Conformational sampling via a self-​regulating effective energy surface

The difficulty of efficiently sampling the phase space of complex systems with rough energy surfaces is well known. Typical solns. to the problem involve accelerating the crossing of barriers, but such methods often have the secondary problem that the low-​energy states of interest are inadequately sampled, unless the parameters of the search algorithm are modified as the system evolves. A method is presented to improve the sampling with particular emphasis on the low-​energy conformations, which make the most important contributions to the thermodn. of the system. The algorithm proposed here samples the details of the min., while easily surmounting barriers. This is achieved by introducing a self-​regulating sampling variable which depends on the current state of the system. Two replicas of the system are introduced and the sampling variable is treated as a particle coupled to the phys. system. The method is illustrated with a simple model system and is applied to the realistic example of barrier crossing in a protein-​ligand complex.