[Dock-fans] positive energy with water molecule

[Scott Brozell]

Hi,

On Mon, 8 Sep 2008, Anuradha Mittal wrote:

> I am carrying out amberdock recoring calculations on a set of ligands. There
> is a water molecule in the protein that I need to include while performing
> the calculations.
> 
> I am getting positive amber score for all the ligands. However if I remove
> that water molecule all scores are negative. I prepared all structures using
> sybyl but did change the nomenclature according to amber.

What exactly does this mean ?
Did you execute prepare_amber.pl ?
If a water exists in the receptor pdb file argument to prepare_amber.pl
then a TIP3P water model will be used.  So if the pdb has, e.g.,
TER
ATOM  30277  O   WAT  9659       0.444  -0.413 -43.191  1.00  0.00
ATOM  30278  H1  WAT  9659      -0.232  -0.514 -42.521  1.00  0.00
ATOM  30279  H2  WAT  9659       1.267  -0.569 -42.727  1.00  0.00
then the prmtop will have a WAT residue.

> What am I doing wrong here? Is it that Amber-dock doesn't support water
> molecules? Or do I need to explicitly define the parameters for water? Any
> suggestions?

Since there is a change in the Amber rescores with and without water,
the water seems to be there.  The simple diagnostic route starts with
visualization, and probably the extra details from verbose dock6 (-v)
calculations will be informative.  The most likely source of the positive
energy is a clash which should show up in both of the above, but the phase
space of possibilities is large.

Scott



Scott