Equilibrium sampling by weighting non-equilibrium simulation trajectories
* SPEAKERS
Name
Affiliation
E-mail
Cheng Yang
UCAS
* HOST(Applicant)
Name
Affiliation
E-mail
Yongseok Jho
APCTP
yongseokjho(at)gmail.com
* DATE / TIME
2015-04-22, 16:30pm
* ABSTRACT
When the free energy surface of a physical system is very rugged, it is expensive to extract its equilibrium information with standard molecular dynamics simulations. On the other hand, a non-equilibrium trajectory with the system Hamiltonian changing over time can explore the whole conformational space more efficiently. In this work, we develop a novel simulation method, the weighting non-equilibrium trajectories (WNT) method, to obtain a system’s equilibrium information according to the Re-weighted Ensemble Dynamics method (RED) (L.Gong, et al., Phys. Rev. E. 80, 026707, 2009). The WNT method can produce the global equilibrium distribution via giving each non-equilibrium trajectory a weight which can be calculated through the simulation data. The test runs on two particular systems, one-dimensional potential and Lennard-Jones potential systems, demonstrate that this method is more efficient in sampling than the RED method.
