Disease Systems Biology

Advances in medical research revealed that a disease phenotype is the result of pathobiological processes that interact in complex networks. The group works on computational data integration approaches where clinical and phenotypic data are combined with molecular level data, predominantly exome and transcriptome information. One important aim is to identify the genetic contributors to human variation in drug efficacy, which is an important aspect in personalized medicine. By longitudinal exome analyses of the tumor heterogeneity we can significantly reduce the number of hits for functional validation. Another focus is to apply RNA sequencing of human tissues and network biology to study the complexity of the biochemical networks at multiple scales for understanding the development of metabolic diseases, and specifically diseases linked to energy storage.

The research group is included in the Germany-wide Research network „Virtual Liver“ to establish the systems biology of human liver. The group at the HKI analyses transcriptome data to infer gene regulatory networks that drive the central metabolism of the liver under nutritional challenges and under the pathophysiological conditions of Non-Alcoholic Fatty Liver Disease (NAFLD; e.g. steatosis, steatohepatitis). The network models are constructed by fitting the simulated gene expression profiles to the time series obtained from DNA-microarrays using structural network templates supported by the prior knowledge from biomolecular databases and by text mining from scientific journals. The inferred networks are used to predict hypotheses on molecular interactions and to design follow-up experiments. Further information: www.virtual-liver.de/wordpress/en

Systems biology of fungal infection aims at understanding the interaction of the host, in particular the immune system, with components of the fungal pathogens by analysing two interacting networks. Gene regulatory network models were and will be reconstructed (inferred) based on the analysis of high-throughput data integrating prior knowledge. The inference of transcriptional regulatory networks allows for an incorporation of both the gene expression data together with promoter sequence information. As a result, a set of potential target genes can be predicted, which are presumably responsible for the interaction of the fungal pathogens with the host. This information, in turn, can be used as a starting point for the investigation of the host’s response to the pathogen’s attack.