VMD-L Mailing List
From: Francesco Pietra (chiendarret_at_gmail.com)
Date: Sat Dec 13 2014 - 16:44:12 CST
- Next message: zoran: "Re: FFTK most probably doesn't work with molecules containing new atom types"
- Previous message: zoran: "Re: FFTK most probably doesn't work with molecules containing new atom types"
- In reply to: Axel Kohlmeyer: "Re: FFTK most probably doesn't work with molecules containing new atom types"
- Next in thread: zoran: "Re: FFTK most probably doesn't work with molecules containing new atom types"
- Reply: zoran: "Re: FFTK most probably doesn't work with molecules containing new atom types"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
While I obviously agree with Axel about the basic science illustrated,
there are situations where such dirty parameterizations of transition metal
clusters for classical MD prove useful. That is the case when the focus is
not the metal complex, while conservation of the geometry of the metal
complex during MD is what matters. A case in point is PNAS ͉ August 14,
2007 ͉ vol. 104 ͉ no. 33 ͉ 13319 –13324 If you read the PhD thesis behind
that, you see that that work was carried out with those limitations in
mind. In my view it proved useful. Consequently, tools to develop params
for tm complexes may have their interest. Such as the MCPB toolkit in amber
quarters, if it goes beyond Zn. cheers francesco pietra
On Sat, Dec 13, 2014 at 10:15 PM, Axel Kohlmeyer <akohlmey_at_gmail.com> wrote:
>
> can't let this go uncommented...
>
> On Sat, Dec 13, 2014 at 7:50 AM, zoran <zmatovic_at_kg.ac.rs> wrote:
>
> [...]
>
> > I am very sorry that scientists making force fields for macromolecules
> > forget on transition metals (in general) and that TM make a substantial
> part
> > of biological processes.
>
> if you knew a bit more about force fields and the chemistry of
> transition metals, you wouldn't make such a general attack. people
> don't do parameters for transition metals for classical (pairwise
> additive) force fields, because it is pretty damn near impossible to
> get them right. already divalent cations are a big stretch in most
> cases and rarely represented well. force fields are always a
> compromise, but in the case of transition metals, you have to deal
> with the problem of having to describe interactions that have
> substantial directional and manybody character (i.e. the interaction
> of the transition metal depends on how many and what neighbors are
> where) with a spherical functional form. of course, you can follow the
> recipe for force fields and derive some parameters, but in most cases,
> the resulting simulations are a waste of time. considering the fact
> that many force field users are very uncritical about how to use
> parameterizations, it is quite understandable, that people, who
> specialize in force field parameters, do not want to lead people down
> that road and rather not generate and publish sub-standard parameters.
>
> if you know a way how to do these parameterizations so they are
> producing meaningful calculations, please don't hold back and let us
> know. we most certainly do not need people that blame everybody else
> but themselves.
>
> axel.
>
>
> > With my best regards to all of you
> > Zoran
> >
>
>
>
> --
> Dr. Axel Kohlmeyer akohlmey_at_gmail.com http://goo.gl/1wk0
> College of Science & Technology, Temple University, Philadelphia PA, USA
> International Centre for Theoretical Physics, Trieste. Italy.
>
>
- Next message: zoran: "Re: FFTK most probably doesn't work with molecules containing new atom types"
- Previous message: zoran: "Re: FFTK most probably doesn't work with molecules containing new atom types"
- In reply to: Axel Kohlmeyer: "Re: FFTK most probably doesn't work with molecules containing new atom types"
- Next in thread: zoran: "Re: FFTK most probably doesn't work with molecules containing new atom types"
- Reply: zoran: "Re: FFTK most probably doesn't work with molecules containing new atom types"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]