[gmx-users] Energy Minimisation and Equilibration Problems

Thu Aug 6 17:58:20 CEST 2009

The energies simply do not seem to come down any further within several
thousand steps.

I start with a .mol2 file which contains the structure of ethylene glycol
(ethanediol).  These are the commands that I use to set up and run the
minimisation:

$ .../topolbuild1_2_1/src/topolbuild -n ethanediol -dir
.../topolbuild1_2_1/dat/gromacs -ff gmx53a6

the above command outputs the following files:

ethanediol.gro
ethanediol.log
ethanediolMOL.mol2
ethanediol.top
ffethanediol.itp
posreethanediol.itp

I then proceed to enlarge the box and solvate the molecule:

$ editconf -f ethanediol.gro -o ethanediol_box.gro -box 5 5 5

$ genbox -cp ethanediol_box.gro -cs spc216.gro -o ethanediol_box.gro -p
ethanediol.top -shell 1

I then use grompp to configure the minimisation:

$ grompp -f minim.mdp -c ethanediol_box.gro -p ethanediol.top -o em.tpr

This is my .mdp file for minimisation:

===============================
define        = -DPOSRE
integrator    = steep
emtol        = 10.0
emstep          = 0.01
nsteps        = 2000

nstlist        = 1
ns_type        = grid
rlist        = 1.0
coulombtype    = PME
rcoulomb    = 1.0
rvdw        = 1.3
pbc        = xyz

pme_order    = 4
constraints     = none

nstxout        = 1
nstvout        = 1
nstenergy    = 1
nstlog        = 1
nstcomm         = 1
Tcoupl          = no
Pcoupl          = no
gen_vel         = no
===============================

$ mdrun -v -deffnm em

the minimisation runs without error and these are the last few lines of
output:

===========================================
Step= 1990, Dmax= 1.1e-05 nm, Epot= -1.76427e+04 Fmax= 1.25524e+04, atom=
304
Step= 1991, Dmax= 1.3e-05 nm, Epot= -1.76433e+04 Fmax= 3.13881e+04, atom=
484
Step= 1992, Dmax= 1.6e-05 nm, Epot= -1.76443e+04 Fmax= 2.11785e+04, atom=
484
Step= 1993, Dmax= 1.9e-05 nm, Epot= -1.76445e+04 Fmax= 4.22489e+04, atom=
484
Step= 1994, Dmax= 2.3e-05 nm, Epot= -1.76456e+04 Fmax= 3.33021e+04, atom=
484
Step= 1996, Dmax= 1.4e-05 nm, Epot= -1.76468e+04 Fmax= 1.26110e+04, atom=
304
Step= 1997, Dmax= 1.6e-05 nm, Epot= -1.76474e+04 Fmax= 3.81619e+04, atom=
484
Step= 1998, Dmax= 2.0e-05 nm, Epot= -1.76486e+04 Fmax= 2.68240e+04, atom=
484
Step= 2000, Dmax= 1.2e-05 nm, Epot= -1.76496e+04 Fmax= 1.26501e+04, atom=
304

writing lowest energy coordinates.

Steepest Descents did not converge to Fmax < 10 in 2001 steps.
Potential Energy  = -1.7649566e+04
Maximum force     =  1.2650057e+04 on atom 30
Norm of force     =  1.7544460e+03
===========================================

If I proceed to equilibration after doing the above, the water molecules
simply fly apart (although not immediately).  Additionally, it seems that
there are no interactions between the waters during equilibration.

I have tried to run the minimisation for a larger number of steps, but it
does not help.  I have also tried to delete individual water molecules from
the structure files, but doing so simply causes the minimisation to "fixate"
on another two molecules.  I am not sure what values of the energies are
reasonable for this system, and how to minimise it further.  Please advise.

Thank you.

Nancy

On Thu, Aug 6, 2009 at 8:38 AM, Bruce D. Ray <brucedray at yahoo.com> wrote:

> On Wednesday, August 5, 2009 at 6:03:51 PM, Nancy <nancy5villa at gmail.com>
> wrote:
> > I am trying to run equilibration on my solvated ethylene glycol
> (ethanediol) system.  I started
> > with an ethanediol mol2 file from which topolbuild generated various
> files.  I used editconf
> > to enlarge to box of the "ethanediol.gro" file:
> >
> > $ editconf -f ethanediol.gro -o ethanediol_box.gro -box 5 5 5
> >
> > and then I used genbox to solvate it:
> >
> > $ genbox -cp ethanediol_box.gro -cs spc216.gro -shell 1 -o
> ethanediol_solv.gro -p ethanediol.top
> >
> > I ran energy minimisation on it:
> >
> > $ grompp -f minim.mdp -c ethanediol_solv.gro -p ethanediol.top -o em.tpr
> >
> > my minim.mdp file is:
> >
> > ==========================
> > define        = -DFLEXIBLE
> > integrator    = steep
> > emtol        = 1000.0
> > emstep          = 0.01
> > nsteps        = 5000
> >
> > nstlist        = 1
> > ns_type        = grid
> > rlist        = 1.0
> > coulombtype    = PME
> > rcoulomb    = 1.0
> > rvdw        = 1.3
> > pbc        = xyz
> >
> > pme_order    = 4
> > constraints     = none
> >
> > nstxout        = 1
> > nstvout        = 1
> > nstenergy    = 1
> > nstlog        = 1
> > ==========================
> >
> > $ mdrun -v -deffnm em
> >
> > and the energies converge (although I think that the values are still too
> large for this system):
> >
> > ==========================
> > Steepest Descents converged to Fmax < 1000 in 12 steps
> > Potential Energy  = -1.1206108e+04
> > Maximum force     =  9.4016180e+02 on atom 31
> > Norm of force     =  3.4989252e+02
> > ==========================
>
> {remainder snipped}
>
> The energies do seem large.  What happens if you do energy minimization
> with emtol = 10.0  ?
>
>
> --
> Bruce D. Ray, Ph.D.
> Associate Scientist, and Operations Director
> NMR Center
> IUPUI
> Physics Dept.
> 402 N. Blackford St.
> Indianapolis, IN 46202-3273
>
>
> *From:* Nancy <nancy5villa at gmail.com>
> *To:* Discussion list for GROMACS users <gmx-users at gromacs.org>
> *Sent:* Wednesday, August 5, 2009 10:28:47 PM
> *Subject:* [gmx-users] Energy Minimisation and Equilibration Problems
>
>
>
> _______________________________________________
> gmx-users mailing list    gmx-users at gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at http://www.gromacs.org/search before posting!
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-request at gromacs.org.
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20090806/b7e0cee9/attachment.html>