[gmx-users] Re: Output for rmsd and radius of gyration doubts

[Justin Lemkul]

On 3/4/13 10:37 AM, Ewaru wrote:
> Hi Justin,
>
> Thank you for your prompt reply! As this is my first approach to molecular
> dynamics, I still have lots of to understand but I'm constantly reading and
> trying to work things out. I predicted 2 structures, one using Modeller, and
> another one with I-Tasser. I followed your very informative tutorial for the
> approach of energy minimization and equilibration using Gromacs. My mistake
> for assuming that RMSD should fall within 0.1nm. I'm supposed to choose the
> best model to proceed with docking. Is it okay if I run just a 10ns
> simulation for both the predicted structure and then decide the best one by
> comparing the stability of the protein, from the output of the RMSD and
> radius of gyration?
>

Radius of gyration is probably not the best criterion here; I just demonstrated 
it in the tutorial because it is a rather fool-proof program.  RMSD is a decent 
indicator of structural changes, but by itself doesn't necessarily indicate 
convergence or tell you anything useful about the structure.  You need to apply 
some real biological knowledge before concluding which model is "best."  Since 
successful docking requires an accurate representation of the binding site, you 
need to probably examine the stability of critical residues, interactions among 
them, etc.  There are far more metrics that can be assessed than just trivial 
things like RMSD and Rg.  The tutorial is by no means comprehensive (nor does it 
claim to be), so spending some time in the literature and reading textbooks 
about these techniques should be prerequisite.

-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

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