Humboldt-Universität zu Berlin - Collaborative Research Center for Theoretical Biology

Theory-based prediction of drug targets in biological networks

Theory-based prediction of drug targets in biological networks is still in its infancy. This holds both for diseases that are caused by intracellular changes (such as genetic disorders, cancer) as well as for diseases that are induced by parasites. Currently, potential drugs are pre-dominantly identified by testing the effect of high number of compounds on an individual target. Network effects are largely not considered. We will develop a theoretical framework for the prediction of drug targets and for the assessment of effect of a drug application, based on the knowledge of the affected network and availability of a sensible mathematical model. Moreover, we will develop a search schema for combinations of drug targets that together allow for a more effective and – in case of host-parasite interactions – more host-gentle than single treatments. Specifically, we will also test different types of interactions (inhibition and activation) of the drugs with their targets.

This type of research is based on the development of computational models describing the dynamics of cellular networks, preferentially in form of sets of ordinary differential equations. Those networks may comprise metabolic reactions, but also signalling pathways and other types of regulatory networks. The analysis will make use of classical sensitivity analysis, including the metabolic control analysis, which quantifies the effect of infinitesimally small perturbations, but it will first be extend to finite perturbations, second to combinations of perturbations, and will third include the search for optimal perturbation strategies. Since it becomes understood, that cellular regulation is not restricted to the classical view of individual pathways, we will investigate networks, which involve virtually all relevant reactions and interactions (e.g. metabolic reactions together with the regulation of the genes encoding their enzymes).

Beyond exploring general principles of target identification and target combination, the approach will be applied to networks and diseases under study. The first application example is response of yeast cells to external stress, specifically to osmotic stress and pheromone. We closely collaborate with Prof. Hohmann (University Göteborg) on these pathways, such that much experience both on the experimental and on the modelling site has accumulated. The second application example is a network of signalling pathways (TGFβ and canonical Wnt pathway), which are relevant in colon cancer. We study these pathways in collaboration with the Sers/Schäfer (Charité, Berlin) and Knaus (Free University Berlin) groups. The third application is the metabolic syndrome, which is a major cause of disease in industrialized societies. We established collaboration with Prof. van Driel and Dr. Bakker (Vrije Universiteit Amsterdam) in order to study means to combat such diseases.

german version