[gmx-users] limit to the number of QM atoms in a QM/MM simulation?

[Benson Muite]

Hi Kristina,

Cannot answer your question directly, but in addition to Orca, there is 
some effort in this area that may allow using software other than Orca 
for the QM part, for example:
Zalevsky A.O., Reshetnikov R.V., Golovin A.V. (2019) New QM/MM 
Implementation of the MOPAC2012 in the GROMACS
https://doi.org/10.1007/978-3-030-05807-4_24

Benediktsson B. and Bjornsson R. (2017)
QM/MM Study of the Nitrogenase MoFe Protein Resting State: 
Broken-Symmetry States, Protonation States, and QM Region Convergence in 
the FeMoco Active Site
https://github.com/RagnarB83/chemshell-QMMM-protein-setup


Olsen, Bolnykh, Meloni, Ippoliti, Bircher, Carloni and Rothlisberger (2019)
MiMiC: A Novel Framework for Multiscale Modeling in Computational Chemistry
https://doi.org/10.1021/acs.jctc.9b00093
http://manual.gromacs.org/2019/reference-manual/special/mimic-qmmm.html 
(I belive this is still in progress


Regards,
Benson
On 9/3/19 10:42 AM, Kristina Woods wrote:
> Hello:
>
> I have a very basic question about the QM/MM implementation in gromacs (and
> yes - I know that it is no longer supported).  I am using ORCA as the
> quantum chemistry package for a QM/MM simulation of a photo-protein in
> gromacs.  I have successfully used the combination of ORCA and gromacs with
> other photo-protein QM/MM simulations without problem but in the previous
> cases I have had a smaller number of QM atoms (below 200 atoms).  In my
> current protein, I would like to consider close to 700 atoms but I notice
> that everytime I set-up a simulation in gromacs - gromacs fragments the
> geometry of the system so that only a limited number of the QM atoms are
> contained within the simulation box.  The rest of the atoms are translated
> out of the simulation box (sorry if this is not a good description).  I
> have tried compiling different combinations of gromacs and orca and this
> doesn't seem to change anything.  I have also studied the input geometry of
> my system and that seems to be fine (the initial geometry comes from the
> output of a 100 ns all-atom MM simulation of the entire system).  I then
> tried to simulate only of a subset of the QM atoms (up to 200 atoms) of
> interest and with a smaller system everything seems to work beautifully.
> So my question is if my problems are attributed to the fact that there is a
> limit to the number of QM atoms that can be considered in a QM/MM
> simulation?
>
>
> Thank you,
>
> Kristina