[gmx-users] nrexcl = 2 or 3?

[Mark Abraham]

On 15/12/11, Yun Shi <yunshi09 at gmail.com> wrote:

> Hi Mark,
> 
> I do not quite understand. For example, in amber ff, 1-4 interactions (except the part from dihedral interactions) are calculated according to non-bonded parameters and then scaled by 1/2 or 5/6.

 
Yes, see discussion of gen-pairs in manual section 5.7.1 
 

>  When setting nrexcl = 3, which is the default, aren't 1 - 4 interactions excluded from using non-bonded parameters?

 
This is a separate mechanism. Exclusions for non-bonded interactions are applied "on the fly" in mdrun. That doesn't apply to 1-4 parameter generation, but it is necessary for the set of choices of nrexcl, gen-pairs, fudgeXX, [atomtypes], [nonbond_params], [pairtypes] and [pairs] to be consistent. Details vary with the particular force field. Parameters for "bonded" 1-4 interactions (a.k.a. pairs) are generated in grompp according to the rules of the force field - either from the [pairtypes], [pairs] or from scaling depending on the force field and interaction type.
 
Mark 

> 
> 
> 
> Thanks,
> Yun
> 
> 
> 
> On Wed, Dec 14, 2011 at 4:33 PM, Mark Abraham <mark.abraham at anu.edu.au> wrote:
> 
> 
> > 
> > 
> >  
> >  
> > On 15/12/11, lq z <lqz.fsu at gmail.com> wrote: 
> > 
> > > Dear GMXers,
> > > 
> > > I need to use charmm ff (same question for amber ffs), and am confused with what value I should give to nrexcl in [ moleculetype ], 2 or 3?
> > 
> >  
> > 
> > Whatever value pdb2gmx generates for CHARMM, probably 3. 
> > 
> >  
> > 
> > > 
> > > 
> > > According to the manual (v 4.5), "nrexcl = 3 stands for excluding non-bonded interactions between atoms that are no further than 3 bonds away." In charmm ff, 1-4 interactions are needed and it is "no further than 3 bonds away," 
> > 
> >  
> > 
> > ... but 1-4 interactions are regarded as bonded interactions. 
> > 
> >  
> > 
> > > so I should use nrexcl = 2. However, it is 3 in charmm27.ff/ folder.
> > 
> >  
> > 
> > Yes, CHARMM has special parameters for some 1-4 interactions, which are dealt with seperately.
> > 
> >  
> > 
> > > 
> > > 
> > > 
> > > By the way, I did a simple test with a dihedral a-b-c-d. Only a and d have charges. If I use nrexcl =3, no electrostatics energy is reported while the value matches with the one calculated by hand when I set nrexcl to be 2.
> > > 
> > 
> >  
> > 
> > As expected.
> >  
> > 
> > Mark 
> > 
> > --
> > 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/Support/Mailing_Lists/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/Support/Mailing_Lists
> > 
> 
> 
> 
> 

 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20111215/33299c36/attachment.html>