[gmx-users] Free Energy Calculations

[Justin Lemkul]

On 7/17/15 3:35 AM, Live King wrote:
> Hi Everyone,
> I use ancient version of Gromacs 4.5.5 for free energy calculations. I have
> two different questions.
>
> 1) When I run g_bar for results, I get this
>
> lambda  0.000 -  0.050,   DG -2.16 +/-  0.00
> lambda  0.050 -  0.100,   DG -2.16 +/-  0.01
> lambda  0.100 -  0.150,   DG -2.09 +/-  0.01
> lambda  0.150 -  0.200,   DG -1.98 +/-  0.01
> lambda  0.200 -  0.250,   DG -1.83 +/-  0.01
> lambda  0.250 -  0.300,   DG -1.59 +/-  0.00
> lambda  0.300 -  0.350,   DG -1.26 +/-  0.00
> lambda  0.350 -  0.400,   DG -0.80 +/-  0.00
> lambda  0.400 -  0.450,   DG -0.17 +/-  0.00
> lambda  0.450 -  0.500,   DG  0.67 +/-  0.00
> lambda  0.500 -  0.550,   DG  1.79 +/-  0.00
> lambda  0.550 -  0.600,   DG  3.30 +/-  0.01
> lambda  0.600 -  0.650,   DG  4.35 +/-  0.11
> lambda  0.650 -  0.700,   DG  1.99 +/-  0.10
> lambda  0.700 -  0.750,   DG  0.61 +/-  0.04
> lambda  0.750 -  0.800,   DG  0.04 +/-  0.02
> lambda  0.800 -  0.850,   DG -0.44 +/-  0.04
> lambda  0.850 -  0.900,   DG -0.46 +/-  0.02
> lambda  0.900 -  0.950,   DG -0.43 +/-  0.04
> lambda  0.950 -  1.000,   DG -0.49 +/-  0.02
>
> total   0.000 -  1.000,   DG -3.13 +/-  0.21
>
>
> Where these results are stored. I don't seem to find them anywhere.
> Barint.xvg seems to contain something else. Please let me know if they get
> stored or not and If they don't please let me know how to store it ? ( of
> course without copy and pasting into the text file).
>

g_bar -f *.xvg | tee bar.out

>
> 2) The results I showed above always show this warning :-
>
>
>
>
> *"WARNING: Some of these results violate the Second Law of Thermodynamics:
>           This is can be the result of severe undersampling, or (more
> likely)         there is something wrong with the simulations".*
>
> I have checked for both possible suggestions.I don't seem to find anything
> wrong with my simulations. Sometimes this warning goes away if I increase
> the time span. But, most of the times it keeps coming back. What does that
> mean ? How can I get rid of that ?
>

The warning is kind of cryptic, but in my experience indicates either 
insufficient sampling, inadequate topology, or wrong .mdp settings.

-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

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