[gmx-users] Query on generate velocities and gen_temp in mdp files for REMD simulations

[Justin Lemkul]

On 1/16/13 3:56 AM, Neha Gandhi wrote:
> Dear Help,
>
> I am trying to run REMD simulations and found couple of interesting
> discussions on GROMACS mailing list regarding ensembles which can be used
> during REMD.
>
> I will be running REMD of small peptide in explicit solvent and I am
> confused about gen_temp and gen_vel parameters for these simulations below:
>
> 1. NPT ensemble at 310 K to get density right. What should be the
> parameters gen_vel and gen_temp set to?

That depends.  If you do prior equilibration, like NVT, then you should not be 
re-generating velocities upon moving to NPT.  If NPT is your first phase of 
equilibration, then gen_vel = yes and gen_temp is set to whatever the desired 
temperature is.  The same is true for any equilibration protocol.

> 2. Equilibration at each temperature -NVT ensemble. Although equilibration
> run will be carried out for a reference temperature range, should the
> gen_temp still be set to 310K?

If gen_vel = no, then the gen_temp setting is ignored.  You should only ever 
initialize velocities once, at the start of whatever the first phase of 
equilibration is.  Regenerating velocities later makes the previous 
equilibration pointless and may cause instabilities in your system.

> 3. Replica exchange -What should be gen_temp and gen_velocity set to?
>

See point #2.

-Justin

-- 
========================================

Justin A. Lemkul, Ph.D.
Research Scientist
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================