On Mon, 2014-04-28 at 12:17 +0300, Thomas Evangelidis wrote:
> Hi,
>
> Is it because I do not use the latest version of AMBER but AMBER 12
> instead?
No. Nothing has changed about the dihedral potential in a very long
time (decades at least).
> I get the following error when I try to run it with with
> pmemd.cuda Version 12.3:
>
> | New format PARM file being parsed.
> | Version = 1.000 Date = 04/23/14 Time = 14:17:23
>
> | Note: 1-4 EEL scale factors are being read from the topology file.
>
> | Note: 1-4 VDW scale factors are being read from the topology file.
> CUDA (GPU): Prmtop appears to contain modified off-diagonal elements for
> VDW A coef.
> GPU code does NOT support such force field modifications.
> Found Aij = 787979.7920942737 expected
> 479622.7030000000
> i = 20 j = 1 Matrix element = 191
This error occurs because you are using Amber 12 instead of Amber 14.
The iPolQ force field employs a NBFIX approach to Lennard-Jones
interactions. NBFIX is terminology from CHARMM in which different pairs
of atom classes interact differently with each other than they do with
others. There's a discussion of this topic on my wiki:
http://jswails.wikidot.com/parmed#toc7
Because pmemd.cuda used to (in versions 12 and earlier) back-out the
original sigma and epsilon parameters for each atom and recombine them
when computing the Lennard-Jones interactions. This approach does not
work when certain pair interactions are changed, hence this warning.
HTH,
Jason
--
Jason M. Swails
BioMaPS,
Rutgers University
Postdoctoral Researcher
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Apr 28 2014 - 05:30:04 PDT
