[gmx-users] Peptide-membrane simulation force field parameters

[Justin Lemkul]

On 2/4/15 2:28 AM, tarak karmakar wrote:
> Dear All,
> I need a suggestion related to simulation of peptide-membrane system.
> Are CGenFF parameters (from ParamChem) for peptide analogues good enough to
> couple with Charmm36 parameters for the lipids in a membrane? (provided
> very low charge and param penalties)
>

The protein force field is highly optimized, and we generally recommend against 
trying to shoehorn CGenFF parameters into it.  Low penalties just indicate that 
we cover the chemical space reasonably well in the force field, but that's not 
necessarily an indicator that those parameters will properly represent a protein 
moiety.  What you have to remember about generalized force fields (CGenFF, GAFF, 
etc) is that transferability comes at the expense of accuracy.

What kind of unnatural residue(s) do you have?  Can it be parametrized by 
analogy from existing functional groups without going to CGenFF?  I would 
recommend relying on the protein force field as much as possible.  If you have 
to derive anything via CGenFF, absolutely run the gamut of normal parameter 
validation before relying on the parameters for production simulations.

-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

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