From: Mayne, Christopher G (
Date: Wed Sep 28 2016 - 09:35:46 CDT


It looks like there is only one or two conformations in the MM PES that are causing the shift. Most of the time when this occurs there it is due to a strange conformation upon MM-based minimization, a bad LJ parameters, or something else that gives such a large offset in the reported energy using MM force fields. In the latest alpha version of VMD, I have added a new feature under the “Advanced Settings” section of the “Opt. Torsions” tab that will export a trajectory containing the MM-minimized conformations that are used to compute the PES. This option is activated by checking the “Save MM Traj.” option.

I would suggest re-running your optimization with this option selected, and then take a look at the trajectory to see if you can pinpoint any structural issues with the MM structures. You should also be able to use the “Torsion Explorer” (under the “Scan Torsions” tab) to load in the QM conformations and compare the two.

Christopher Mayne

> On Sep 28, 2016, at 5:19 AM, Mohan maruthi sena <> wrote:
> Dear FFTK experts,
> I am trying to generate parameters for an
> organic molecule containing 41 atoms. I have followed the FFTK
> tutorial and generated charges, bond and angle parameters. I am trying
> to fit dihedrals and I have not considered the dihedrals ending with
> hydrogen atoms (except polar hydrogens). I have fitted the dihedrals using
> simulated annealing method, the fit seems reasonably good except there
> is difference in minima between quantum and fitted curve. I have
> tried many ways changing various k and delta values but failed to get
> the minima coinciding with QM part. Could you please provide some
> insights regarding the shift in minima. Please have a look at the
> plots (links provided below) containing normal and enlarged views of
> minima's.
> --
> Thanks and Regards,
> Mohan Maruthi,
> Ph.D. student,