[Product Releases]

Most recent post


Can we trust docking results?
Sept 2010

IBM Systems and Technology Group releases a white paper with eHiTS and Cell
Oct 2008

EPA's ToxCastTM project will use SimBioSys' eHiTS as docking engine
Nov, 2007


243rd ACS
Mar 25-29, 2012
San Diego, CA
see >> more


Presentation at the:

 IUCr 2008 Osaka, Japan

XXI Congress and General Assembly of the International Union of Crystallography, 23-31 August, 2008

Sunday, August 24 - Monday, August 25, 2008 

03. Computers in Analysis, Molecular Modelling and Molecular Design
04. Rational Drug Design

Effects of initial conformations of small ligands on computational docking accuracies

Authors: Akifumi Oda1, Ohgi Takahashi1, Noriyuki Yamaotsu2, Shuichi Hirono2

(1) Tohoku Pharmaceutical University, 4-4-1, Komatsushima, Aoba-ku, Sendai,  Miyagi, 981-8558, Japan
(2) Kitasato University, 5-9-1, Sirokane, Minato-ku,
Tokyo, 108-8641, Japan
Abstarct: Computational ligand docking is one of the most important techniques  of Structure-Based Drug Design, which makes the most of 3D-structures of drug  target proteins determined by experimental studies, such as NMR or  crystallographic analyses for the drug discovery and development. In this  study, the effects of initial conformations of ligands on computational  docking were investigated, and appropriate settings of conditions for computational docking were determined. Five types of initial conformations were prepared, and docking calculations were carried out by using each conformation as inputs. Furthermore, several settings of docking parameters were used (default, accurate, high throughput, etc), and robust settings for various initial structures were investigated. GOLD and eHiTS were used as
docking software, and structurally known protein-ligand complexes were used  as test set. Root mean square deviations between computational and  experimental structures (RMSD) were adopted for criteria for evaluations, and  the docking pose with RMSD < 2.0 were defined as "reasonable poses". When  at least one of the generated poses by a docking trial was reasonable, the  trial was defined as "success", and when the top ranked pose, i.e. the pose
with the lowest binding free energy, was reasonable, the trial was defined as "top pose success". The search abilities of docking were evaluated by "success rate" and "top pose success rate". As the results, bad initial conformations, which were much different from crystal ligand structures, cause the worst success rate and the worst top pose success rate in all initial conformations. Comparing GOLD and eHiTS, eHiTS was better than GOLD to obtain reasonable poses regardless of rankings, but GOLD was better to obtain reasonable top poses.

Keywords: computer-aided drug design, conformational

Copyright (c) 2008 SimBioSys Inc., All rights reserved.