Data and Knowledge Mining in Biomedical Research   

Chairperson: Roderic Guigó
Centre de Regulació Genòmica
Barcelona, Spain

Anthony J. Brookes
University of Leicester
Leicester, United Kingdom
"New Developments in Genotype-To-Phenotype Databasing"
The era of high-throughput DNA analysis is rapidly evolving to include a focus upon the genetic etiology of disease and other phenotypes. Previous 'genomics' work was supported by extensive databases, and such depositories must now be enhanced, adapted, and integrated to support the new genotype-phenotype emphasis of current research. This talk will consider some of the key challenges and strategies being used to move the field towards creating a unified 'knowledge grid for genotype-phenotype information'. Such a development will help advance the use of Mendelian genetics in medicine, and bolster emerging systems for personalised healthcare.

Alfonso Valencia
Centro Nacional de Investigaciones Oncológicas (CNIO)
Madrid, Spain
"Computational Methods for the Analysis of Biomedical Information"
The information on molecular systems is growing at increasing speed including together with the original experimental information (sequences, structures, interactions, expression data) additional information on the experimental conditions and the results of the scientific analysis (i.e. aspects related with the predicted function of genes and proteins). All this information is accessible to the large community of experimental biologist in a large variety of databases, heterogeneous methods accessible via web and scientific publications, including 12 million abstracts and an increasingly growing number of full text publications. In the context of international networks such as Biosapiens, ENFIN and Embrace, and as part of the group of the Spanish National Bioinformatics Institute our group has developed an integrated set of computational methods that facilitate the annotation of genomes and for the analysis of potential cancer related genes. A description of these methods and their integration as workflows will be presented during the conference.

Jordi Mestres
GRIB, IMIM/Universitat Pompeu Fabra
Barcelona, Spain
"Knowledge-Based Approaches to Pharmacological Profiling"
Recent efforts towards the construction of annotated chemical libraries have contributed to the establishment of the knowledge base for integration of chemical and biological data. There is now an urgent need for novel computational tools that allow for efficiently navigating through this knowledge base, while exploiting it to provide an estimation of the pharmacological profile of molecules.

Patrick Ruch
Geneva University Hospitals
Geneva, Switzerland.
"Automatic Annotation of Pathological Functions in UniProt"
Although the UniProt KnowledgeBase is not a medical oriented database, it contains information on more than 3’000 human proteins involved in pathologies. In order to make these data easily accessible to clinical experimentalists, we have developed a procedure to automatically link the curated medical information of UniProtKB/Swiss-Prot entries to the Medical Subject Diseases (MeSH). The approach relies on two types of evidences: the MeSH descriptors directly provided by UniProt references to MEDLINE, the description of medically-relevant (e.g. DISEASE, PHARMACEUTICAL, FUNCTION) fields in UniProt, which will be normalized using an automatic text categorizer. The method is assessed on a benchmark set, whose scope has been restricted to pathological categories.

Dietrich Rebholz-Schumann
European Bioinformatics Institute
Cambridge, United Kingdom
"Streaming Facts from Scientific Publications to the Scientist"
Scientific literature is increasingly available in electronic form and early on after its acceptance for publication. Techniques to analyse the literature for contained facts are then applied to deliver individual facts directly to the scientist. This leads to the integration of the scientific literature into the infrastructure of existing IT data resources. This talk will explain how scientists in the biomedical domain profit from an infrastructure consisting of services for information extraction. All services automatically process the documents and interlink them with bioinformatics data resources. In addition they can be integrated into external IT solutions to directly couple experimental results with annotations from the scientific literature.

close window