Tetramers are not stable

[Kyle Kihn]

Hi all,
I am attempting to run an MD simulation of a tetramer system. We noticed that the tetramer immediate blows apart into either two dimers (in one system tested) or into four monomers. We tried this on four different systems and never saw the tetramer stay intact. We do not seem to have this problem though when generating the inputs through the CHARMM-GUI. Attached is the script used to run these simulations (generated from the openMM Gui). As a secondary question we have run other monomer systems using this script, if there is an issue is it effecting the results of our monomer system or is this issue unique to the oligomer states?
Thank you for the help,
Kyle

[Peter Eastman]

Did you modify that script by hand? There are some odd things in it. These lines especially don't do what they clearly were meant to:

Code: Select all

pdb.addSolvent(forcefield, padding=1.0*unit.nanometers, model='tip3p')
pdb.addSolvent(forcefield, ionicStrength=0.15*unit.molar, positiveIon='K+')
pdb.addSolvent(forcefield, ionicStrength=0.15*unit.molar, negativeIon='Cl-')

The first one will build a water box with no extra ions beyond the ones needed to neutralize it. The next two lines will do nothing. The box is already completely full of water, so there's no space for it to add any more. I think what you really meant to do was this:

Code: Select all

pdb.addSolvent(forcefield, padding=1.0*unit.nanometers, model='tip3p', ionicStrength=0.15*unit.molar, positiveIon='K+', negativeIon='Cl-')

By the way, the old openmm-builder web app is quite out of date. I recommend using OpenMM-Setup instead, since it does much more.

Anyway, it's hard for me to speculate on why you get different results with the openmm-builder script than the CHARMM-GUI one. Clearly there are differences between them, and probably a whole lot of differences. But I don't know which ones are responsible for the difference in behavior.