Shunji Matsumoto

The Feasibility of an Efficient Drug Design Method with High-Performance Computers

In this study, we propose a supercomputer-assisted drug design approach involving all-atom molecular dynamics (MD)-based binding free energy prediction after the traditional design/selection step. Because this prediction is more accurate than the empirical binding affinity scoring of the traditional approach, the compounds selected by the MD-based prediction should be better drug candidates. In this study, we discuss the applicability of the new approach using two examples. Although the MD-based binding free energy prediction has a huge computational cost, it is feasible with the latest 10 petaflop-scale computer. The supercomputer-assisted drug design approach also involves two important feedback procedures: The first feedback is generated from the MD-based binding free energy prediction step to the drug design step. While the experimental feedback usually provides binding affinities of tens of compounds at one time, the supercomputer allows us to simultaneously obtain the binding free energies of hundreds of compounds. Because the number of calculated binding free energies is sufficiently large, the compounds can be classified into different categories whose properties will aid in the design of the next generation of drug candidates. The second feedback, which occurs from the experiments to the MD simulations, is important to validate the simulation parameters. To demonstrate this, we compare the binding free energies calculated with various force fields to the experimental ones. The results indicate that the prediction will not be very successful, if we use an inaccurate force field. By improving/validating such simulation parameters, the next prediction can be made more accurate.

Large-scale integrated super-computing platform for next generation virtual drug discovery

Traditional drug discovery starts by experimentally screening chemical libraries to find hit compounds that bind to protein targets, modulating their activity. Subsequent rounds of iterative chemical derivitization and rescreening are conducted to enhance the potency, selectivity, and pharmacological properties of hit compounds. Although computational docking of ligands to targets has been used to augment the empirical discovery process, its historical effectiveness has been limited because of the poor correlation of ligand dock scores and experimentally determined binding constants. Recent progress in super-computing, coupled to theoretical insights, allows the calculation of the Gibbs free energy, and therefore accurate binding constants, for usually large ligand–receptor systems. This advance extends the potential of virtual drug discovery. A specific embodiment of the technology, integrating de novo, abstract fragment based drug design, sophisticated molecular simulation, and the ability to calculate thermodynamic binding constants with unprecedented accuracy, are discussed.

ES/SDEM - software development engineering methodology for expert systems

Abstract ESs (Expert Systems) are being developed by an increasing number of organizations, and more people are interested in this new technology. While knowledge representation and AI tools are discussed among AI enthusiasts, the notion of an ES building methodology still seems to be widely considered as an artistic endeavour. To a certain extent, this is true, because KA (Knowledge Acquisition) includes artistic aspects. However, it is important to provide a methodology to standardize the ES development process, to reduce the workload of KEs (Knowledge Engineers), and to educate new KEs. We have developed ES / SDEM (Software Development Engineering Methodology for Expert Systems) by examining successful case histories from among 120 ESHELL ∗ (FUJITSU EXPERT SHELL) applications. In this paper, the outline of ES / SDEM will be given.