Data analysis, machine learning, and modelling in the bio-medical domain

The amount of available data in the life sciences has been increasing dramatically in recent years. Newly developed and increasingly affordable techniques contribute to a 'data explosion' in, e.g., biology and the environmental sciences. In medicine, for instance, besides imaging and traditional lab technology, the acquisition of genomics, proteomics and other "-omics" data plays an increasing role in, both, basic research and clinical practice. This development triggers a strong need for efficient computational and automated analyses, but also for new approaches to the faithful modelling of underlying biological processes. Just a few of the current challenges and most important questions are:

• the identification and efficient use of relevant bio-markers for reliable diagnosis support tools
• the visualization and analysis of very high-dimensional data as in, e.g., gene expression or proteomics
• the inference of, for instance, underlying metabolic or regulatory networks from the available data
• the efficient combination of 'mixed data' in a unified framework, e.g. concerning patient data from a large variety of sources and technical platforms

Mostly, these questions cannot simply be addressed by application of existing methods "from the shelf", but require the development of novel computational and modelling tools, tailored to the specific needs of bio-medical research. Progress in this area relies on constant dialogue and interaction between the disciplines. The aim of this workshop is, therefore, to bring together active researchers from many of the involved areas and facilitate intense scientific exchange. This should lead to the identification of novel directions of research and potential novel collaborations.