Dear Vald,
You can try to test these parameters. We don’t have much time to do extensive tests of these parameters now and if others are interested in doing it they are welcomed.
I think for the Na+, K+, Rb+ and Cs+ ions, the 12-6 model is pretty good. But for some other monovalent cations such as Li+, Tl+ and Ag+, and the halogen ions, the 12-6 model has considerable errors, where the 12-6-4 model is suggested to be used.
My opinion is Joung and Tom’s parameters from and our HFE parameters are all designed specifically for the PME simulation, and may could better reproduce HFE values than Smith&Dang’s parameters in the simulation using PME approach.
York and co-workers have benchmarked different nonbonded models for simulating Mg2+ ion and showed the 12-6-4 model showed improvement over the 12-6 ones. Especially for the 12-6-4 model with SPC/E water, it could reproduce several different kinds of properties at the same time. (Panteva, M. T.; Giambaşu, G. M.; York, D. M. J. Comput. Chem. 2015, 36, 970.)
After that, they have also performed refinement of 12-6-4 parameters for nucleic acid systems. (Panteva, M. T.; Giambaşu, G. M.; York, D. M. J. Phys. Chem. B 2015, 119, 15460.)
As Jason said, up to now there are only AMBER and OpenMM support the 12-6-4 model.
Kind regards,
Pengfei
> On Feb 12, 2016, at 5:38 PM, Jason Swails <jason.swails.gmail.com> wrote:
>
> On Fri, Feb 12, 2016 at 12:51 PM, Vlad Cojocaru <
> vlad.cojocaru.mpi-muenster.mpg.de> wrote:
>
>>
>> Did anyone use the 12-6-4 set in NAMD ? Is NAMD able to read the
>> additional lines necessary in the AMBER topology files ?
>>
>
> I'm pretty sure if you want to use the 12-6-4 potential, you have two
> options: Amber and OpenMM. And if you want to run on GPUs, your only
> option is OpenMM. ParmEd has an "OpenMM" action that will behave like
> sander and pmemd (it will read sander input files and run an equivalent
> simulation with OpenMM). If you have a 12-6-4 potential set there, then it
> should use that correctly. This is described in the Amber manual.
>
> No other code out there supports this
>
> potential form (at least that I'm aware of).
>
> HTH,
> Jason
>
> --
> Jason M. Swails
> BioMaPS,
> Rutgers University
> Postdoctoral Researcher
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sat Feb 13 2016 - 13:30:03 PST
