[gmx-users] define new bond

[Justin Lemkul]

On 4/28/14, 10:15 AM, fantaman wrote:
> Dear GROMACS users,
>
> if I want to define a new bond, should I define it in the [ bonds ] section
> of the corresponding [ aminoacid ] section in top/aminoacid.rtp or should I
> leave this section (I mean [ bonds ]) empty and introduce the bond in the
> file forcefield.ff/ffbonded.itp? If I have to introduce the bond in the

It's not a matter of either/or.  The molecule level of the topology needs to 
know that a bond exists between two atoms.  The force field level needs to know 
what parameters are assigned to that bond whenever it occurs in the system.

> ffbonded.itp, I wouldn't know how to make it since the file I have inherited
> presents lists for gb's, ga's, gi's and gd's and then in the end sections
> for special bonds, but new bond parameters are never there...
>

The Gromos force fields are a bit different from all the others.  The gb_*, 
ga_*, etc declarations are just cpp macros that get parsed when grompp runs 
through ffbonded.itp - see that file for the way it works.  In reality, it isn't 
any different from any of the other force fields.  You can either declare your 
new bonded parameters via the #define mechanism or just write them explicitly 
per the format shown in Chapter 5 of the manual.

-Justin

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

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

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 601
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

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

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