[gmx-users] Average simulation time for correct surface tension

[Justin Lemkul]

On 7/18/16 10:16 AM, gozde ergin wrote:
> Seems there is no systematic drift and trend but fluctuating is, please see on
>  http://imgur.com/Vfs0wqw <http://imgur.com/Vfs0wqw>
>

This is not what I suggested.  The raw time series shows lots of noise.  What 
happens when you plot the average of 1-ns chunks of time (with error bars), e.g. 
0-1, 1-2, 2-3, etc?  This helps you to determine if there are systematic changes.

>> How did you decide that this division of time was appropriate?
>
> This is exactly my question how can I decide the simulation time is enough or not. Also this surface tension value is not correct either.
> It should be around 1000-1100 not 1244 bar nm.
>

Well, this could be any number of things: insufficient sampling, poor topology, 
system size, etc.

-Justin

>
> On 18 Jul 2016, at 16:06, Justin Lemkul <jalemkul at vt.edu> wrote:
>>
>>
>>
>> On 7/18/16 6:07 AM, gozde ergin wrote:
>>> Dear users,
>>>
>>> I am trying to estimate the surface tension of cis-pinonic covered water surfaces.
>>> Li. et al 2010 did the same thing and his simulation time was 2 ns.
>>> On the other hand there are some other studies mentioned that simulation time should be long around 200 ns because surface tension is difficult to estimate accurately, since fluctuations are large and convergence is slow (Engin et. al 2010).
>>>
>>> I would like to give a an example below. I simulated the system for 20 ns. First 6 ns is for equilibration and last 14 ns sampling.
>>>
>>
>> How did you decide that this division of time was appropriate?
>>
>>> For first 6 ns :
>>>
>>> Energy                      Average   Err.Est.       RMSD  Tot-Drift
>>> -------------------------------------------------------------------------------
>>> #Surf*SurfTen               1181.02         22    2327.09    32.8309  (bar nm)
>>>
>>> For last 14 ns:
>>>
>>> Energy                      Average   Err.Est.       RMSD  Tot-Drift
>>> -------------------------------------------------------------------------------
>>> #Surf*SurfTen               1244.15         10    2339.36    -63.287  (bar nm)
>>>
>>> RMSD is larger than Average value so i am confused how long should I run?
>>>
>>
>> What happens if you plot your results in 1-ns chunks of time?  Any systematic drift or trend?
>>
>> -Justin
>>
>> --
>> ==================================================
>>
>> 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
>>
>> ==================================================
>> --
>> Gromacs Users mailing list
>>
>> * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!
>>
>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>
>> * For (un)subscribe requests visit
>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-request at gromacs.org.
>

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

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

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