[gmx-users] radius of gyration in REMD of protein

[Justin Lemkul]

On 8/18/16 10:28 PM, YanhuaOuyang wrote:
> Hi, I am running a REMD of a protein,  which ranged from 270 to 500K in
> explicit water model with 60 replicas,  and I stop to analyse the REMD data
> after running for 50ns. I use the demux.pl script to get the continue the
> coordinate for each replica. Then I calculate the Rg value for each replica.
> However, the Rg fluctuation of each replica differs dramatically, some
> fluctuate dramatically, some converge well. Do someone knows the reason, or
> the fluctuation is normal ? How can I check my REMD is running correctlly?

Suddenly switching to a high temperature can (and usually does) induce large 
structural changes, hence the point of REMD.  This is why people most often 
(nearly always) focus on structure properties of the ensemble-continuous ground 
state.  Behavior at high temperature is usually very artificial and you usually 
care most that you are getting sufficient exchanges to cross barriers so that 
the ground-state sampling is enhanced.

It could also very well be that you have some disordered loops/termini that 
simply go a bit crazy at high temperature, which is not likely very relevant to 
the actual dynamics.  Watching the demuxed trajectory(ies) that are showing 
large Rg variation should point out the exact behavior.  Step 1 is always to 
watch the trajectory!

-Justin

> the commad of Rg calculation is below: gmx gyrate -s md_0_0.tpr   -f
> *_continue_trajectory_pbc_mol.xtc  -o gyrate_*.xvg
>
> Best regards, Ouyang
>

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

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

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