On Mon, 2014-11-03 at 21:12 +0800, 王漠野 wrote:
>
>
> Dear AMBER users,
>
> I had a REMD process with amber. One simulation time is not enough and
> I restarted the simulation of second times . I fixed the **.mdin
> files. Reset the parameters irest=1, ntx=5 and the groupfile commands
> were as follows:
>
> -A -rem 1 -remlog rem2.log -i remd.mdin.00X -o remd2.mdout.00X -c
> remd2.rst.00X -r remd2.rst.00X -x remd2.mdcrd.00X -inf
> remd2.mdinfo.00X -p protein.prmtop
>
> The simulation ended normally. Then I checked the rem.log file and the
> rem2.log file. Then I found that the exchange ratios was
> inconsecutive. The followed picture was the screen capture of my
> rem.log file and my rem2.log file:
This is normal. The only thing that sander or pmemd accept as input are
the mdin file, prmtop file, and the coordinate file. Nowhere in any of
those files is the "REMD exchange ratio" information stored, so there is
no choice but to start from "zero".
The only thing that "restarting" a REMD simulation does is take the
velocities *and temperature* from the coordinate file (the temperature
from the coordinate file replaces the one defined in the mdin file).
cpptraj from AmberTools 14 includes support for reading REMD log files
and doing more in-depth analysis of them. I would encourage you to look
into that if you need cumulative statistics over multiple log files.
HTH,
Jason
--
Jason M. Swails
BioMaPS,
Rutgers University
Postdoctoral Researcher
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Nov 03 2014 - 06:30:02 PST
