[gmx-users] Simulation of polarizable Carbon nanotubes

[Justin Lemkul]

On 6/25/15 6:21 PM, S.M. Golam Mortuza wrote:
> Hi Justin,
> Thanks a lot for your reply and suggestion. I have pulled the master branch and performed "git checkout drude". I think, the file named md.cpp in /src/program/mdrun is the main file to run the code. Please let me know if I am wrong.

There's a whole lot more than just md.cpp; that is the main mdrun code but 
there's a lot more under the hood.

> I am just wondering how can I execute the code in my current version of gromacs (4.6.1), which is installed in my computer and cluster. Do I need to re-install the the clone? I've tried to explore the information from Git Tutorial in Gromacs website, but could not understand much. Please accept my apology.

You need to actually compile and install the new code; it compiles like any 
other version of GROMACS.  4.6.1 won't do what you need.

> Regarding anisotropic polarization: I see that [water_polarization] directive was used in the topology file in the past to model anisotropic polarization of water molecule, while [polarization] was used for isotropic polarization. I am assuming by mentioning "Build the manual for the syntax of the new topology directives" in your previous email, you meant to create a new directive, for instance [cnt_polarization], in my topology file. I am not sure how to do that. I do not see any information regarding polarization directives in the gromacs manual as well. I would highly appreciate if you please provide some information how can I proceed with this.

You need to compile the development version of the manual to see the new syntax 
of the [anisotropic_polarization] directive.  You should not create any new 
topology directives.

I also took another quick look at the paper; I think the way they are specifying 
the polarization is different from what GROMACS requires.  Their polarization 
value is massive, orders of magnitude larger than what we typically use for 
atomic polarizability.  They appear to have specified the polarization of the 
entire unit cell, so likely this value has to be divided over the number of 
shells in the system, but that's just a guess.  They say they used a modified 
version of GROMACS, so you should contact them to get their topologies and/or 
code to make sure it's consistent.

-Justin

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

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

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