ULM UNIVERSITY

Ulm University
Core Unit Medical Systems Biology 
Albert-Einstein-Allee 11 
89081 Ulm, Germany 
Website

Project leader

PD Dr. Hans A. Kestler, Dipl.-Ing.

Ulm University,
Core Unit Medical Systems Biology

Phone: +49 731 50 24248
Fax: +49 731 50 24156

email contact form
Institute presentation

The Research Group “Bioinformatics and Systems Biology” at Ulm University (PD Dr. Kestler) group has a well-documented and broad range expertise in the analysis and data-mining of high-throughput data and the mathematical modelling of cellular processes. Our publications cover the analysis of CGH, matrix-CGH, and gene expression data in different types of tumours especially pancreatic carcinoma (Gress et al., 2012; Kleger et al., 2012; Meyer et al., 2011). We develop new methods in the fields of machine learning, computational statistics, and visualisation (Kestler et al., 2005). Recently, we established the Bioinformatics Unit of the Genomics Facility (University Ulm). The current general transition to NGS platforms leads to extraordinary increases of the amount of data generated. To handle this large amount of data we also develop parallelised algorithms (Kraus and Kestler, 2010; Mussel et al., 2010). In recent cooperative projects, we have established various mathematical models of biological processes. Using a delay-differential equations model for myeloid and lymphoid hematopoietic stem cell (HSC) and progenitor populations, which can undergo apoptosis, proliferation and differentiation, we show that enabling the differentiation of myeloid HSCs to lymphoid HSCs allows a better explanation of the measured cell counts upon stimulation by irradiation (Wang et al., 2012). We also devised an extended ODE model of the Wnt signalling pathway (Wawra et al., 2007) and analysed the concentration-dependent activation of Wnt pathways in silico (Kestler and Kuhl, 2011). Supported by our standard toolkit for the analysis and simulation of Boolean networks (Mussel et al., 2010), a Boolean model of the heart development was established (Herrmann et al., 2012). We also developed new approaches to reconstruction of large-scale Boolean networks from time series data (Maucher et al., 2011).