[gmx-users] MD after equilibration phase

[Justin A. Lemkul]

bipin singh wrote:
> Thanks for your comments
>  I have a very crude question which may not be even logical to ask here 
> but I just want to know your opinion on this:
> 
> If I will perform two simulations in following two ways :
> 
> *RUN:1*
> 
> (1) Did NVT with POSRES on protein and
>     gen_vel=yes
>     gen_temp= 350
> 
> (2) Did NPT with POSRES on protein and
>     continuation=yes
>     gen_vel=no ;Without reading velocities from NVT
> 
> (3) Did MD with No POSRES (NPT)
>      continuation=yes
>      gen_vel=no ; No velocities read from previous run
> 
> 
> *
> RUN:2
> 
> *(1) Did NVT with POSRES on protein and
>     gen_vel=yes
>     gen_temp= 350
> 
> (2) Did NPT with POSRES on protein and
>     continuation=yes
>     gen_vel=no ; Reading velocities from NVT run using -t option
> 
> (3) Did MD with No POSRES (NPT)
>      continuation=yes
>      gen_vel=no ; Velocities read from previous run using -t option
> 
> 
> 
> What will be the expected flaws of the RUN:1 over RUN:2 and will the 
> results got from the RUN:1 is unreliable ? If yes why ?
> 

I doubt there is a sound way to predict the errors that will arise.  In the case 
of method #1, you obliterate any benefit of prior equilibration.  At the outset 
of your unrestrained MD, your equilibrated ensemble is no more.  Thus, your 
system will essentially start at some undefined ensemble and require a long time 
to re-establish equilibrium.  There may be algorithmic problems with thermostats 
and barostats as they try to establish the target values, which will depend on 
exactly which algorithms you're using.

-Justin

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

Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

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