[Dock-fans] A huge difference in results

[Scott Brozell]

Hi,

On Thu, 9 Aug 2007, hassan shallal wrote:

> I am a beginner of using dock. I just have finished my first dock experiment
> using the tutorials posted in
> http://dock.compbio.ucsf.edu/DOCK_6/tutorials/index.htm.
>
> I compared the results I obtained in My experiment (attached are two files,
> My_flex_scored.mol2 and My_rigid_scored.mol2) and the results posted in the
> tutorial (attached are other two files, (Tutorial_flex_scored.mol2 and
> Tutorial_rigid_scored.mol2).
>
> I found that there is a good agreement between the obtained and the posted
> VDW results. However, a huge difference within the range of 8-9 was observed
> in both the case of grid score and electrostatic interaction.

########## Name:     1ABE.pdb
##########    Grid Score:          -41.291992
##########           vdw:          -21.404663
##########            es:          -19.887329

########## Name:     1ABE.pdb
##########    Grid Score:          -36.147507
##########           vdw:          -21.315907
##########            es:          -14.831600
      6 O1         14.6521   57.3674   56.3454 O.3     1  ARA307     -0.5790

########## Name:     UNNAMED
########## Grid Score:          -33.120888
##########          vdw:          -22.171465
##########           es:          -10.949423

########## Name:     UNNAMED
########## Grid Score:          -28.794184
##########          vdw:          -22.208521
##########           es:           -6.585662
     19 O1         14.5687   57.5120   56.4257 O.3     1  ARA1       -0.3649


> I actually have two questions:
> The first: why this difference happened? is there a problem in my compilers
> that has led to this difference? if yes, how can I solve it?
> The second: how can I calculate the free energy of binding (delta G) by
> using the above values?

Counting ?'s there are 4 questions:
1.  The charges are significantly different for all the atoms.
So the difference is likely due to Coulomb's law.
2.  Unknown, but no evidence of that yet.
Run the test suite to validate your installation.  See
http://dock.compbio.ucsf.edu/DOCK_6/dock6_manual.htm#Installation
and the blurb at the end of that section for other sources of error.
3.  This question does not exist yet.
4.  One could approximate the binding energy as the score.
http://dock.compbio.ucsf.edu/DOCK_6/dock6_manual.htm#OverviewoftheDOCKSuiteofPrograms
But see previous posts for warnings:
http://blur.compbio.ucsf.edu/pipermail/dock-fans/2007-March/000973.html

Scott