Re: [AMBER] relaxation time to maintain quasi-static condition

From: Adrian Roitberg <roitberg.ufl.edu>
Date: Tue, 20 Jun 2017 14:29:56 -0400

Hi

I believe you need to dig deep into the literature before trying these
type of calculations.

What are you trying to do ?

Just pull? Pull and get good structures? pull and get free energies?

All of those things have been done by many others, so you need to go
check their papers and ask why they did what they did.

I believe your confusion has to do with your first question:

"But wait, doesn't it mean that I am only 0.5 A0 far from the equilibrium if I have pulled it in 0.5 A0 pulling increments?"

You seems to believe that you can ONLY be at equilibrium when you pulled
zero angstroms. That is NOT correct. You are pulling, and for EACH
distance you would like to have ALL other degrees of freedom in your
system, be equilibrated. How do you do that? only ONE way: run an
infinite amount of time.
Since that is not possible, you need to come up with your own measure or
reuse someone else's.

Adrian

On 6/20/17 12:40 PM, Ramin Salimi wrote:
> Dr. Roitberg
>
> Thanks for the response
>
> So it follows:
>
> 1- How to figure out how far out out of equilibrium I am after pulling?? I mean is there any postprocessing method for that? But wait, doesn't it mean that I am only 0.5 A0 far from the equilibrium if I have pulled it in 0.5 A0 pulling increments?
>
> 2- During the relaxation period, as you correctly mentioned, I heard some do energy minimization during the relaxation period, and some others do equilibration. Which is that? Additionally, after each pullings, do I need to restrain the molecule to the new length with harmonic restraints during relaxation periods?
>
> 3- So you finally suggested to try different rates for both pulling, and relaxation periods. Then how to check if it is still in equilibrium or not?(I think this question partially overlaps with #1, my apologies)
>
> Best,
>
> Ramin
>
>
> ________________________________
> From: Adrian Roitberg <roitberg.ufl.edu>
> Sent: Tuesday, June 20, 2017 7:48:05 AM
> To: AMBER Mailing List
> Subject: Re: [AMBER] relaxation time to maintain quasi-static condition
>
> It might be worse tan you think.
>
> The amount of time you need to spend during what you call
> 'equilibration' depends very much on how fast you pull during that
> pulling part of the simulation. For example, if you pull 1 A very
> slowly, then you do not need to spend much time equilibrating. If you
> pull very fast, then you better spend lots of time equilibrating
> (technically, relaxing).
>
> So, it is not only the relaxation time of your system you need to worry
> about, but more importantly, you need to figure out how far out of
> equilibrium you are after pulling.
>
> Unfortunately, there are no formulas or set recipes for any of this, you
> need to try different ways.
>
> adrian
>
> On 6/20/17 8:28 AM, David Case wrote:
>> On Tue, Jun 20, 2017, Ramin Salimi wrote:
>>> I am trying to pull my DNA molecule in increments of 1 A0, and then
>>> run equilibration for 50ps, then do another pull for 1 A0, then do the
>>> 50ps equilibration, and keep doing it until the molecule reaches my
>>> target length. My primary purpose is to keep the system in quasi-static
>>> condition as much as possible.
>>>
>>> So, based upon what I understood, to keep the system in quasi-static
>>> during both pullings, and equilibration periods, the equilibration
>>> periods between increments along with pulling periods should be larger
>>> than the relaxation time of the system.
>> Generally, there is no one number that qualifies as "the" relaxation time for
>> a system. I'm no expert on pulling simulations, but my guess is that you will
>> have to try different equilibration periods to examine convergence.
>>> Does it have anything to do with correlation time, and the use of IRED
>>> vectors, and IRED matrices in cpptraj?
>> No.
>>
>> ....dac
>>
>>
>> _______________________________________________
>> 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%7C66b2a79fa60e43c069d108d4b7da7db0%7C990436a687df491c91249afa91f88827%7C0%7C0%7C636335596423429301&sdata=8KuAcW1sThBofutm0DdqzEv1CW%2F31o47LINxGf0gAoE%3D&reserved=0
> --
> Dr. Adrian E. Roitberg
> University of Florida Research Foundation Professor
> Department of Chemistry
> University of Florida
> roitberg.ufl.edu
> 352-392-6972
>
>
> _______________________________________________
> 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%7C66b2a79fa60e43c069d108d4b7da7db0%7C990436a687df491c91249afa91f88827%7C0%7C0%7C636335596423429301&sdata=8KuAcW1sThBofutm0DdqzEv1CW%2F31o47LINxGf0gAoE%3D&reserved=0
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber

-- 
Dr. Adrian E. Roitberg
University of Florida Research Foundation Professor
Department of Chemistry
University of Florida
roitberg.ufl.edu
352-392-6972
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Jun 20 2017 - 11:30:03 PDT
Custom Search