[gmx-users] RE: Energy Minimization/ NpT settling problem

[Justin Lemkul]

On 12/21/12 2:57 PM, emmanuelle wrote:
> Thanks for your reply.
>
> 1.My cutoffs are OK. Indeed, changing them leads to the same problem.

This outcome points to a topology or configuration issue.

> 2.My original box is 20nm in length. I generate it using genconf to avoid
> any lack of memory that genbox could provide for a system of such a size.
> Then I use editconf to reach the required density of 1000g/l.

Scaling with editconf can shorten or elongate bonds and has never been a very 
robust approach in my hands.  It is better to generate a box that has a 
reasonable density and use sufficient equilibration to achieve the target value 
(within the accuracy of the force field being used).

> 2b.Atom 13090 is a H atom of an MEA molecule (HO-CH2-CH2-NH2). More
> specifically, it is the hydrogen directly bonded to the oxygen.
>
> Surprisingly, when I simulate a similar system for which I only substitute
> the monoethanolamine molecules by ethanolamine molecules, everything works
> perfectly fine.
>
> Could you help me on this?
>

Probably the editconf scaling adjusted the coordinates in a bad way.

-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

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