[gmx-users] average structure

[Mark Abraham]

On 31/10/2011 7:39 PM, larifsofiene wrote:
> Greeting
> i'm doing a MD simulation for a monomeric protein over 10 ns, my 
> question is :
>
> * for my MD how do i know that it has equilibrated and suitable for 
> study ?,is it RMSD , Radius of gyration and potential energy stability 
> enough ? OR should i look for other parameters and do multiple 
> simulation with the same results ?

If those and any observables related to your point of interest look 
converged over time (compare averages from different parts of the 
trajectory) then things are probably fine. It depends how ordered your 
protein should be...

> * for simulation structure study should i use an average structure 
> (g_covar or grmsf) OR should i use the last trajectory from the 
> simulation (the 10 th ns trajectory )
>
> *if i use average structure should i delete hydrogen before looking 
> for hydrogen bonds ?,because average structure with it's original H 
> give abnormal H bonds.

See the http://www.gromacs.org/Documentation/FAQs about average structures.

Mark