21.11.2014 Views

ayout 1 - EMBL Grenoble

ayout 1 - EMBL Grenoble

ayout 1 - EMBL Grenoble

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

<strong>EMBL</strong> Research at a Glance 2009<br />

Dietrich<br />

Rebholz-<br />

Schuhmann<br />

Master in Medicine, 1988,<br />

University of Düsseldorf.<br />

PhD 1989, University of<br />

Düsseldorf.<br />

Master in Computer Science,<br />

1993, Passau.<br />

Senior scientist at gsf,<br />

Munich, Germany and LION<br />

bioscience AG, Heidelberg.<br />

Facts from the literature and biomedical<br />

semantics<br />

Previous and current research<br />

Text mining comprises the fast retrieval of relevant documents from the whole body of the literature<br />

(e.g. Medline database) and the extraction of facts from the text thereafter. Text mining solutions<br />

are now becoming mature enough to be automatically integrated into workflows for<br />

research work.<br />

Research in the Rebholz-Schuhmann group is focussed on fact extraction from the literature. It is<br />

our goal to automatically connect literature content to other biomedical data resources (e.g. bioinformatics<br />

databases) and to evaluate the results. Ongoing research targets the identification of relationships<br />

between genes and diseases, molecular interactions and other types of information.<br />

Over the past two years, the team has generated several public resources: a lexicon of biomedical<br />

terms, an ontology for gene regulatory events and recently an authoring service (PaperMaker).<br />

The work in the research group is split into different parts: 1) research work in named entity recognition<br />

and its quality control (e.g. UKPMC project, CALBC); 2) knowledge discovery tasks, e.g.<br />

for the identification of gene–disease associations; and 3) development of a modular IT infrastructure<br />

for information extraction (Whatizit). All parts are tightly coupled.<br />

Future projects and goals<br />

Group leader at <strong>EMBL</strong>-EBI<br />

since 2003.<br />

The following goals are priorities for the future. Firstly we will continue our ongoing research in<br />

term recognition and mapping to biomedical data resources to establish state-of-the-art text mining<br />

applications. We will develop this by focussing on automatic means to measure and evaluate<br />

existing options to identify the most promising solutions (UKPMC project, CALBC support action).<br />

Secondly, we will invest further effort into the extraction of content from the scientific literature. Such solutions will be geared towards the<br />

annotation of diseases and the generation of fact databases. As part of this research we will investigate workflow systems where text mining<br />

supports bioinformatics information retrieval solutions. One solution is the integration of public biomedical data resources into the data<br />

from the biomedical scientific literature.<br />

Finally, we will increase the availability of information extraction solutions based on SOAP web services for the benefit of the bioinformatics<br />

community. This requires standards in the annotation of scientific literature and will automatically lead to semantic enrichment of the scientific<br />

literature.<br />

Overview of the categorisation of information retrieval<br />

tools on the basis of their input and output formats.<br />

Selected references<br />

Beisswanger, E. et al. (2008). Gene Regulation Ontology (Gro):<br />

Design principles and use cases. Paper presented at Studies in<br />

Health Technology and Informatics 2008, 136, 9-1<br />

Jaeger, S. et al. (2008). Integrating protein-protein interactions and<br />

text mining for protein function prediction. BMC Bioinformatics, 9,<br />

Article S2<br />

Kim, J.J. et al. (2008). MedEvi: Retrieving textual evidence of<br />

relations between biomedical concepts from Medline. Bioinformatics,<br />

2, 110-112<br />

Kim, J.J. & Rebholz-Schuhmann, D. (2008). Categorization of<br />

services for seeking information in biomedical literature: A typology<br />

for improvement of practice. Brief. Bioinform., 9, 52-65<br />

Rebholz-Schuhmann, D. et al. (2008). Text processing through web<br />

services: Calling Whatizit. Bioinformatics, 2, 296-298<br />

70

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!