[gmx-users] zero value for non-Protein components in energygrps

[Mohammad Hassan Khatami]

Thanks Justin!

That was the issue.

MHK


On 2017-10-31 01:35 PM, Justin Lemkul wrote:
>
>
> On 10/31/17 12:07 PM, Mohammad Hassan Khatami wrote:
>> Hi,
>>
>> I have an issue with calculating energy groups in GMX2016.4, while 
>> running a simple protein in water tutorial using the default oplsAA 
>> forcefield in gromacs. I searched my problem it in the mailing list 
>> but did not find my answer.
>>
>> In addition to the default tutorial results, I wanted to print out 
>> the energy values for different components of the simulation, i.e., 
>> protein and water, in the .edr file. So I added the line below to the 
>> .mdp file to calculate values for different energy groups:
>>
>>
>> energygrps = Protein SOL
>>
>> After running a short simulation, I calculated the different energy 
>> values and the output is as follow:
>>
>> Energy                      Average   Err.Est.       RMSD Tot-Drift
>> ------------------------------------------------------------------------------- 
>>
>> LJ-14                           3004.95        5.3     54.675 
>> 44.9933  (kJ/mol)
>> Coulomb-14                 7927.05        4.1    44.0138 -4.19923 
>> (kJ/mol)
>> LJ (SR)                         21978.8        110    500.505 
>> 752.675  (kJ/mol)
>> Coulomb (SR)               -208384        220    1438.51 -1321.11 
>> (kJ/mol)
>> Coul-SR:Protein-Protein -208384        220    1438.51   -1321.11 
>> (kJ/mol)
>> LJ-SR:Protein-Protein      21978.8        110    500.505 752.675 
>> (kJ/mol)
>> Coul-14:Protein-Protein   7927.05        4.1    44.0138 -4.19923 
>> (kJ/mol)
>> LJ-14:Protein-Protein       3004.95        5.3     54.675 44.9933 
>> (kJ/mol)
>> Coul-SR:Protein-SOL               0          0          0 
>> 0                (kJ/mol)
>> LJ-SR:Protein-SOL                   0          0 0          0         
>>     (kJ/mol)
>> Coul-14:Protein-SOL               0          0          0 0        
>> (kJ/mol)
>> LJ-14:Protein-SOL                   0          0 0          0         
>>      (kJ/mol)
>> Coul-SR:SOL-SOL                    0          0 0          0        
>> (kJ/mol)
>> LJ-SR:SOL-SOL                        0          0 0          0     
>>            (kJ/mol)
>> Coul-14:SOL-SOL                    0          0 0          0        
>> (kJ/mol)
>> LJ-14:SOL-SOL                        0          0 0 0                 
>> (kJ/mol)
>>
>> Here, all the energy values for the Protein-SOL and SOL-SOL are zero, 
>> which is not what I expected. Moreover, the general LJ-14, 
>> Coulomb-14, LJ (SR) and Coulomb (SR) are exactly the values 
>> calculated for only protein.
>>
>> When I run them individually, i.e., only water system and only 
>> protein system, I get non-zero values, which are expected.
>>
>> Is there any other flag except  "energygrps" that I forgot to 
>> implement in my mdp file? and why is it only calculating the values 
>> of the protein  in the general LJ-14, Coulomb-14, LJ (SR) and Coulomb 
>> (SR) energies.
>>
>
> If you did the simulation on GPU, energygrps are not supported. Do an 
> mdrun -rerun on the trajectory using just CPU hardware to obtain these 
> energies.
>
> -Justin
>