Monte Carlo simulation is commonly used to compute pathways and thermodynamic properties of proteins. A simulation run is a series of random steps in conformation space, each perturbing slightly some degrees of freedom of the molecule. An attempted perturbation is accepted with a probability that depends on the resulting change in the molecule's energy. This work investigates general algorithmic
techniques to accelerate Monte Carlo simulation by speeding up the update of the energy function during simulation, without changing the step generator or the outcome of the acceptance test. It also investigates applications of these techniques, for example, to quickly generate compact conformations.