[gmx-developers] Fluctuations printed by mdrun

[Justin A. Lemkul]

Christian Mötzing wrote:
> Hi, I have to re-open this issue of mine.
> 
> In Gromacs 4.0.x the fluctuation output looked like this.
> <---
>     <======  ###############################  ==>
>     <====  R M S - F L U C T U A T I O N S  ====>
>     <==  ###############################  ======>
> 
>    Energies (kJ/mol)
>         LJ (SR)        LJ (LR)   Coulomb (SR)   Coul. recip.
> Potential
>     8.83363e+02    6.11178e+00    3.11788e+03    1.40733e+02    2.38529e
> +03
> <---
> 
> Was my assumption correct that the Potential column lists the
> fluctuation of the potential energy? Does the RMSD value from g_energy
> in 4.5.x correspond to this value?
> 
> g_energy gives me this output:
> 
> <---
> Last energy frame read 80 time   80.000           
> 
> Statistics over 80001 steps [ 0.0000 through 80.0000 ps ], 1 data sets
> All statistics are over 16001 points
> 
> Energy                      Average   Err.Est.       RMSD  Tot-Drift
> -------------------------------------------------------------------------------
> Potential                  -36331.7        290    1329.64   -2056.31
> (kJ/mol)
> <----
> 
> Im not sure what this tools wants to tell me. Does "Last energy frame
> read..." mean that the last frame was copied 80 times? "Statistics over

No, that means g_energy has identified the last frame in your .edr file as one 
corresponding to time = 80 ps.

> 80001 steps..." sound good to me as I simulated 80000 steps. But what
> does 1 data set mean? Did it only find 1 set of data to work with in
> this 80000 steps? The nstenergy value was set to 1000 so I hoped to have
> 80 data sets.
> 

The nstenergy setting is simply output control, indicating that you will be able 
to output (i.e. in an .xvg file) the values of different energies that are saved 
every 1000 steps.  You don't get 80 data *sets*, you get 80 data *points*, 
although the averages in the .edr file are exact over all MD steps.

> The reason why I am asking is that our results did not turn out as I
> expected. I simulated the waterbox with stepsize (dt) going from 0.00025
> to 0.004. I was told that the fluctuation should go up with growing
> stepsize. But it does not. It stays perfectly constant.
> 

Fluctuations are dependent upon the input .mdp settings.  I suppose that if you 
use accurate methods for nonbonded interactions and (especially) contraints for 
bonds, then I don't see how the fluctuations should inherently go up with the 
step size.

-Justin

> Any help is appreciated.
> 
> Christian
> 
> 
> 
> 
> Am Mittwoch, den 01.12.2010, 18:55 +0100 schrieb Berk Hess: 
>> You don't need any option to g_energy.
>> RMSD are the fluctuations that used to be printed in the md.log file.
>> These are fully accurate RMSD's over all MD steps (not only the ones you
>> get in energy.xvg).
>> Tot-Drift is the drift over the total time of a linear fit to the energy.
>>
>> Berk
>>
>> On 12/01/2010 06:43 PM, Christian Mötzing wrote:
>>> Hi, I looked into g_energy. There is a -aver and -fluc option. Both give
>>> me the same output:
>>>
>>> Energy                      Average   Err.Est.       RMSD  Tot-Drift
>>> -------------------------------------------------------------------------------
>>> Potential                  -36331.7        290    1329.64   -2056.31
>>> (kJ/mol)
>>>
>>>
>>> Is the RMSD the fluctuation? What ist Tot-Drift?
>>> Why do both option give the same ouput? Or am I looking in the wrong
>>> ouput? With both options nothing seems to change in the generated
>>> energy.xvg
>>>
>>> Christian
>>>
>>> Am Montag, den 29.11.2010, 12:47 +0100 schrieb Berk Hess: 
>>>   
>>>> I removed it because it requires storing more history over the whole
>>>> system in checkpoint files.
>>>> g_energy will print this for you.
>>>>
>>>> Berk
>>>>
>>>> On 11/29/2010 12:03 PM, Christian Mötzing wrote:
>>>>     
>>>>> Hi,
>>>>>
>>>>> in version 4.0.x mdrun printed the fluctuations of a simulation (I am
>>>>> doing a waterbox simulation) to the log file.
>>>>>
>>>>> In version 4.5.x this output is gone and I can't find a way to turn it
>>>>> on again.
>>>>>
>>>>> How do I get this kind of output in 4.5.x?
>>>>>
>>>>> Thx
>>>>>
>>>>> Christian
>>>>>
>>>>>   
>>>>>       
>>>>     
>>>   
> 
> 

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

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

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