VMD-L Mailing List

From: Trollmann, Marius (marius.trollmann_at_fau.de)
Date: Thu Nov 10 2022 - 10:36:32 CST

Dear VMD-Users,

I recently started to work with the ffTK plugin to parametrize the small
molecule Thiazolidin (there are some similar molecules in CGenFF, but I
want to use this workflow for more complex molecules later on). However,
I have some trouble with the optimization of the partial atomic charges.
Recent literature (e.g. Additive CHARMM36 Force Field for Nonstandard
Amino Acids, CHARMM General Force Field: A Force Field for Drug-Like
Molecules Compatible with the CHARMM All-Atom Additive Biological Force
Fields) suggests for sulfur-containing compounds the usage of MP2/6-31G*
together with the Basis set superposition error (BSSE). I have the
following questions:

1) How can I load a Gaussian .log file into ffTK generated with the
Counterpoise-Method (to take the BSSE into account) and Gaussian16revA3?
If I want to do this with the actual version of ffTK (2.1 and
vmd_1.9.4a57), I get the following error message "vecscale: parameters
must have data" originating from the QMTools routine
"get_cartesian_from_zmat".

2) I ran already the QM water interaction calculations using HF/6-31G*,
MP2/6-31G* and MP2/6-31G* with BSSE to compare the results (I applied
the calculations on all heteroatoms (S,N) and hydrogens in the
molecule). In case of HF/6-31G* and MP2/6-31G* some of the .log files
contained the warning "Optimization completed on the basis of negligible
forces" and showed distances above 2.5A between the hydrogen/oxygen in
the water molecules and the donor/acceptor atoms. Can I ignore this
files for the subsequent MM-optimization or could I edit some the input
arguments to get a better convergence of the optimization routine? What
really puzzles me is that I get different results (more atoms with a
minimum distance above 2.5A) if I include the counterpoise method into
the calculation (MP2/6-31G* with BSSE). Is that somehow expected or is
this suspicious?

3) If I use the MP2 level of theory for the calculation of the QM
interaction energies, I need also to adapt the level of theory in the
water-sp.gau file, right? By default it uses RHF, can I safely replace
this with MP2?

I want to thank you already in advance for your answers!

Cheers,
Marius