On Mon, 12 May 2003, David A. Case wrote:
> On Mon, May 12, 2003, Ioana Cozmuta wrote:
>
> > > > Warning: Encounter 10-12 H-bond term
>
> Note that there are not really any 10-12 bond terms, in all probability.
> Amber uses this notation for atom pairs with no nonbonded interactions
> (of any sort, 10-12 or 6-12).
In amber I am using the parm94 ff and I understand that the 10-12 Hbond
term was embedded into the van der Waals interaction between the oxygen
atoms. As hydrogens bonded to oxygen atoms have zero radii, basically the
interaction between an oxygen and a hydrogen atom not bonded to it is like
the oxygen sees the other oxygen (the radii of the oxygen atoms was
extended to consider the hydrogen). And basically the same idea is
followed in all cases where h-bonds could be encountered.
So the parametrization in water is:
-nonbonded for OW-OW (6-12 LJ potential)
-bonded interactions for OW-HW interactions (diag harmonic)
-the 1-4 interactions are reduced with the corresponding constants
What I do not understand is if I turn on amber in NAMD and I use the same topology and
coordinate files as in amber in NAMD why is NAMD complaining about the
10-12 H-bond term when amber is not?
So you mean that NAMD still sees the hydrogens available for non-bonded
interactions?
> >
> > Ewald PARAMETER RANGE CHECKING:
> > parameter skinnb+cutoffnb: (nonbond list cut) has value 0.10500E+02
> > This is outside the legal range
> > Lower limit: 0.00000E+00
> > Upper limit: 0.89703E+01
> > Check ew_legal.h
>
> You might be able to get away with reducing your skinnb to a small value.
> Otherwise, you need to make a bigger system...Amber is not set up to handle
> very small unit cells (of the order of the cutoff).
>
It was just a test system that I could have indeed make it larger :)
> ..good luck..dac
Thank you,
Ioana
>
Received on Tue May 13 2003 - 00:53:01 PDT
