Dear jz,
You might first consider using R.E.D. Server and automatically
generating the N-term, C-term and central fragments of a
regular/modified amino-acid residue starting from a single dipeptide:
Please, see
http://q4md-forcefieldtools.org/Tutorial/Tutorial-3.php
http://q4md-forcefieldtools.org/Tutorial/Tutorial-3.php#24
versus
http://q4md-forcefieldtools.org/Tutorial/Tutorial-3.php#25
&
http://q4md-forcefieldtools.org/Tutorial/P2N/All-frag-Pept/listing-6mol.pdf
lists the files generated.
then, once understood you could continue to use this tool or try
following non-automated approaches using R.E.D. Server/R.E.D. IV or
III.x...
Concerning multiple molecular orientations the idea of the rigid-body
reorientation algorithm (RBRA) is only to control the mol. orientation
of each optimized geometry independently of the initial orientation
selected by the user or that generated by the QM program after
geometry optimization. In general, we use pairs of orientations using
non-linear X Y Z atoms (the code checks by now if the selected atoms
are linear & print a warning in this case) defined as it follows: X Y
Z | Z Y X (or Y X Z | Z X Y | ...). This allows cancelling out the
effect of the 1st orientation by the 2nd one. R.E.D.
Server/Ante_R.E.D. 2.0 automatically proposes by now sets of atoms to
be involved in the RBRA approach.
Concerning multiple conformations you could select conformation(s)
based on experimental data; or following the approach by Cieplak et
al. (JCC 1995) or Duan et al. (JCC 2003); or adopt your own approach
based on rules you decided; for instance the lowest minimum/minima
obtained after geometry optimization (with small dE values between
conformations).
regards, Francois
> i am trying to derive the charges for a modified amino acid and am working
> through the collage of information out there. i am trying to utilize the RED
> programs that the RED guys were nice enough to put together. i am using
> gaussian for the geo-optimization and charge calculation and the feeding it
> into the proper places in the AmberTools suite.
>
> i've convinced myself, using a simple test system, that it is time to move
> on to the more complicated, modified amino. i am now kind of confused at the
> proper way to go about generating the multiple confirmations necessary to
> fit this more complicate molecule. is there a typical way to generate
> multiple conformations and orientations for molecules, such as drug
> molecules or drug molecules covalently bound to an amino, that have not been
> parametrized yet? or if there is no typical way, how do any of you do it?
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Oct 27 2011 - 02:00:05 PDT
