[gmx-users] Potential energy from a previous configuration mismatch

[Justin Lemkul]

On 11/29/12 9:59 AM, juan-manuel.castillo at mv.uni-kl.de wrote:
> Dear Gromacs friends,
>
> I am running some simulations of a hexane/ethanol mixture in a box, using a
> United-Atom OPLS model. My Gromacs version is the 4.5.5.
>
> I fill the simulation box with genbox (I have also used an equilibrated pdb file
> generated by other simulation code), and I run a energy minimization simulation,
> wich gives the following energies for the final configuration:
>
>     Energies (kJ/mol)
>            Angle   Fourier Dih.          LJ-14    Coulomb-14        LJ (SR)
>      6.62915e+00    5.17493e+02   -2.61388e+01   0.00000e+00   -4.01422e+03
>    Disper. corr.   Coulomb (SR)       RF excl.     Potential Pres. DC (bar)
>     -9.65799e+01    0.00000e+00    0.00000e+00  -3.61282e+03    0.00000e+00
>
> Potential Energy  = -3.6128213e+03
> Maximum force     =  2.7169607e+01 on atom 541
> Norm of force     =  3.9986315e+00
>
> I take the final configuration, and using exactly the same force field
> definition and simulation parameters (cutoff radious, electrostatic method,
> etc), I start a molecular dynamics simulation in the NVE ensemble. No matter
> what I do, I always have a blow-up problem. I have checked all the points that
> could be giving the problem, in particular the ones given here:
>
> http://www.gromacs.org/Documentation/Terminology/Blowing_Up
>
> and I have also tried to modify one by one different parameters, but I never
> succeed to obtain a running simulation. I have seen that I always have a much
> larger energy at the beginning of the simulation than the energy at the end of
> the minimization:
>
>     Energies (kJ/mol)
>            Angle   Fourier Dih.          LJ-14    Coulomb-14        LJ (SR)
>      5.43743e+04    2.06120e+03    6.02532e+01   0.00000e+00    1.62817e+07
>    Disper. corr.   Coulomb (SR)       RF excl.     Potential    Kinetic En.
>     -9.65799e+01    0.00000e+00    0.00000e+00   1.63381e+07    1.65906e+10
>     Total Energy    Temperature Pres. DC (bar) Pressure (bar)   Constr. rmsd
>      1.66069e+10
>
> but the potential energy should be exactly the same in both cases, as it only
> depends on the atomic positions. Anybody can tell me what I am doing wrong?
>

Can you post a complete .mdp file?  In theory, if one evaluates the same 
configuration under the same parameters (i.e. in a single-point energy 
evaluation), yes, the energies reported should be the same.  But doing EM and 
doing MD are not the same, especially if you are introducing constraints or 
other parameters.

-Justin

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

Justin A. Lemkul, Ph.D.
Research Scientist
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

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