[gmx-users] SWM4-NDP Simulation

[Justin Lemkul]

On 8/23/18 4:38 AM, Julian Müller wrote:
> Hello,
>
> i'd like to simulate pure water and sodium chloride solutions using 
> the drude polarizable SWM4-NDP model. I am new to Gromacs and don't 
> know how to setup the calculation. I simulated SWM4-NDP boxes in 
> LAMMPS before, but LAMMPS lacks the hardwall constraint for the 
> charged particle on a spring.
>

SWM4-NDP should never have Drudes hitting the hard wall, so for pure 
water, you should never need that algorithm.

> I successfully did small test calculations with the TIP4P water model 
> and want to transfer to the polarizable model from there.
>
> In the mailing list I found some hints from 2014, that the preparation 
> of polarizable solvent boxes via pdb2gmx will be available in the near 
> future but no further information on that in the 2018 manual or the 
> help function of pdb2gmx. I also tried to build a SWM4-NDP.itp from 
> the SW.itp but failed miserably at the unfamiliar .itp syntax. Any 
> pointing in the right direction would be greatly appreciated.
>

GROMACS doesn't yet officially support the extended Lagrangian code, but 
it has supported SCF-based polarization for quite some time. The 
extended Lagrangian that you're likely referring to was coded into 
GROMACS and added to the development repo a few years ago, but 
unfortunately we're still struggling with some domain decomposition 
compatibility issues that I haven't had sufficient time to sort out. 
OpenMP works just fine, but that's obviously going to limit you to small 
systems.

My recommendation at this point is that you use NAMD or OpenMM until we 
can get official support for all the Drude stuff in GROMACS. If you want 
a GROMACS-compatible .itp file for SWM4-NDP, see 
http://mackerell.umaryland.edu/charmm_drude_ff.shtml

-Justin

-- 
==================================================

Justin A. Lemkul, Ph.D.
Assistant Professor
Virginia Tech Department of Biochemistry

303 Engel Hall
340 West Campus Dr.
Blacksburg, VA 24061

jalemkul at vt.edu | (540) 231-3129
http://www.thelemkullab.com

==================================================