[CPMD-list] how to include water in QM part in a QMMM calculation

[Axel Kohlmeyer]

On Wed, 16 May 2007, Yong Zhang wrote:

dear yong,


YZ> What is the difference between 999 and -999? When would those be used?

please see
http://www.cpmd.org/manual/node47.html

and for more details have a look at detdof.F lines 230-250

there are some oddities in your input, can you please explain
how you did the setup your system. did you equilibrate it with
gromos or amber? you gromos input looks as if you used amber,
but then you have to set AMBER in the &QMMM section.

YZ> &QMMM
YZ>   VERBOSE
YZ>   GROMOS
YZ>   WRITE LOCALTEMP
YZ>   SPLIT
YZ>   TOPOLOGY

YZ> # NPM >= 0 number of (identical) solute molecules
YZ> # NSM >= 0 number of (identical) solvent molecules
YZ> #      NPM     NSM
YZ>        1      429
YZ> END

also, if you want to make water molecules quantum, that are
in the solvent section, you first have to convert them into
solute molecules. for that you have to create a new topology
and input file (either manually, or via the FLEXIBLE WATER
keyword). see:
http://www.cpmd.org/manual/node111.html


YZ> > > The water is part of solute (residue H2O). It is weird because no shake
YZ> > > or constraint is used in either cpmd input or gromos input. Anybody has

this is not correct. _every_ solvent molecule in gromos is rigid!
if you want to make it flexible, you have to convert it into a 
solute molecule. please note that the usual water potential parameters
in your input are not suitable for making the potential flexible.
so it is advisable to have only those waters flexible, that are
converted into quantum molecules. otherwise you'd have to re-equilibrate
your whole system with adapted water potential parameters.

if you want to see the difference, please do the following experiment:

set up and equilibrate a small protein in water with amber and then
create two sets of gromos inputs: one via amber7togromos.x -rigid
and one via amber7togromos.x -flexible
set those up for a qm/mm run with cpmd in the same way, and then 
run them both with MOLECULAR DYNAMICS CLASSICAL (this is not 
parallelized, but it runs pretty fast on a current desktop anyways).

this should seamlessly continue your classical MD. please follow 
the enerergies and see the difference.

YZ> > > met the same problem? Is this related to the SPC water model? How to
YZ> > > overcome this problem? Thanks a lot.

if you run with amber, you actually use TIP3P unless you 
changed the water potential.

cheers,
   axel.

YZ> > >
YZ> > > best wishes,
YZ> > >
YZ> > > Yong
YZ> > >
YZ> > >
YZ> > > ------------------------------------------------------------------------
YZ> > >
YZ> > > _______________________________________________
YZ> > > CPMD-list mailing list
YZ> > > CPMD-list at cpmd.org
YZ> > > http://cpmd.org/mailman/listinfo/cpmd-list
YZ> >
YZ> > --
YZ> > Fleurat-Lessard Paul, Lecturer  e-mail: Paul.Fleurat-Lessard at ens-lyon.fr
YZ> > Laboratoire de Chimie
YZ> > Ecole Normale Supérieure de Lyon              Tel: + 33 (0)4 72 72 81 54
YZ> > 46, Allée d'Italie                            Fax: + 33 (0)4 72 72 88 60
YZ> > 69364 Lyon Cedex 07
YZ> >
YZ> 

-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
   Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.