –allows for classes as instances, modification of RDF and OWL vocabularies OWL DL –the part of OWL Full that fits in the Description Logic framework –known to have decidable reasoning OWL Lite –a subset of OWL DL –easier for frame-based tools to transition to –easier reasoning 8/10/20119

10. Biomedical Informatics OWL Example 8/10/201110

11. Biomedical Informatics Manchester OWL syntax Restrictions 8/10/201111

12. Biomedical Informatics Manchester OWL syntax Boolean Class Constructors 8/10/201112

13. Biomedical Informatics Example Person and hasChild some (Person and (hasChild only Man) and (hasChild some Person)) Describes the set of people who have at least one child that has some children that are only men (i.e. grandparents that only have grandsons) 8/10/201113

14. Biomedical Informatics OWL vs. RDF OWL and RDF are much of the same thing, but OWL is a stronger language with greater machine interpretability than RDF. OWL comes with a larger vocabulary and stronger syntax than RDF. OWL supports the identification of inconsistencies (e.g. disjoint classes man & woman, instances and classes cannot be both) 8/10/201114

15. Biomedical Informatics Example of Ontology 8/10/201115

16. Biomedical Informatics How to create an ontology? Determine the scope of the ontology Consider reusing existing ontologies Enumerate important terms in the ontology Define the classes and the class hierarchy Define the properties of classes-slots Define the constraints of the slots Create instances In reality, there is no order for these process… 8/10/201116

17. Biomedical Informatics Determine the scope What is the domain that the ontology will cover? For what we are going to use the ontology? For what types of questions the information in the ontology should provide answers? 8/10/201117

18. Biomedical Informatics Determine the scope What is the domain that the ontology will cover? e.g. Cardiology, food… For what we are going to use the ontology? e.g. Question answering, information extraction For what types of questions the information in the ontology should provide answers? e.g. what is the treatment for patient of cognitive heart failure? 8/10/201118

19. Biomedical Informatics What to reuse? Domain specific ontologies –UMLS Semantic Network Semantic type and Relationship –SNOMED CT –National Drug File – Reference Term IEEE Upper Ontology –Suggested Upper Merged Ontology 8/10/201119

20. Biomedical Informatics Where to get important terms? From –your mind if you are domain experts –Other persons From –Books –Guidelines –Literatures –etc. 8/10/201120

21. Biomedical Informatics Classes Classes usually constitute a taxonomic hierarchy (a subclass-superclass hierarchy) A class hierarchy is usually an IS-A hierarchy –an instance of a subclass is an instance of a superclass If you think of a class as a set of elements a subclass is a subset Multiple inheritance, (A is a B), (A is a C) 8/10/201121

22. Biomedical Informatics Properties rdf:Property –owl:ObjectProperty Link individuals to individuals hasParent –owl:DatatypeProperty Link individuals to data values 8/10/201122 Go to protégé, see NDF-RT.owl

23. Biomedical Informatics Property restrictions Value constraints –Put constraints on value range for a particular property Cardinality constraints –Constraints on the number of values for a particular property e.g. hasParent –Value constraints: allValuesFrom “Human” –Cardinality constraints: maxCardinality “2” 8/10/201123

24. Biomedical Informatics Ontology Reasoning Determine the consistency of ontology Identify subsumption relationships between classes Reasoners: –RacerPro –FaCT++ C++-based reasoner –Pellet –HermiT 8/10/201124

25. Biomedical Informatics Evaluation of Ontology Assessment by human against a set of criteria Natural language evaluation techniques Evaluate use of ontology in an application Comparison of ontology against a source of domain data Using reality as benchmark Ontology accreditation, certification, maturity model 8/10/201125

26. Biomedical Informatics Applications of Ontology Global Health Monitor –BioCasterBioCaster Question answering –QALL-ME NLP/IE –Extended Syndrome Surveillance Ontology 8/10/201126

27. Biomedical Informatics Institutes & People W3C web ontology ONTOLOG Stanford –Thomas R. Gruber –Mark A. Musen, M.D., Ph.D Buffalo ontology site –Barry SmithBarry Smith University of Manchester –e.g. Alan Rector 8/10/201127

28. Biomedical Informatics Processing Tools & Triple Store Ontology Engineering Environments –Ontolingua –Protégé –Altova SemanticWorks (Commercial) Parser/Serializer –Rapper: Raptor RDF parsing and serializing utilityRapper SPARQL Query –Jena ARQ Triple Store –Jena SDB –http://www.w3.org/wiki/LargeTripleStoreshttp://www.w3.org/wiki/LargeTripleStores 8/10/201128

29. Biomedical Informatics Resources all about OWL OWL Web Ontology Language Overview OWL Web Ontology Language Guide OWL Web Ontology Language Reference OWL Web Ontology Language Semantics and Abstract SyntaxOWL Web Ontology Language Semantics and Abstract Syntax OWL Web Ontology Language Test Cases OWL Web Ontology Language Use Cases and Requirements OWL Web Ontology Language XML Presentation Syntax OWL Web Ontology Language Parsing OWL in RDF/XML 8/10/201129

30. Biomedical Informatics QUESTIONS? 8/10/201130

31. Biomedical Informatics Practice Protégé –How to work with Protégé Jena SDB Triple Store –Load the RDF file into triple store 8/10/201131

32. Biomedical Informatics PROTÉGÉ 8/10/201132

33. Biomedical Informatics Protégé - Windows Ontology metrics Preference  Render –Show the classes in different way 8/10/201133

34. Biomedical Informatics Inverse Properties Each object property may have a corresponding inverse property. If some property links individual a to individual b, then its inverse property will link individual b to individual a. 8/10/201134

35. Biomedical Informatics Functional Properties If a property is functional, for a given individual, there can only be at most one individual to be related via this property. –For a given domain, range must be unique Functional properties are also known as single valued properties. 8/10/201135

36. Biomedical Informatics Inverse Functional Properties If a property is inverse functional, then its inverse property is functional. –For a given range, domain must be unique. 8/10/201136

37. Biomedical Informatics Functional vs. inverse functional properties FunctionalProperty vs InverseFunctionalProperty domainrangeexample Functional Property For a given domain Range is unique hasFather: A hasFather B, A hasFather C  B=C InverseFunctional Property Domain is unique For a given range hasID: A hasID B, C hasID B  A=C 8/10/201137

38. Biomedical Informatics Transitive Properties If a property is transitive, and the property related individual a to individual b, and also individual b to individual c, then we can infer that individual a is related to individual c via property P. 8/10/201138

39. Biomedical Informatics Protégé – DL Query Quickly test definitions of classes to see that they subsume the appropriate subclasses. Or check for class membership of arbitrary descriptions without having to create named class placeholders. Follow Manchester OWL SyntaxManchester OWL Syntax http://protegewiki.stanford.edu/wiki/DLQueryTab Example –For pizza.owl –Pizza and hasTopping only VegetarianTopping 8/10/201139

40. Biomedical Informatics Protégé - OWLViz http://protegewiki.stanford.edu/wiki/OWLViz Graphviz –Installation –Set path 8/10/201140

41. Biomedical Informatics TRIPLE STORE 8/10/201141

42. Biomedical Informatics Install Mysql server Latest mysql connector jar file Jena SDB Cygwin 8/10/201142

43. Biomedical Informatics THANK YOU! 8/10/201143