[gmx-users] results of md

[Justin Lemkul]

On 12/4/12 4:15 PM, mohammad agha wrote:
> Dear Justin,
>
> Thank you very much from your response.
> It means that this problem is not important. Should I use -reprod option in running of md.mdp? I want to obtain micelles as monodisperse with identical Nagg(aggregation number) in the one simulation, but distribution of sizes is broad and it is opposite of experimental results.

The -reprod option only turns off various optimizations to help diagnose 
potential bugs.  It is not relevant here.  MD is chaotic; even the exact same 
.tpr file on the same hardware will not necessarily produce binary identical 
results.  Hence the sampling issue.

> Can you help me about this problem, Please?
>

I have no expertise in micelle simulations.  In general, either longer 
simulations or several more simulations of intermediate length will give you 
better sampling from which you can gather more statistically reliable estimates 
of experimental observables.

Note that the quality of the topology will also dictate whether or not the 
results will reflect reality, so proper force field derivation and validation is 
necessary before any results are trustworthy.

-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

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