Application of discrete molecular dynamics to protein folding and aggregation

With the rapid increase in computational speed and memory, simulations of proteins and other biological polymers begin to gain predictive power. However, in order to simulate a folding trajectory of a moderate size protein or an aggregation process of a large number of peptides, traditional molecular dynamics methods based on explicit solvent and accurate force field models still must gain several orders of magnitude in speed. Under these circumstances, simplified models which capture the essential features of the system under study may shed light on the problem in question. One of these simplified methods is discrete molecular dynamics (DMD). DMD replaces the interaction potentials between atoms and covalent bonds by discontinuous step functions. This simplification as well as coarse graining of the model (replacing groups of atoms by one effective bead) and replacing the effect of solvent by varying the strength of inter-bead interactions can speed up simulations sufficiently to generate many folding-unfolding events and to track the aggregation of many peptides. This increase in speed is gained mainly due to the ballistic motion of either secondary structures of the protein or individual peptides. This ballistic motion is a characteristic feature of the DMD method. This chapter will review successes and failures of the DMD method in protein folding and aggregation.