[gmx-users] opla atom types and charges for nitrile / thienopyridine

[Justin A. Lemkul]

Sebastian Kruggel wrote:
> hello,
> 
> i generated a lig.itp file with topolgen.pl to perform ligand/protein md 
> simulations. as my ligand contains nitriles and a thienopyridine 
> scaffold, topolgen failed assigning atom types.
> 
> i found opls_753/4 in ffoplsaa.atp for the nitrile but changing the *itp 
> file i didn't find any hint for the charge nor the chgr entry to put. 

See below for parameterization advice.  See the manual for information on charge 
groups - the .atp file won't help you there.

> additionally i am not sure about the treatment of the pyridine nitrogen 
> as i only see the opls_520 and i don't know what the 6-31G* is meaning.
> 

6-31G* is a QM charge calculation method often used in deriving parameters; see 
below.

> in the mailing list i found
> http://lists.gromacs.org/pipermail/gmx-users/2009-May/042069.html

The more applicable message in the thread above is actually the next message:

http://lists.gromacs.org/pipermail/gmx-users/2009-May/042068.html

Virtual sites can be tricky with the nitrile group.

> but wondered if oplsaa shouldn't be the better choice for a md 
> simulation of ligand and protein - or is the gromos53a6 (or the gmx 
> force field as used in the drug/enzyme tutorial
> http://davapc1.bioch.dundee.ac.uk/prodrg/gmx.pdf
> ) not too bad?

Either way, parameterization is going to be hard.  Please consult the following, 
and consider that deriving parameters for any new species is a very advanced 
topic and should only be undertaken with a thorough knowledge of derivation 
procedures and the intrinsics of the desired force field:

http://www.gromacs.org/Documentation/How-tos/Parameterization

You could certainly derive parameters suitable for either force field (OPLS or 
53a6), but it will take a considerable amount of time.  The topolgen.pl script 
is just a cute little utility I wrote some time ago to give you a starting point 
for topology generation.  Its output should not necessarily be considered 
accurate or appropriate for immediate simulation, anyway.  You've got to do the 
legwork of deriving parameters from there!

-Justin

> 
> maybe anybody has done simulations with similar molecules, thanks in 
> advance for any help!
> 
> sebastian
> 
> 
> 

-- 
========================================

Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================