Interdisziplinäres Zentrum
für Bioinformatik


Home Suche  |  Sitemap  |  Impressum  |

LIFE - Transcriptomics Group

Bioinformatik Uni Leipzig

http://www.life.uni-leipzig.de

MediGRID – Networked Computing Resources for Biomedical Research (Project Ontology Tools)


Toralf Kirsten, Michael Hartung
Interdisciplinary Centre for Bioinformatics
University of Leipzig

Markus Löffler, Katrin Loebe
Institute for Medical Informatics, Statistics and Epidemiology
University of Leipzig

Erhard Rahm
Dept. of Computer Science
University of Leipzig


Grid computing is regarded as one of the critical success factors in biomedical research. The joint project MediGRID unifies well known research institutes in the area of medicine, biomedical informatics and life sciences into a consortium. Numerous associated partners from industry, healthcare and research facilities ensure a broad representation of these communities.



From raw data to research results
The main goal of MediGRID is the development of a Grid middleware integration platform enabling eScience services for biomedical life science. Therefore the consortium allocated the tasks in different modules. The four methodological modules (middleware, ontology, resource fusion and eScience) plan to incrementally develop and provide a Grid infrastructure while taking into account the need of the biomedical users. The user communities are represented in three research modules for biomedical informatics, image processing and clinical research.

MediGRID is initially funded for three years. Within this time period our goal is to demonstrate the benefit of Grid computing in use cases covering different areas of biomedical informatics. The first cases will cover highly grid enabled projects from biomedical informatics in order to develop best practices for grid-enabling other biomedical projects.

Whereas Grid projects are pretty similar in the lower layers, the biomedical community has to face particular challenges in the upper layers. The top layer represents the heterogeneous biomedical data sources. Beyond the problem to find the relevant data sets via metadata description, access control to the data is of paramount interest, as the owner of the data are foremost patients. Due to the heterogeneity of the data we need an additional ontology layer to homogenize the data. Given semantic interoperability the researcher can correlate and analyze the data with biomedical informatics methods. Finally the result data can be presented.

The successful usage of services, data and physical resources in a Grid environment requires beside technical a semantic interoperability between organisations and nodes. Ontologies are an elementary instrument for describing different resources in a semantic unique way.




The module ontology tools focuses on two purposes:


  • The “OntoBuilder” tool will be provided to manage and build simple ontologies (controlled vocabularies) used by selected MediGRID applications.

  • Existing biomedical ontologies (e.g. GeneOntology) are integrated and utilized for MediGRID applications. This fusion service is based on the existing GenMapper tool


An ontology provides the conceptualisation of a domain. With a system of semantically unique defined terms an ontology can be used to structure the knowledge in the domain and to describe it for the exchange between applications clearly. Instances of ontologies vary from standardized vocabularies, taxonomies (is-a hierarchies) up to complex networks of concepts including different types of relations.

top