[gmx-users] Fwd: Change in dihedral angle

[Justin Lemkul]

On 10/31/16 8:31 AM, Sudip Das wrote:
> Dear All,
>
> I am performing simulation of a protein containing cyclic peptide ring (not
> an regular amino acid) using GROMOS54a7 force field with GROMACS 5.0.5
> package. After equilibration and NVT production run, the final structure
> having a C-CHn-CHn-C dihedral angle has a value of -18.5 degree, whereas
> after optimizing the final structure quantum chemically in Gaussian09
> software, the same dihedral angle gives the value equal to +18.1 degree.
>
> Can anyone please suggest any solution? Should I need to change the
> dihedral angle parameter in the force field? Or this is just a problem of
> sign convention between the above to computing softwares?
>

That can easily be checked.  If the order of the atoms is the same, then the 
dihedral should be the same.

Note that the final structure of any run may not be representative of the entire 
run.  You should compute a distribution of the dihedrals and obtain an ensemble 
average.  Note, too, that sampling in solution may be different from anything 
that happens in vacuo.  The connection to QM with GROMOS is a bit tenuous.  A 
1-D potential energy surface of the dihedral in question may be useful to see if 
perhaps there are multiple minima with small barriers between them, for instance.

-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

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