[gmx-users] Water model properties
jalemkul at vt.edu
Mon May 26 16:23:47 CEST 2014
On 5/25/14, 11:00 PM, Harshkumar Singh wrote:
> I'm using version 4.5.4 and very new to this software.
> Trying to follow this paper-
> * A systematic study of water models for molecular simulation: Derivationof
> water models optimized for use with a reaction field David van der Spoel,a)
> Paul J. van Maaren, and Herman J. C. Berendsenb)*
> 1.It says to use the SETTLE algorithm for constraining bonds but I can only
> get the options of using Lincs or Shake otherwise it shows an error.
The SETTLE algorithm is always used for water, unless you define waters as
flexible in the .mdp file (define = -DFLEXIBLE); LINCS and SHAKE only apply to
> 2.I went ahead using the Lincs algorithm but many of the properties do not
> give the same results.How do I resolve that?
> My steps were
> 1.converting the pdb file and then using editconf for making the box.
> 2.used grompp as the same it was in the tutotials with the changes made
> accordingly in the .mdp file according to the paper.
> 3.then the analysis with g_dipoles and g_energy.
> 4.only dipole moment was correct and rest other properties like
> epsilon,kirkwood factors,potential energy,density,temperature,pressure
> showed too much deviation.
Without seeing the actual contents of the .mdp files, knowing how long your
simulations were, how converged they were, and what the exact numbers you got
were, it's not really possible to comment. Note that if properties like
temperature and pressure are not converged, you're sampling a non-equilibrium
state and your results likely reflect this. The dipole moment works out because
the waters are rigid, so the value of the dipole moment is a fixed, intrinsic
Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow
Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 601
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201
jalemkul at outerbanks.umaryland.edu | (410) 706-7441
More information about the gromacs.org_gmx-users