[gmx-users] pulling time

[Justin Lemkul]

On 10/5/16 5:24 PM, Alexander Alexander wrote:
> Dear gromacs user,
>
> I was wondering how long normally should be the pulling time in one step
> before the final umbrella sampling?
> For instance in the Aβ42 protofibril umbrella sampling used in tutorial,
> the applied simulation times are as following referring to the original
> paper:
>
> Production Normal MD: 100 ns
> Pulling time :  500 ps
> MD in each windows : 10 ns
>
>

These are like any other simulation.  They need to be sufficiently long to 
converge the properties of interest.

> The second question is that can we limit the pulling region to pull away
> for example A from B as 3 nm in the pulling run(not in the final umbrella
> sampling)?
>

The length of the reaction coordinate should be sufficient to obtain 
"non-interacting" species to fully describe the binding free energy.  Of course, 
such states are not generally possible in MD with PME since there are always 
long-range forces, but if the molecules are sufficiently beyond the short-range 
cutoffs, it's usually good enough.

-Justin

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

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

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