[gmx-developers] REMD(replica-exchange molecular dynamics) + GROMACS

[masakatsu-ito at aist.go.jp]

Hi Developers,

Peter Tieleman wrote : (subject: Implementing Replica Exchange )
> We're planning to implement replica exchange in the next couple of 
> months in Gromacs.

I am very glad to know that REMD (replica-exchange molecular dynamics)
would be implemented in GROMACS, because I am also interested in doing
REMD simulations to investigate biomolecular processes like
ligand-receptor binding.

Since I have developed REMD toolkit which is designed as an
object-oriented framework in C++ to incorporate a force field
implementation, I would be happy if our toolkit contributes to those
who are interested in REM(replica-exchange method)/REMD simulations.
I think the toolkit can be combined with GROMACS to build a variant of
REM and REMD programs. Our REMD toolkit is going to be released under
GPL, and we have examined its extensibility and parallel performance
by combining it with Mindy (serial version of NAMD). But we have not
yet performed a large scale simulation which requires an efficient
force field implementation like GROMACS.

Now I am planning to write an adaptor class which receives a force
vector from GROMACS and pass atomic positions to GROMACS.  So could
somebody tell me how easy or difficult it will be to reuse GROMACS as
an software component which receives atomic positions and answer a
force vector and potential energy?

I would appreciate any suggestion or hints.

Thanks in advance.

Masakatsu Ito , Ph.D

Grid Technology Research Center
National Institute of Advanced Industrial Science and Technology
Tsukuba Central 2, Tsukuba, Ibaraki 305-8568 Japan
Phone : +81-29-861-5730  Fax : +81-29-861-5301
E-mail masakatsu-ito at aist.go.jp
URL http://met.sourceforge.net/