Investigating Protein-Protein and Protein-Ligand Interactions by Molecular Dynamics Simulations

In recent years, the earlier view of proteins as relatively rigid structures has been replaced by a dynamic model in which the internal motions and resulting conformational changes play an essential role in their function. In this context, molecular dynamics (MD) simulations have become an important computational tool for understanding the physical basis of the structure and function of biological macromolecules. Also in the process of finding new drugs MD simulations play an important role. Our workgroup uses molecular dynamics simulations to study proteins of biological and medical relevance, e.g. signal transduction proteins or human integrin complexes. The general aim of these investigations is to find possible new lead structures or drugs and also to understand the basic and essential mechanisms behind the mode of action of our target systems. In MD simulation, the problem size is fixed and a large number of iterations must be executed, so the MD simulation suites have to scale to hundreds or thousands CPUs to get detailed view inside biomolecular systems. The used programs AMBER and GROMACS scale well up to 64 or 32 CPUs, respectively. A typical run for about 100 ns simulation time consumes 5500 up to 21000 CPU hours.