[gmx-users] No group in index file?

[Justin Lemkul]

On 10/31/13 4:44 PM, Xu Dong Huang wrote:
> @ Justin,
>
> I did not define [angles] in the topology but I do have the angles result
> from All-atom run using OPLS forcefield. The reason for me not to include it
> in MARTINI forcefield topology is because I wanted to see if it will produce
> similar angle result of atom 1&2, 1&3 and 2&3 in martini water since the
> behavior of particle types I chose for my 3 atoms are well defined by martini
> at this point. My interest is to find for which force constant K value will
> the martini model match the all-atom model using OPLS.
>
> And you’re right, sorry, I got confused about the angles, there should only
> be 1. However, even the angle value I received doesn’t match the traditional
> all-atom result.
>
> What will happen if I define the [angles] part with the angles I received
> from traditional simulation? Because ultimately I am trying to find the k
> value parameter that will make the 3 particle display an angle similar to
> traditional run.
>

If you define an equilibrium angle for 1-2-3, it will oscillate harmonically 
around that value, with the distribution determined by the value of k.  I 
suspect that is how you can tune the parameters, by matching the distributions 
produced in the two simulations rather than a single target value, because with 
the angle defined in the topology, you're sort of predetermining the outcome.

-Justin

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

Justin A. Lemkul, Ph.D.
Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 601
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441

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