Systems Biology Group
Mathematical modeling and simulation techniques have turned out to be valuable tools for the understanding of complex systems in different areas of research and engineering. In recent years, this approach has come to application frequently also in biology, resulting in the establishment of the research area systems biology. Systems biology tries to understand the behaviour of complex biological systems by means of mathematical approaches. This requires the integration of qualitative and quantitative experimental data into coherent models. A challenging task is the development of comprehensive models that can be used in medical and pharmaceutical research for the establishment of a personalized medicine. The Systems Biology Group has its research interests in the mathematical modeling of cellular processes with respect to complex diseases, such as cancer. Within the group different systems biology resources and tools for the modeling and simulation of biological systems have been designed and implemented. These tools are used in current projects for the modeling of cancer-related signal transduction and metabolic pathways, their subsequent gene regulatory network, and the effect of mutations and drugs. Moreover, the group is working on the modeling of stem cell biology. The research is driven by the integration of diverse ‘omics data, as generated by current (high-throughput) technologies.
Scientific methods and findings
The Systems Biology Group is hosting the PyBioS modeling and simulation system (http://pybios.molgen.mpg.de). PyBioS has a web-based user interface (Figure 15) and it makes use of well established methods for the mathematical description of biochemical reaction systems based on ordinary differential equation systems and Petri nets, and novel interfaces to biochemical pathway databases (e.g., Reactome, KEGG, ConsensusPathDB). In addition PyBioS provides several functionalities for model analysis and visualization. Moreover, the systems biology group is involved in the development of the ConsensusPathDB database that is hosted by the Bioinformatics Group of Ralf Herwig.
(A) The PyBioS software (http:// pybios.molgen.mpg.de) acts as a repository for mathematical models of biological systems. Different tabs of the web interface provide functions for model design, simulation, analy- sis, and visualization (B).[less]
(A) The PyBioS software (http:// pybios.molgen.mpg.de) acts as a repository for mathematical models of biological systems. Different tabs of the web interface provide functions for model design, simulation, analy- sis, and visualization (B).
Structure and behavior of any cell and any organism are determined by converting information in the genome and the environment into the phenotype through a series of molecular processes. Dysfunctions in the molecular interaction network can cause sever diseases such as cancer. Curing the disease often involves by itself complex disturbances in these networks. Progress in the treatment of tumours in individual patients will depend critically on being able to predict the effects of such treatments in the context of the genome involved. The development of predictive models is however complicated by the lack of information on many of the reaction kinetics needed. Information on the kinetics and kinetic parameters is either not available at all, or, at best, is based on experiments often carried out under conditions quite different from those in living cells. Thus, computational modeling approaches must primarily face the challenge of coping with this lack of information. One approach to overcome this limitation can be a rigorous analysis of the model’s parameter space, e.g., by sampling unknown parameters from appropriate random distributions and a subsequent statistical evaluation. Such a kind of Monte Carlo-based approach makes it necessary to run thousands of simulations and thus can only be performed using distributed computing.
This approach has been developed and implemented within the Systems Biology Group. In the course of current research projects (MUTANOM, MoGLI, TREAT20) the Sys- tems Biology Group has established a large model of cancer-related pathways. Currently, the model comprises more than 2800 components and 4400 individual reactions and it covers different signaling pathways, such as EGF-, IGF-, NGF-, Wnt-, Notch-, Hedgehog-, Fas-, Trail-signaling, etc. Furthermore, the model has been extended by the integration of validated microRNA (miRNA) target information and subsequently it was used to analyze the impact of specific miRNA inhibitors. In cooperation with the groups of M.-R. Schweiger, M.-L. Yaspo, and B. Lange the model is used to study the effects of cancer-related somatic mutations and the identification of reasonable intervention points (Figure 16). Moreover, within the MoGLI project we study the transcriptional program and the molecular circuitries regulated by Hedgehog (HH) and GLI proteins (Figure 17). The results will improve our understanding of the complex molecular networks regulated by oncogenic HH/GLI signaling and will accelerate the search for novel molecular targets that represent an opportunity for therapeutic intervention.
(A) Genome-wide time course expression data was used to identify synergistic effects of Hedgehog signaling with other pathways, such as EGF-mediated signaling. (B) Clusters of co-regulated genes (C) and individual synergistic genes were identified. (D) Fur- ther integration of gene regulatory data was used for the development of network models of differentially and synergistically regulated genes.[less]
(A) Genome-wide time course expression data was used to identify synergistic effects of Hedgehog signaling with other pathways, such as EGF-mediated signaling. (B) Clusters of co-regulated genes (C) and individual synergistic genes were identified. (D) Fur- ther integration of gene regulatory data was used for the development of network models of differentially and synergistically regulated genes.
Within the IMGuS project on systems biology of steatohepatitis, a model of liverrelated metabolic processes has been developed that is further analyzed and ex- tended also in the LivSYSiPS project on stem cell reprogramming and differentiation related to steatohepatitis (in cooperation with J. Adjaye). The established resources, tools, algorithms, and models build a foundation for the application of systems biology strategies in medical and pharmaceutical research and, based on data from high-throughput genome, transcriptome, and proteome analysis, it enables the development of a personalized medicine.