Hi Hannes
Thanks so very much for the reply because I really need clarification.
Regarding the first question, I am confused between either one of these two:
1- minimize, heat up, equilibration. Then pick 'n' frames from the equilibration trajectory, and do SMD on each
OR
2- minimize, heat up, equilibration. Then do the production MD with the RC restrained to the initial value. After that, pick 'n' frames from the trajectory of the production MD, and do the SMD on each.
I also have the two following questions:
How can we get the entropy (difference) through this method?
How to check the standard deviation to see if it is more than a few KBT or not?
I already apologize if these sound basic. I really need guidance.
Kind Regards
Ramin
________________________________
From: Hannes Loeffler <Hannes.Loeffler.stfc.ac.uk>
Sent: Sunday, April 30, 2017 11:05:15 AM
To: amber.ambermd.org
Subject: Re: [AMBER] Restraint in Steered MD/Jarzynski method
On Sat, 29 Apr 2017 17:55:20 +0000
Ramin Salimi <ramin.salimi01.utrgv.edu> wrote:
> Dear Amber users
> I am trying to pull my DNA molecule through Jarzynski method.
> However, there are some things that I do not understand quite well:
>
> 1- when we equilibrate the system using the restraint to keep the
> reaction coordinate in the initial value, do we use nmropt? Because
> during the equilibration phase, I restrain the DNA using
> restraintmask option and relieve the restraint in a few steps by
> making the value of restraint_wt smaller and smaller until I remove
> the restrainmask. I am a bit confused what to do when it says to
> restrain the RC.
I am not sure what "it" you are referring to. But non-equilibrium
simulations still require you to start from a pre-equilibrated
ensemble. This means creating unbiased and, in principle, uncorrelated
starting points. So I would not know why you would want to restrain
your RC to a particular value.
> 2- when we pick a large number of frames from the equilibrated
> ensemble trajectory, do we need to strip solvent molecules to avoid
> the contribution of solvent-solute interactions, before starting SMD?
If you plan to do your simulation in solvent you will necessarily have
to continue from your equilibrated solvent starting points.
> 3- At every value of the RC, we have 'n' number of dist_vs_t files,
> then we average as such:
> D={exp(-w1/kT)+exp(-w2/kT)+........+exp(-wn/kT)}/n Then, delta F= -kT
> Ln(D) which gives the PMF at one specific value of the reaction
> coordinate. Is this method correct?
If you "pull" your system along a RC you will receive a work value for
every simulation you do. All this work values will have to be
exponentially averaged with the Jarzynski formula. So your formula
seems to be correct.
Cheers,
Hannes.
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Flists.ambermd.org%2Fmailman%2Flistinfo%2Famber&data=02%7C01%7Cramin.salimi01%40utrgv.edu%7C36d42ef325074f5d3f4808d48fe2d577%7C990436a687df491c91249afa91f88827%7C0%7C0%7C636291651783729608&sdata=Bs20%2BZ0IT%2BCQwZzLkgj4ILXBcatJZ8kpI%2BaT%2Bx9VqaA%3D&reserved=0
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sun Apr 30 2017 - 12:00:03 PDT
