Re: About QM-MM

From: Dhiraj Srivastava (dhirajks_at_gmail.com)
Date: Wed Aug 29 2007 - 12:05:26 CDT

atleast it will be good if we can do DFT level calculation. although very
high level QM calculation may not be possible but atleast we can do DFT
level calculation. it is more reliable and also not too costly
computationaly.
Dhiraj

On 8/29/07, Francesco Pietra <chiendarret_at_yahoo.com> wrote:
>
> Hi
> I also want to add to my previous reply, that semiempiricals within MD
> pose
> special (serious) problems that semiempirical alone may not have. So,
> perhaps
> the adaptation is not so straightforward. This is purely guess because I
> have
> little with standalone semiempiricals, actually I found them unreliable
> anytime
> I tried them (I am talking about conformational problems).
>
> francesco
>
> --- Dhiraj Srivastava <dhirajks_at_gmail.com> wrote:
>
> > Hi
> > from GAMESS-US you can do semiempirical, DFT and other higher level
> > calculation. so you can do any level of calculation you want with
> > GAMESS-US. I was thinking NWChem is not open source.
> > Dhiraj
> >
> >
> > On 8/29/07, Francesco Pietra <chiendarret_at_yahoo.com> wrote:
> > >
> > > I don't know Gamess-US (I use NWChem). Are you talking of
> semiempirical or
> > > "ab
> > > initio" QM? Of course, "ab initio" wouldn't be viable for molecules
> from
> > > medium
> > > size (60 atoms) on.
> > > francesco
> > >
> > > --- Dhiraj Srivastava <dhirajks_at_gmail.com> wrote:
> > >
> > > > Hi
> > > > If you can interface NAMD with some open source QM package like
> > > GAMESS-US
> > > > then it will be good. then both NAMD and QM package will be open
> source.
> > > > Dhiraj
> > > >
> > > > On 8/29/07, Francesco Pietra <chiendarret_at_yahoo.com> wrote:
> > > > >
> > > > > The problem is not open or closed source (you can have ONION in
> open
> > > > > source QM
> > > > > suites, as I have it). ONION is quite a different affairs from
> QM-MM.
> > > > >
> > > > > francesco
> > > > >
> > > > > --- Eduardo Tejera <edutp00_at_yahoo.com> wrote:
> > > > >
> > > > > > These are great news without any questions.
> > > > > >
> > > > > > I think that is very intersting the problem of interaction
> between
> > > > > small
> > > > > > molecules and proteins were the QM-MM can be applied as a very
> > > usefull
> > > > > tool.
> > > > > > We can use ONION in Gaussian with good results in several case
> but
> > > this
> > > > > > software is not open source like NAMD and with the posibility of
> > > > > "scripts"
> > > > > > the spectrum of applications is very wide. I think that the use
> of
> > > > > > semiempirical methods as well as DFT could be available.
> > > > > >
> > > > > > Best
> > > > > > Eduardo
> > > > > >
> > > > > > Francesco Pietra <chiendarret_at_yahoo.com> escribió:
> > > > > > Hi Chris:
> > > > > > These are wonderful news.
> > > > > >
> > > > > > Of my interest is the interaction of the "organic molecule"
> (read:
> > > > > natural
> > > > > > products, drugs) with biopolymers. Even the first simple step of
> > > > > studying the
> > > > > >
> > > > > > conformations of the "organic molecule" under the influence of
> the
> > > > > > environment
> > > > > > could be best approached by QM-MM.
> > > > > >
> > > > > > I foresee two problems:
> > > > > > (1) Semiempiricals like MNDO, PM3, AM1 deal very poorly with
> > > > > conformational
> > > > > > problems. Perhaps, DFTB is better to this respect.
> > > > > > (2) Parameters for the "organic molecules". However, Jan Saam
> just a
> > > few
> > > > > days
> > > > > > ago has promised to develop Paratool, once is PhD examinations
> are
> > > over.
> > > > > > Paratool could solve these problems, provided it is made simple
> > > > > (automatic)
> > > > > > enough to become manageable by the organic chemist.
> > > > > >
> > > > > > Best wishes
> > > > > >
> > > > > > francesco pietra
> > > > > >
> > > > > > --- Christopher Harrison wrote:
> > > > > >
> > > > > > > Hi Francesco,
> > > > > > >
> > > > > > > We are currently in the early stages of planning and
> developing a
> > > QM-
> > > > > > > MM implementation for NAMD. There's obvious interest from the
> > > > > > > community as well as our own group for this functionality.
> > > > > > >
> > > > > > > We'd be happy to hear about any specific types of QM-MM
> > > calculations,
> > > > > > > functionalities or applications that might be useful to those
> > > > > > > interested.
> > > > > > >
> > > > > > > Best,
> > > > > > > Chris
> > > > > > >
> > > > > > > Christopher Harrison
> > > > > > > Postdoctoral Research Associate
> > > > > > > Theoretical and Computational Biophysics Group
> > > > > > > University of Illinois at Urbana-Champaign
> > > > > > > char_at_ks.uiuc.edu
> > > > > > >
> > > > > > >
> > > > > > > On Aug 28, 2007, at 8:18 AM, Francesco Pietra wrote:
> > > > > > >
> > > > > > > > I wonder why no QM-MM in NAMD. Is that because of disbelief
> in
> > > > > > > > semiempirical
> > > > > > > > electronic calculations, such as MNDO AM1 PM3 DFTB, or is
> just
> > > > > > > > because no
> > > > > > > > developer has yet had time to consider that facet?
> > > > > > > >
> > > > > > > > The other side of the coin is MD included in well
> established QM
> > > > > > > > code suites,
> > > > > > > > which become unmanageable with molecules of common interest------=_Part_1398_23219153.1188407126449--

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 05:20:33 CST