[gmx-users] combining umbrella sampling with replica exchange

[Sanku M]

Hi,
  I sent this email few weeks back and thought sending it again in case someone can give some opinion.
I have a technical question regarding feasibility of  combining replica exchange with umbrella sampling or any other pulling simulations in gromacs.
Since the umbrella sampling or any other pulling simulations are non-equilibrium simulation due to presence of external bias, I wonder whether the detailed balance in  replica exchange simulation  will be maintained .
I ask this because replica exchange  uses Monte Carlo metropolis algorithm to swap configurations and so I wanted to know what is the energy-difference ( New energy - old energy) in metropolis algorithm gromacs will use : Is it the energy difference after taking into account the biasing potential or is is only the internal potential energy-difference ( i.e not including biasing potential ) ?

Any response from the users or developer will be highly appreciated.

Thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20120103/69641d4c/attachment.html>