Physics Colloquium: Sampling transition paths: some applications to biological systems
Prof. Henri Orland, Institute for Theoretical Physics, Saclay, France
Transition paths are the stochastic paths that a system follows from one state of a system to another. In presence of high barriers, the transitions are exponentially rare events that require very long simulation times (Kramers time). However, most of the simulation time is spent waiting for the transition to occur.
On the other hand, the transition path time distribution (time over which the system changes its configuration) can be computed and it can be shown that these times are very short compared to barrier crossing times, thus allowing for short time simulations.
In this talk, I will show how transition paths can be formulated as conditioned stochastic paths, and how they can be generated exactly by a modified Langevin equation.
The method is illustrated on some simple analytical models, on the knotting-unknotting of circular DNA (in presence of topoisomerase) and on some allosteric transitions in proteins.
Event Organizer: Prof. Amiel Sternberg