[gmx-users] Is there any difference with gromacs and other MD programs?

[Justin Lemkul]

On 1/22/18 12:43 AM, 설진규 (자연과학부) wrote:
> Dear gmx users,
>
>
> I’ve started to run MD simulation of ion aggregation of aqueous solution. (JCP 145, 174501 (2016))
>
> So I followed adding initial molecules(I gained BF4 ion topology from https://atb.uq.edu.au/molecule.py?molid=26889) , em, ensembles and md simulation. Then I calculated RDF but their g(r) values of O...O(figure 1-a in given journal) are slightly different between simulation on paper and I did.
>
> My simulation reports g(0.464)=1.044.
>
> I might guess the problem is that I used gromacs with gromos54a7 force field and the paper simulated with AMBER. I thought the molecules of the system is so simple that those force fields couldn’t be too different.

That's a bad assumption. Each force field is parametrized under 
different conventions and different target data, so you should not 
expect equivalent results.

>
> My questions are;
>
> 1. Are there any difference of calculating method between GROMACS and AMBERMD? If so, how can I find its details?

Refer to each program's reference manual and published literature. Every 
MD engine is a little different in its algorithms, features, 
optimizations, etc.

> 2. If I select AMBER force field and edit topology and run it by GROMACS, can I solve the simulation with completely same result with this journal?

Within statistical error, yes.

-Justin

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

Justin A. Lemkul, Ph.D.
Assistant Professor
Virginia Tech Department of Biochemistry

303 Engel Hall
340 West Campus Dr.
Blacksburg, VA 24061

jalemkul at vt.edu | (540) 231-3129
http://www.biochem.vt.edu/people/faculty/JustinLemkul.html

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