From: Jason Smith (
Date: Mon Mar 08 2021 - 11:04:23 CST

Gaussian has forcefield methods that might be a better starting point, though I don't know if they have something ubiquitous like mmff94 or similar to run with whatever your ligand is. You could use VMD to manually move the waters to more reasonable positions and then run. My personal suggestion would be to move the waters to somewhere reasonable in VMD (can do it with the mouse, there are tutorials around somewhere), and then run something fast to get a good starting point (pm7, sto-3g, many free programs around like mopac, discovery studio), then use those coords for the fftk steps.

Note that I believe fftk optimisation relies on the user having sampled the water binding spots sufficiently well, so someone with more experience on the fftk side of things might have a safer idea.


Jason Smith
Post-Doctoral Fellow
Robert N. Young Lab, Simon Fraser University
8888 University Drive, Burnaby, BC, V5A-1S6

From: <> on behalf of Prof. Eddie <>
Sent: Monday, 8 March 2021 6:46:07 AM
To: Vmd l
Subject: vmd-l: fftk wat. opt. when charges overlap

Hi all,
I've gone through the fftk tutorials and they tell you to check the placement of the waters before running the gaussian files. Good advice. I found that a few of my waters would overlap with some of the ligand atoms. However, now I am not sure what to do? Do I just not optimize those atoms or is there a way to "fix" this? I'm unfamiliar with gaussian and don't have gaussview.
Thanks for the help!!

Edward Ackad, Ph.D<*7Eeackad__;JQ!!DZ3fjg!srQPZ1jc311MKadXRrCkW9cOwjF2nJe5oOtA73jdIscAX1onhY341wS-s5OBmicW7Q$>
Associate Professor of Physics
Computational Nanophotonics
Southern Illinois University Edwardsville
(618) 650-2390