Research
In order to understand complex diseases such as diabetes or Parkinson's disease and to develop adequate treatment strategies, it has become clear that systemic approaches have to be taken. One step in such a challenging analysis is the integration of heterogeneous data into a particular tissue model in order to quantify perturbation resulting in disease states. The scientific goal of the group Computational Modeling in Biology (CMB) is to develop such quantitative models. Recently we have been focussing on stem cell differentiation, neural development and disease phenotypes resulting from misregulation. Since a theoretical model provides only limited value to our biological partners, we feed back model estimations in an iterative refinement cycle. The particular novelty in our approach is to include detailed prior information – which is readily available in many applications from molecular biology – in the inference process, and to do so on different levels of model accuracy. Realizing that no single model can ever accurately reflect reality, we use Bayesian inference to derive ensembles of models, fitting to the data in varying quality. We are convinced that applied research feeds back to theory and vice versa. So, we aim at joining our knowledge in the theoretical disciplines of statistical machine learning with dynamical systems and applied modeling in systems biology.
stem cells & development
regulatory networks
metabolomics
statistical inference
dynamical systems