Molecular dynamics is often the only way to elucidate functional motions of biomolecules and in particular proteins. The probability of such transitions decreases exponentially with the height of the free energy barrier between the stable states. This often makes straightforward simulation of functional motions infeasible. There are many ways to improve the sampling of transitions, but most do not work effectively on complex proteins systems. We are working on improving sampling in biologically interesting systems by developing and applying methods, e.g. extended ensemble and replica exchange methods.