[Dock-fans] gbsa hawkins

Scott Brozell sbrozell at scripps.edu
Tue Mar 13 18:42:57 PST 2007


Hi,

On Mon, 12 Mar 2007, Andreia Palmeira wrote:

> I tried to do a docking considering the solvent effect. I've used
> gbsa_hawkins.in and the mol2 file of the molecules that I've obtained from a
> previous docking (from which I'd had very nice negative grid scores). When I
> considered the solvent effect the values of GB/SA score became huge, higher
> than 100 positive!

It may be that good scoring poses for the Grid Score are not good
scoring poses for the GB/SA Score; after all fortuitous cancellation
of error is the bread and butter of much computational chemistry.
And GB is on the bleeding edge.

However, you didn't include any details of your investigation of the
problem.  One can imagine several scenarios, such as a mere translation
of scores, bad contacts in poses scored with one score because the
pose was generated with a different score, one really bad contact
in all the poses for which one score is very sensitive, etc.

What is the comparison between the Grid Score and the GB/SA Score
score breakdowns (see test suite snippets below) ?
How does the ranking compare between the scores ?
What can you glean from the verbose ( -v ) output ?
How thorough was your receptor preparation ?

Note the warnings about the need for minimization in the manual:
http://dock.compbio.ucsf.edu/DOCK_6/dock6_manual.htm#HawkinsGBSA
If you send your Hawkins GB/SA Score Parameters
then a Hawkins GB/SA Score expert might spot something.


> My ligands have negative charges... I don't know if that
> may be the problem when including the solvent. Is there anyway that I can
> re-score my negatively charged molecules considering the solvent effect?

For negative ions the salt concentration might be important.

You could try the Amber Score: this might give you more control
over the GB calculation but at the price of potentially much
greater preparation.

See these links to perhaps similar questions about GB:
http://amber.ch.ic.ac.uk/archive/200608/0220.html
http://amber.ch.ic.ac.uk/archive/200209/0008.html

Scott

"grid1.dockout.save"
Molecule:  C1

 Elapsed time for docking: 0 seconds

 Anchors:      1
 Orientations:    1
 Conformations:      1

    Grid Score:          -29.223606
           vdw:          -26.868692
            es:           -2.354913


1 Molecules Processed
Total elapsed time:  0 seconds

"hawkins_gbsa1.dockout.save"
Molecule:  C1

 Elapsed time for docking: 37 seconds

 Anchors:      1
 Orientations:    1
 Conformations:      1

   GB/SA Score:          -58.245907
           vdw:          -53.120110
            es:          -16.514385
            gb:           16.258278
            sa:           -4.869689


1 Molecules Processed
Total elapsed time:  75 seconds
~


More information about the Dock-fans mailing list