Re: amber-developers: REMD and PTRAJ manuals

From: Daniel Roe <>
Date: Wed, 23 Apr 2008 12:29:28 -0400

On Wed, Apr 23, 2008 at 12:12 PM, David A. Case <> wrote:

> Thanks. I've put this information into amber11/doc/AmberTools.lyx, so it
> will
> be available at the next release.
> Do replica exchange runs automatically create the right sort of trajectory
> to be used in this way? Is there an example in the manual of what to do,
> or
> (better) a tutorial that runs replica exchange with the new procedures?

The new format REMD trajectory is automatically generated when rem>0 now.
This is necessary because of the new implementation - if the REMD header
were not written there would be no way of telling what temperature any
given coordinate frame is at. So there's nothing a user needs to do except
be aware that this new format is used only for REMD runs. The changes are
described in section 4.5.1 of the amber manual.

I should mention one potentially significant issue: this is only for
formatted writes, i.e. netcdf trajectories do not get this REMD information.
I'm not familiar enough with the netcdf format to know if this would be an
easy thing to incorporate into a netcdf trajectory.


Daniel R. Roe, Ph.D.
Research Chemist
National Institute of Standards and Technology
100 Bureau Drive, Stop 8443
Gaithersburg, MD 20899-8443
(301) 975-8741
Received on Sun Apr 27 2008 - 06:07:15 PDT
Custom Search