Today a wide array of computational methods is complementing experimental efforts in discovering new, bioactive molecules, including drugs, cosmetics, agrochemicals and functional foods. Development and application of these methods is driven by an urgent need to make drug discovery more efficient and limit animal experiments to a non-reducible minimum.
In our lab we focus on the development and application of novel computational approaches for early drug discovery (i.e. the hit identification, hit-to-lead and lead optimisation; see figure one the left). We are particularly interested in developing in silico models for the prediction of xenobiotic metabolism, which is a key factor for the safety and performance of any chemicals in contact with an organism. Further research branches include the development of algorithms and methods for the prediction of (i) 3D structures of small molecules (conformer ensembles), (ii) the biomacromolecular targets of small molecules (polypharmacology) and (ii) bad actors and frequent hitters in biochemical assays.
We also run a large number of projects where we use ligand-based and structure-based methods to identify and optimise bioactive compounds for challenging drug targets. These projects are conducted in close collaboration with on-site, domestic or international partners. We maintain large molecular libraries containing millions of small molecules (in particular also natural products) that are purchasable or can be sourced by our partners. Using methods such as automated ligand docking, we select the most promising compounds in silico, acquire them, and submit them to experimental testing. It is always an exciting experience for us to receive testing results for our predicted molecules. These results are then fed back into the model development cycle.