[gmx-users] Why potential energy not in negative value?

[Justin Lemkul]

On 10/31/12 4:10 AM, Nur Syafiqah Abdul Ghani wrote:
> Dear All,
>
> Currently I'm doing the minimization part for my solvent which is
> hexafluoroisopropanol with water.
> I'm packing them with packmol then use the forcefield gromos96.
> Now  by using the command
> grompp -f em1.mdp -c minimize_301012.gro -p topol.top -o minimize_311012.tpr
> mdrun -deffnm minimize_311012 -v
>
> And its give the result as follow :
>
> writing lowest energy coordinates.
>
> Back Off! I just backed up minimize_311012.gro to ./#minimize_311012.gro.3#
>
> Steepest Descents converged to Fmax < 15 in 36 steps
> Potential Energy  =  1.88042503678994e+05
> Maximum force     =  1.46843345643871e+01 on atom 30092
> Norm of force     =  1.94397813942079e+00
>
>
> I already minimize it with the bigger step that use about 10ns.
> Next I increase the step which take about 15ns and the result show
> same as above..
>

Note that EM is not run in time, so your settings below are misleading.  The 
only settings that matter are nsteps and emstep; dt is ignored completely when 
running EM.  Note that your cutoff settings are incorrect for Gromos96, as you 
say you are using.  How did you generate the HFIP topology?

The positive potential indicates that your system has a net repulsive energy, so 
the configuration isn't very happy.  The maximum force is very good, but there 
may be unsatisfied interactions that yield the positive energy.  Perhaps a 
gentle equilibration will resolve the situation.

-Justin

> My mdp file was like this :
>
> ;
> ;       Preprocessing
> title                    = STEEP
> cpp                      = /usr/bin/cpp
> include                  = -I/usr/local/gromacs/share/gromacs/top
> define                   = -DFLEXIBLE
> ;
> ;       Run Control
> ;
> integrator               = steep
> nsteps                   = 75000000 ; 75000000 * 0.2 = 15000 ps = 15 ns
> dt = 0.0002 ; 0.2 fs
> ;
> ; Energy Minimization
> emstep                   = 0.0001 ; used with steep
> emtol                    = 15
> nstcgsteep               = 500 ; do a steep every 10 steps of cg
> energygrps               = System
> ;
> ;       Neighbor Searching
> nstcomm                  = 1
> nstlist                  = 10
> ns_type                  = grid
> pbc                      = xyz
> rlist                    = 1.2
> ;
> ; Electrostatics
> coulombtype              = pme
> rcoulomb                 = 1.2
> rcoulomb-switch          = 0
> vdw-type                 = cut-off
> rvdw                     = 1.2
> rvdw-switch              = 0
> fourierspacing = 0.2
> ;
> ; Output Control
> nstenergy = 10
> optimize_fft = yes
> ;
> ;       OPTIONS FOR BONDS
> ; OPTIONS FOR BONDS
> constraints              = none
>
>
> Is the potential energy need to be in negative value for my simulation?
> What happened if it not in negative value?Can I just proceed to the
> next step for equilibration?
> I already find and read the paper that relate to my simulation and
> they did not mention the energy that they got from their
> experiment.
>
> Thanks in advance u guys,
> --
> Best Regards,
>
> Nur Syafiqah Abdul Ghani,
> Theoretical and Computational Chemistry Laboratory,
> Department of Chemistry,
> Faculty of Science,
> Universiti Putra Malaysia,
> 43400 Serdang,
> Selangor.
> 013-7188131
> alternative email : syafiqahabdulghani at gmail.com
>

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

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

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