[gmx-users] simulating AMP covalently linked to a protein

[Justin Lemkul]

On 5/25/17 3:12 PM, Gilberto Valdes wrote:
> Thanks for your answer,
> I would like to use charmm27 force field, it covers the hole AMP molecule
> if I use the ADE residue with the 5PHO and 3TER patches implemented in
> charmm software.
> The problem is how to patch the ADE equivalent residue (named RA) in
> gromacs, and then how to linked to the epsilon N group of the Lys.

Use our CHARMM36 port:

http://mackerell.umaryland.edu/charmm_ff.shtml#gromacs

It has ATP, from which you can easily make an AMP unit linked to lysine.  You'll 
likely have to parametrize the linkage.  This will involve a geometry 
optimization, charge assignment (most can be taken by analogy, just a few atoms 
around the linkage should change), water interactions, a bonded refinement 
including potential energy scans for dihedrals.  You can probably use CGenFF as 
a starting point but you should ultimately not mix general atom types with 
protein and nucleic acid types.

-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

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