[Dock-fans] Need some help about amber score

Scott Brozell sbrozell at scripps.edu
Mon Apr 7 15:47:43 PDT 2008


Hi,

On Mon, 7 Apr 2008, Äî Áõ wrote:

>   First of all, I really appreciate Carlos' reply. Maybe because I didn't interpret my last problems clearly, the reply is not what I expected. 
>   The fact is that I have tried the amber tutorial and got the result "dock.out". And I found that the result in the dock.out file is as follows:
>    
>   Molecule: AAH
>    Elapsed time for docking: 71 seconds
>    Anchors:  1
>  Orientations:  1
>  Conformations:  1
>      Amber Score:           -5.119654
>        complex:        -4658.695412
>       receptor:         4584.941716
>         ligand:           68.634042


The above is the web dock.out.  Your output also looks fine;
note that the receptor is not movable and its energy is identical:

>   I have tried the amber score tutorial and found that the result in the output_scored.mol2 file is as follows:
>   ########## Name:  AAH
> ##########   Amber Score:           -4.254279
> ##########       complex:        -4657.768593
> ##########      receptor:         4584.941716
> ##########        ligand:           68.572598


> The problem puzzling me is what does the negative (or positive) of complex (or receptor) represent for? I know that negative amber socre represent for good binding .But how about the others? The result of my own complex is as follows:

I tried to explain this in the manual:
http://dock.compbio.ucsf.edu/DOCK_6/dock6_manual.htm#AMBERScore
The AMBER score output in the DOCK output contains energy terms for each species (complex, receptor, and ligand) such that the sum of the terms is equal to the score.


But this might be a better output format:
########## Name:  AAH
##########   Amber Score:           -4.254279
##########       complex:        -4657.768593
##########     -receptor:         4584.941716
##########       -ligand:           68.572598

Note the negative signs which may help to indicate that the
receptor and ligand energies are shown negated.


>   ##########  Name:  ZINC00242930
> > ##########   Amber Score:          -23.744446
> > ##########       complex:              1312969.508014
> > ##########      receptor:              -1312992.442160
> > ##########        ligand:               -0.810299
> Why the energy of my complex is positive? Is that bad?

Because for this molecular mechanics force field the internal energy 
is positive indicating a highly unstable system.
The large positive receptor internal energy suggests that its coordinates
have not been relaxed with this force field.  See
http://amber.scripps.edu/tutorials/basic/tutorial1/index.htm
Since the Amber score is E(Complex) - [ E(Receptor) + E(Ligand) ],
it may not be bad, and if it is not bad then it would be a clear example of
cancellation of error; the bread and butter of much computational chemistry.
Naturally, it would be better if the receptor was relaxed.

>   Sorry for another question. I noticed that the amber tutorial said :
>   EMM = EvdW + Ees + Eint ---- obtained from AMBER MM potentials.
>   What is "Eint". I remember that, in grid based score, Eint=Evdw+Ees. How about the equation above? 

Yes, Eint (internal energy) is a misnomer here.  It is sloppy nomenclature
because that term is actually only the bond, angle, and dihedral
http://amber.scripps.edu/eqn.txt
contributions to the internal energy.
http://en.wikipedia.org/wiki/Internal_energy

Scott




More information about the Dock-fans mailing list