Dear Min,
in addition to what Greg Warren wrote, to my knowledge, there is still a
big uncertainty in the calculation of desolvation energies. A colleague
of mine, a theoretical physicist, when I was at Hoffmann-La Roche, tried
to calulcate the desolvation energy contribution to the total binding
energy of drug compounds to protein active sites, and estimated that the
errors in the desolvation energies had about the same magnitude as the
total net binding energy. So, this is another big source of inaccuracy,
especially with very different drug compounds.
Best regards,
Dirk.
Am 22.07.10 23:13, schrieb Greg Warren:
Dear Min,
There is no software that can accurately calculate the binding energy
of a predicted (docking) or solved (X-ray structure) protein-ligand
complex particularly for ligands that are different chemically -- not
part of a congeneric series. While there are a number of
publications on methods (LIE, FEP, MD, MMPB(GB)SA, QM/MM, even
semi-empirical QM methods) for predicting binding energy, most of
these publication show results for congeneric series compounds not
chemically diverse ligands which is the result you will get from
screening a chemical database with Autodock. There are publications
by SP Brown J Chem Inf Model. 2007 Jul-Aug;47(4):1493-503, J Chem Inf
Model. 2006 May-Jun;46(3):999-1005 using MMPBSA showing a correlation
between predicted and measured binding energy of around 0.8. The 0.8
correlation may not be in the papers but might have only been shown in
presentations given by Scott. If you are looking for accurate
predictions a 0.8 correlation is not very accurate. If you were
thinking about using docking scores, a publication that looked at
using docking scores to predict binding energy is Warren et al. J Med
Chem 2006, 49, 5912. The last third of this publication showed that
using docking scores to predict binding energy was a very bad idea
even for congeneric series.
Regards,
Greg
*From:* CCP4 bulletin board [mailto:[email protected]] *On Behalf
Of *WEI MIN
*Sent:* Thursday, July 22, 2010 1:51 PM
*To:* [email protected]
*Subject:* [ccp4bb] which software can calculate binding energy
Dear All
I have been using the Autodock software for docking the drug candidate
compounds into the receptor structures. I would like to calculate
accurately the binding energy. Would anyone introduce me a software?
Any help would be greatly appreciated.
Best
Min
--
*******************************************************
Dirk Kostrewa
Gene Center Munich, A5.07
Department of Biochemistry
Ludwig-Maximilians-Universität München
Feodor-Lynen-Str. 25
D-81377 Munich
Germany
Phone: +49-89-2180-76845
Fax: +49-89-2180-76999
E-mail: [email protected]
WWW: www.genzentrum.lmu.de
*******************************************************
