Re: running Multiple-replicas metadynamics

From: Osvalds Verners (owv5005_at_psu.edu)
Date: Fri May 16 2014 - 14:21:28 CDT

Hello Giacomo,

I was going to use a single colvars configuration file as required by
LAMMPS 'fix colvars'. There is a multi-replica running option for LAMMPS
as well, but then I should ask what naming convention I should use for
the multiple replica input files, so that these would recognized by
LAMMPS (maybe a question for LAMMPS list). Thanks,

Osvalds

On Fri, 2014-05-16 at 15:00 -0400, Giacomo Fiorin wrote:
> Hello Osvalds, if you have already a working configuration file for a
> single replica metadynamics simulation, all you need to do is to make
> N copies of that configuration file, changing the replica ID on each.
> You would then run each replica independently from the others. The
> reason why it doesn't work for you is that maybe you're creating a
> single file with all of the replicas in it?
>
>
> One note, because this is namd-l and not lammps-users: although the
> syntax of what you're doing is the same in NAMD and in LAMMPS, in NAMD
> there are also available implementations of replica exchange and bias
> exchange methods as Tcl code, i.e. with a different syntax that is
> available in NAMD only.
>
>
> Giacomo
>
>
>
>
>
>
> On Fri, May 16, 2014 at 2:01 PM, Osvalds Verners <owv5005_at_psu.edu>
> wrote:
> Dear Sirs,
>
> I would like to ask if there is any published sample for a
> colvars input
> file that could be used with LAMMPS for running
> Multiple-replicas
> metadynamics. The problem that I have come across is how to
> set up the
> multiple replica definitions, which gives me an error about
> name
> repetition if identical metadynamics instances except for the
> replica id
> are defined.
> Thank you,
> --
> Osvalds Verners
> Graduate Student
> Department of Mechanical and Nuclear Engineering
> Pennsylvania State University
>
>
>
>
>

This archive was generated by hypermail 2.1.6 : Wed Dec 31 2014 - 23:22:25 CST