Hi: Re: [gmx-users] Hello: Forcefield and potentials for DNA-zinc finger complex

[Tsjerk Wassenaar]

Hi Nehme,

I've did simulations on TRAIL, which also contains a zinc-finger
domain, involving three cysteines and a chloride ion. But it's not so
simple. Luckily the exact parameters seemed not that vital, as the
zinc apparently functions to keep the three subunits together and was
in any case not directly related to the properties we were interested
in. Still, I spent some time parameterizing it. For that I did quite a
search of the available literature. You'll find some references in our
2008 Proteins paper. One of the most striking things is the still
unresolved issue of the protonation state of zinc finger domains.
Especially in the acidic environment around DNA, this may become an
issue. It is why I have stayed away from zinc-finger domains lately,
as they'd cost me too much work properly parameterizing them, even
though our lab is doing experimental work on them and it would be
interesting to complement that with MD. That's also why I mentioned to
be interested in such parameters. Not in what you can readily
calculate with QM or what you can get from automatic topology builders
or even from literature, including my own. That doesn't solve the
issue of the protonation state or overall charge of the complex. I
won't trust any report on zinc-finger domain MD, unless it is shown to
properly reproduce experimental properties, including what can be
determined in relation to the protonation state.

Cheers,

Tsjerk


On Wed, May 20, 2009 at 10:40 AM, Nehme El Hachem <ne21 at aub.edu.lb> wrote:
> Hi Tsjerk,
>
> I thank you for your help. Indeed, I am very new to GROMACS ( I have a Linux
> Red hat 9.0 and I do have some issues installing the software).
>
> If you are interested, please find a link:
> http://www.epress.com/w3jbio/vol2/harris2/harris.htm
> Molecular Dynamics Simulation Of The Estrogen Receptor Protein In Complex
> With A Non-Consensus Estrogen Response Element DNA Sequence In A 10 Angstrom
> Water Layer.
>
> I am trying to do some molecular dynamics on the oestrogen receptor DNA
> binding domain (PDB code: 1HCQ). I have InsightII with Discover3 and CHARMm
> but I am confused with the type of forcefield that I need. Furthermore, I
> have to add counterions to neutralize the system, so it seems complicated.
> I wanted to try GROMACS, because it has many forcfields (ex: ffAmber) for
> modeling this type of system.
>
> I appreciate your help,
>
> Nehme.
>
>



-- 
Tsjerk A. Wassenaar, Ph.D.
Junior UD (post-doc)
Biomolecular NMR, Bijvoet Center
Utrecht University
Padualaan 8
3584 CH Utrecht
The Netherlands
P: +31-30-2539931
F: +31-30-2537623