Fraunhofer-Institut für Algorithmen und Wissenschaftliches Rechnen SCAI
My research work is concerned with computational drug design. The main focus is to estimate binding affinites of potential drug candidate molecules to a target protein automatically. Therefore several aspects are important:
One the one hand, for the selection and preparation of small molecules as drug candidates a chain of filters and preprocessing modules has to be constituted.
On the other hand, the preparation of the receptor structure may be a very delicate problem. Especially in cases, where no 3D structure of a receptor protein is given, it has to be made available by homology modelling.
The results of both branches merge into docking a selected compound into the prepared receptor. Thereby, complex structures are generated, which in turn have to be judged with respect to their quality.
If a compound finally has passed through this sequence of calculations and then turns out as a promising candidate, its structure may be modified and then re-enter the process. This iteratively leads to an optimized structure, worth to investigate in further wet-lab experiments.
The computations mainly involve force field and MD calculations. It is necessary (and challenging) to choose compatible programs and to figure out the best adjustments for the particular system of interest.