AMBER Archive (2007)Subject: Re: AMBER: antechamber/charges/conformation
From: Francesco Pietra (chiendarret_at_yahoo.com)
Date: Mon Jun 11 2007 - 04:44:21 CDT
Thanks and see below, please.
--- FyD <fyd_at_q4md-forcefieldtools.org> wrote:
> Quoting Francesco Pietra <chiendarret_at_yahoo.com>:
>
> > Three questions related to "charges computed are
> those
> > for the conformation you provide" (manual,
> > antechamber, p. 80, lines 2-3).
> >
> > (1) I wonder what the expectation may be for a MD
> > conformational search of a large non-polymeric
> > molecule (i.e. no repeating units) starting
> > antechamber with an approximately minimized
> structure.
> > Approximately so, because MM programs do not allow
> a
> > preliminary systematic global conformational
> search
> > for so many degrees of freedom. Notice that,
> because
> > of chirality axes and fixed asymmetric centers
> > present, the different conformers are in a
> > diastereoisomeric relationship.
>
> When you do a conformational search, the H-bonds
> available in your
> 'big' molecule will be important driven forces for
> defining your
> minimum. This might not what you wish...
All hydrogens, at C or heteroatoms, are clearly
defined by the MM package, like all "heavy" atoms.
I.e., Cartesian coordinates are provided for them at
the same precision as for all other atoms, and no
hydrogen is missing, otherwise MM could not work.
Problems may arise on loading to MD a MM-generated pdb
file: the MM package provides a pdb file with all
CONECT defined, which typically are not understood by
MD packages, i.e. MD packages do not use true pdb
files. In addition, I have verified for thiophenol
that the Sybyl mol2 file generated by the MM package
differs somewhat from that illustrated on Amber9
manual.
>
> > (2) Are the above prospects better if the large
> > molecule is first cut into fragments and the
> charges
> > computed from them? What about the reliability of
> the
> > dipole moment obtained this way? Notice that
> fragments
> > still contain movable dihedrals (it is
> unconceivable
> > to cut the molecules into such small fragments
> that do
> > not comprise dihedrals).
> >
> > (3) Can "divide and conquer" be manually driven to
> cut
> > the molecules according to chemical intuition,
> i.e.,
> > can breaking bonds be manually selected?
>
> Usually you cut your molecules into fragments "that
> make sense"...
I have experience in creating fragments; I was asking
if "divide and conquer" on Amber9/antechamber allows
the user to make his choice in choosing fragments.
> See for instance Cieplak et al. J. Comput. Chem.,
> 1995, 16, 1357-1377.
>
> You have examples of fragment building in the
> Tutorial available @
> http://q4md-forcefieldtools.org/Tutorial/
>
> We are about to release in R.E.DD.B. many complex
> force field topology
> databases generated by R.E.D. III and IV beta. Check
> R.E.DD.B. & you
> should find examples of fragment building.
>
> regards, Francois
>
>
>
-----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber_at_scripps.edu
> To unsubscribe, send "unsubscribe amber" to
> majordomo_at_scripps.edu
>
____________________________________________________________________________________
Boardwalk for $500? In 2007? Ha! Play Monopoly Here and Now (it's updated for today's economy) at Yahoo! Games.
http://get.games.yahoo.com/proddesc?gamekey=monopolyherenow
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu
|