[gmx-users] water models

Justin Lemkul jalemkul at vt.edu
Thu Mar 3 23:07:02 CET 2016

On 3/3/16 2:21 PM, Irem Altan wrote:
> Hi,
> I have two general questions about TIP5P and GROMACS version 5.1.2. (I am
> using AMBER99SB).
> - Is it normal that TIP5P does not work out of the box? When I do pdb2gmx
> without modifying force field files, I get an error saying Residue 'HO5' not
> found in residue topology database . This is really confusing, as there
> exists a tip5p.gro, tip5p.itp, the LJ parameters exists within
> ffnonbonded.itp, the atoms are defined in atomtypes.atp, but there is nothing
> within aminoacids.rtp and aminoacids.hdb. Is it meant to be that way?

Normally the water topology is simply taken care of by #including the 
appropriate stock topology.  pdb2gmx will try to create bonds between the atoms, 
which is not normal for rigid water models.

> - This is a more general water model question: what is the purpose of the
> #ifndef FLEXIBLE section within TIPXP.itp? (or spc, spce). I understand that
> the bonds and angles sections define the parameters for the desired bond
> lengths and angles, along with high force constants to constrain them. I also

Let's get the terminology straight - a constraint forces a bond length to be 
rigid and therefore has no force constant.

> understand that they are constrained with the SETTLES algorithm, but what
> does #ifndef FLEXIBLE do?

"Flexible" water means the bonds and angles obey normal harmonic vibrations.  So 
the #ifndef FLEXIBLE means "if the user hasn't told us to use a flexible 
representation, apply SETTLE to keep things rigid."  Keep rigid models rigid 
during MD.



Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
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 mailing list