HLA-Modeler: Automated Homology Modeling of Human Leukocyte Antigens.

The three-dimensional (3D) structures of human leukocyte antigen (HLA) molecules are indispensable for the studies on the functions at molecular level. We have developed a homology modeling system named HLA-modeler specialized in the HLA molecules. Segment matching algorithm is employed for modeling and the optimization of the model is carried out by use of the PFROSST force field considering the implicit solvent model. In order to efficiently construct the homology models, HLA-modeler uses a local database of the 3D structures of HLA molecules. The structure of the antigenic peptide-binding site is important for the function and the 3D structure is highly conserved between various alleles. HLA-modeler optimizes the use of this structural motif. The leave-one-out cross-validation using the crystal structures of class I and class II HLA molecules has demonstrated that the rmsds of nonhydrogen atoms of the sites between homology models and crystal structures are less than 1.0 Å in most cases. The results have indicated that the 3D structures of the antigenic peptide-binding sites can be reproduced by HLA-modeler at the level almost corresponding to the crystal structures.

ASEDock-docking based on alpha spheres and excluded volumes.

ASEDock is a novel docking program based on a shape similarity assessment between a concave portion (i.e., concavity) on a protein and the ligand. We have introduced two novel concepts into ASEDock. One is an ASE model, which is defined by the combination of alpha spheres generated at a concavity in a protein and the excluded volumes around the concavity. The other is an ASE score, which evaluates the shape similarity between the ligand and the ASE model. The ASE score selects and refines the initial pose by maximizing the overlap between the alpha spheres and the ligand, and minimizing the overlap between the excluded volume and the ligand. Because the ASE score makes good use of the Gaussian-type function for evaluating and optimizing the overlap between the ligand and the site model, it can pose a ligand onto the docking site relatively faster and more effectively than using potential energy functions. The posing stage through the use of the ASE score is followed by full atomistic energy minimization. Because the posing algorithm of ASEDock is free from any bias except for shape, it is a very robust docking method. A validation study using 59 high-quality X-ray structures of the complexes between drug-like molecules and the target proteins has demonstrated that ASEDock can faithfully reproduce experimentally determined docking modes of various druglike molecules in their target proteins. Almost 80% of the structures were reconstructed within the estimated experimental error. The success rate of approximately 98% was attained based on the docking criterion of the root-mean-square deviation (RMSD) of non-hydrogen atoms (< or = 2.0 A). The markedly high success of ASEDock in redocking experiments clearly indicates that the most important factor governing the docking process is shape complementarity.

Ph4Dock: pharmacophore-based protein-ligand docking.

The development and validation of the program Ph4Dock is presented. Ph4Dock is a novel automated ligand docking program that makes best use of pharmacophoric features both in a ligand and at concave portions of a protein. By mapping of pharmacophores of the ligand to the pharmacophoric features that represent the concaves of the target protein, Ph4Dock realizes an efficient and accurate prediction of the binding modes between the ligand and the protein. To validate the potential of this unique docking algorithm, we have selected 43 reliable crystal structures of protein-ligand complexes. All of the ligands are druglike, and they are varied in nature. The diffraction-component precision index (DPI) originally used in crystallography was applied in this study in order to evaluate the docking results quantitatively. The root-mean-square deviation (rmsd) between non-hydrogen atoms of the ligand in the prediction and experimental results were analyzed using DPI. The rmsd values for 25 structures, consisting of almost 60% of the dataset, are less than three times of the corresponding DPI values. It means that the precision of docking results obtained by Ph4Dock is mostly equivalent to the experimental error in these cases. The present study has demonstrated that Ph4Dock can accurately reproduce the experimentally determined docking modes if the reliable crystal structures are used. Normally the success rate of the docking is judged using rmsd < or = 2.0 A as the criterion. The Ph4Dock marked an appreciably good success rate of 86% based on this criterion.