Re: AW: restraining

From: Axel Kohlmeyer (akohlmey_at_gmail.com)
Date: Fri Aug 24 2012 - 08:25:22 CDT

chris,

On Fri, Aug 24, 2012 at 3:08 PM, Christopher Rowley <cnrowley_at_mun.ca> wrote:

> In any case, it seems inevitable that I will take Axel's suggestion and
> modify Molecule.C to recognize these molecules and add a constraint between
> their two hydrogens. In general, I would suggest that NAMD should consider
> imposing shake on any H-H bond that is defined in the PSF when rigidbonds
> all is on. Many small molecule force fields impose rigidity this way and to
> be consistent with their parameterization, they should be strictly rigid.

you are ignoring some of the "design decisions" that are in NAMD
for better performance. on of those is the hardcoded water detection
for using an optimized constraints implementation. NAMD is not
directly intended to be a generic MD code. if you are looking
for a more generic code with good parallel performance, you have
to set your sights on gromacs or lammps.

once your h2s *is* recognized as water, the constraints
will be applied, you don't need to add them. i am not certain
however, if there are not any other pitfalls. just to reiterate
NAMD doesn't not expect people to use a rigid 3-atom solvent
other than water.

axel.

>
> Thanks,
> Chris
>
>
> On 08/24/2012 04:12 AM, Norman Geist wrote:
>
> Hi,
>
> also, as far as I know, the H-O-H in TIP3P is not rigid because of the bond,
> but because the angle is fixed by default. The SHAKE is, as I think, only
> applied for the O-H bonds. So you can't use SHAKE for that. But you could
> just try to introduce a fixed angle. Therefore you could modify the
> forcefield of your molecule itself by using a high force for the angle for
> example, or use the NAMD feature extrabonds which also allows to introduce
> extra angles to the system with a custom force.
>
> It would also be interesting which force field you are using.
>
> regards
>
> Norman Geist.
>
> -----Ursprüngliche Nachricht-----
> Von: owner-namd-l_at_ks.uiuc.edu [mailto:owner-namd-l_at_ks.uiuc.edu] Im
> Auftrag von Chris Harrison
> Gesendet: Freitag, 24. August 2012 05:30
> An: cnrowley_at_mun.ca
> Cc: namd-l_at_ks.uiuc.edu
> Betreff: Re: namd-l: restraining
>
> Two possibilities:
>
> 1) You should be able to effectively accomplish this by imposing an
> extremely high force constant in either the "restraints/constraints"
> or the "colvars" approach.
>
> 2) To utilize the SHAKE algorithm directly in such a manner, ......
> does your topology/PSF file include an explicit H-H bond for HS?
>
> Chris
>
>
> On Thu, Aug 23, 2012 at 10:20 PM, Christopher Rowley
> <cnrowley_at_gmail.com> wrote:
>
> Sorry, I should have been more clear about this: I want to keep the
>
> hydrogen
>
> sulfide molecules rigid according to the equations by imposing the
>
> SHAKE
>
> algorithm on the H-H distance, like in rigid water molecules.
>
> Thanks,
> Chris
>
>
> On 24/08/2012 12:38 AM, Chris Harrison wrote:
>
> 'rigidbonds all' only refers to hydrogen bonds. It is a parameter
>
> for
>
> the Rattle/Settle algorithm.
>
> You are looking for either "restraints/constraints":
>
> http://www.ks.uiuc.edu/Research/namd/2.9/ug/node27.html#5771
>
> or the "colvars module":
>
> http://www.ks.uiuc.edu/Research/namd/2.9/ug/node53.html
>
> Either option should allow you to set the restraints you describe.
>
> Chris
>
> On Thu, Aug 23, 2012 at 1:55 PM, Christopher Rowley
>
> <cnrowley_at_mun.ca>
>
> wrote:
>
> Hi,
>
> I'm using NAMD for simulations of a model for hydrogen sulphide
>
> (H2S).
>
> The
> model is defined as being rigid, like TIP3P, so I've included the
> command
> "rigidbonds all" Nevertheless, the H-S-H bond is flexible in MD
> simulations.
>
> The PSF file contains a bond between the two hydrogens, but this
>
> doesn't
>
> seem to make a difference.
>
> Unless I've misunderstood, H-H bonds should be identified in
>
> Molecule.C
>
> Molecule.C:
> if ( ! is_water(a2) ) { // H-H but not water
> rigidBondLengths[a1] = ( mode == RIGID_ALL ? x0 : 0.
>
> );
>
> rigidBondLengths[a2] = ( mode == RIGID_ALL ? x0 : 0.
>
> );
>
> }
>
> Is there a way to force NAMD to constrain this distance/angke?
>
> Thanks,
> Chris
>
> --
> Christopher Rowley
> Ph.D. (chemistry)
> Assistant Professor
> Department of Chemistry
> Memorial University of Newfoundland
> http://www.chem.mun.ca/homes/cnrhome/
>
>
> This electronic communication is governed by the terms and
>
> conditions at
>
> http://www.mun.ca/cc/policies/electronic_communications_disclaimer_2012
> .php
>
>
>
> --
> Chris Harrison, Ph.D.
> NIH Center for Macromolecular Modeling and Bioinformatics
> Theoretical and Computational Biophysics Group
> Beckman Institute for Advanced Science and Technology
> University of Illinois, 405 N. Mathews Ave., Urbana, IL 61801
>
> http://www.ks.uiuc.edu/Research/namd Voice: 773-570-6078
> http://www.ks.uiuc.edu/~char Fax: 217-244-6078
>
>
>
> --
> Christopher Rowley
> Ph.D. (chemistry)
> Assistant Professor
> Department of Chemistry
> Memorial University of Newfoundland
> http://www.chem.mun.ca/homes/cnrhome/
>
>
> This electronic communication is governed by the terms and conditions at
> http://www.mun.ca/cc/policies/electronic_communications_disclaimer_2012.php

-- 
Dr. Axel Kohlmeyer  akohlmey_at_gmail.com  http://goo.gl/1wk0
International Centre for Theoretical Physics, Trieste. Italy.

This archive was generated by hypermail 2.1.6 : Mon Dec 31 2012 - 23:21:57 CST