[gmx-users] Topology file of molecule
jalemkul at vt.edu
Mon Dec 4 15:24:14 CET 2017
On 12/4/17 5:31 AM, Krzysztof Kolman wrote:
> Dear Justin,
> Since you last answer, I have been trying to find some description how to
> perform dihedrals scans using some QM software and Gromacs. I have found
> out that it used to be some QM software implemented in Gromacs (Orca) but
> this implementation is not developed anymore and it does not work with
> newer version of Gromacs and Orca, please correct me here if I am wrong. I
The QM/MM interface is not really maintained. This is also not what you
really should be using. The QM dihedral scans you need to apply have no
MM component, so you don't need GROMACS at all. In general:
1. Perform a QM optimization at a suitable level of theory
2. Create a series of input files with systematically varying values of
the target dihedral, and perform an optimization of the structure with
that dihedral frozen.
3. Compute the potential energy surface of those optimized configurations.
> have been also trying to find out how to do the MM part using Gromacs and I
> found out that the simulations should have some unusual parameters
> (cutoffs, etc.) and it is quite tedious work since all the input files need
> to be manually prepared. My question is do you know maybe if there is
> somewhere a tutorial how to perform such scans using Gromacs where all this
> knowledge is combined and also some semi-automatic solution is presented?
No, and this is something that is indeed very laborious in GROMACS.
You'd have to perform a series of optimizations (energy minimizations)
with dihedral restraints, which have different values (could be easily
scripted there), and then a single-point energy value computed in the
absence of restraints. This is trivial to do in a program like CHARMM,
it is more tedious in GROMACS.
Justin A. Lemkul, Ph.D.
Virginia Tech Department of Biochemistry
303 Engel Hall
340 West Campus Dr.
Blacksburg, VA 24061
jalemkul at vt.edu | (540) 231-3129
More information about the gromacs.org_gmx-users