Hi, all
For replica exchange process in order to converge towards an equilibrium distribution, it is necessary to impose a detailed balance condition on the transition probability.
As the kinetic energy in Hamiltonion cancels out, the difference of total energy is in fact the difference of potential energy. REMD uses this potential energy difference to decide whether to change or not.
However, it seems like that in mmtsb tools, REMD still uses the the difference of potential energy which doesn't contain lambda. So will it influence the sampling since the Hamiltonion is supposed to be an extended one?
I get it. Thank your for reply