Re: AMBER: The PB GB energy from mm_pbsa will be positive on high charged system

From: David A. Case <>
Date: Fri, 18 Aug 2006 10:50:32 -0700

On Wed, Aug 16, 2006, Suxin Zheng wrote:

> I am doing mm_pbsa on high charged system as RNA with Amber8.0. While
> some ligand also has negative charge (toward solvent) the PBTOT and
> GBTOT will be positive as below. And the ligand with different
> charges will give much more difference in PB and GB. It's hard to
> compare different ligand's energy.
> How to deal with such situation. I use the default parameters in
> mm_pbsa.
> # -----------------------
> # =======================
> ELE 4009.76 0.00
> VDW -24.66 0.00
> INT 0.12 0.00
> GAS 3985.23 0.00
> PBSUR -2.55 0.00
> PBCAL -3914.80 0.00
> PBSOL -3917.35 0.00
> PBELE 94.96 0.00
> PBTOT 67.88 0.00
> GBSUR -3.67 0.00
> GB -3941.00 0.00
> GBSOL -3944.67 0.00
> GBELE 68.76 0.00
> GBTOT 40.56 0.00

It is hard to say much without more information. I gather(?) that the DELTA
above refers to binding of a negatively charged ligand to an RNA, and that
your concern is that the computed net free energy (not including entropy
terms) is positive. I'm also guessing(?) that you think this is not wrong,
perhaps because there is some experimental data that contradicts this.

As you can see, I'm having to guess a lot about what the real problem is,
especially since to also refer to "ligand with different charges". It is
known that computing the absolute affinity of RNA for charged ligands by
MM-PBSA is quite a difficult chore. See for example (for a positive ligand):

%A V. Tsui
%A D.A. Case
%T Calculations of the Absolute Free Energies of Binding between RNA and Metal
Ions Using Molecular Dynamics Simulations and Continuum Electrostatics
%J J. Phys. Chem. B
%V 105
%P 11314-11325
%D 2001

On the other hand, neutral ligands seem to fare much better, e.g.:

%A H. Gouda
%A I.D. Kuntz
%A D.A. Case
%A P.A. Kollman
%T Free energy calculations for theophylline binding to an RNA aptamer:
%Comparison of MM-PBSA and thermodynamic integration methods.
%%J Biopolymers
%V 68
%P 16-34
%D 2003

Maybe looking at these, or other similar calculations, will provide some
insight. If you in fact have a negative ligand binding to RNA, you might also
want to look at the (computed and experimental) salt-dependence.


The AMBER Mail Reflector
To post, send mail to
To unsubscribe, send "unsubscribe amber" to
Received on Sun Aug 20 2006 - 06:07:20 PDT
Custom Search