Abstract

The semantic web depends on the use of ontologies to let electronic systems interpret contextual information. Optimally, the handling and access of ontologies should be completely transparent to the user. As a means to this end, we have developed a service that attempts to bridge the gap between experts in a certain knowledge domain, ontologists, and application developers. The ontology-based answers (OBA) service introduced here can be embedded into custom applications to grant access to the classes of ontologies and their relations as most important structural features as well as to information encoded in the relations between ontology classes. Thus computational biologists can benefit from ontologies without detailed knowledge about the respective ontology. The content of ontologies is mapped to a graph of connected objects which is compatible to the object-oriented programming style in Java. Semantic functions implement knowledge about the complex semantics of an ontology beyond the class hierarchy and “partOf” relations. By using these OBA functions an application can, for example, provide a semantic search function, or (in the examples outlined) map an anatomical structure to the organs it belongs to. The semantic functions relieve the application developer from the necessity of acquiring in-depth knowledge about the semantics and curation guidelines of the used ontologies by implementing the required knowledge. The architecture of the OBA service encapsulates the logic to process ontologies in order to achieve a separation from the application logic. A public server with the current plugins is available and can be used with the provided connector in a custom application in scenarios analogous to the presented use cases. The server and the client are freely available if a project requires the use of custom plugins or non-public ontologies. The OBA service and further documentation is available at http://www.bioinf.med.uni-goettingen.de/projects/oba

Highlights

  • Ontologies play a major role in the semantic web (Berners-Lee et al, 2001)

  • Returns all classes linked to a developmental stage

  • The annotated phenotypes are linked to these classes Returns all classes not related to a developmental stage, used for the auto-complete function of the search interface Searches in the labels and synonyms of generic classes and an additional previous generated list for classes matching the search string

Read more

Summary

Introduction

Ontologies play a major role in the semantic web (Berners-Lee et al, 2001). Running in the background they provide electronic systems with the expertise of a knowledge domain. By connecting unknown terms with known ones through defined statements, new knowledge can be deduced. This knowledge can be used to provide the user with information that he/she is seeking but could not exactly specify. This is achieved by means of a mandatory class hierarchy, using the “is_a” relation, and other relations, connecting the ontology classes to each other. While the meaning of relations is comprehensible to human users so that they can select the right one for traversing the graph, it is a particular challenge to transfer the logical axioms defined in an ontology into an objectoriented view that is common to most applications (Winston et al, 1987; Burger et al, 2008)

Methods
Results
Discussion
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call