■Development of Efficient Molecular Design Integrated System of Functional Polymers by Elongation Method Yuriko Aoki (Professor, Kyushu University) The conventional quantum chemical approach based on the molecular orbital method is difficult to apply to large systems like biopolymers, etc., though it is useful for small molecules. We are developing the Elongation method that makes us possible to obtain the electronic states of large aperiodic polymers, treating only a few units at a time of the total system. This study develops the construction of molecular design integrated system of functional polymers and the relation software by combining the quantum-chemical method for obtaining physical properties with the efficient Elongation method. This software becomes an assistance of the molecular design of functional polymers, and drastic cost reduction of materials synthesis can be expected. Back ■Large Scale Computation for Correlated Electron Systems Masatoshi Imada (Professor, The University of Tokyo) Efficient algorithms for accurate computation of correlated electron systems are desired from various fields of materials research and applications. This is because many-body electrons in matter play crucial roles in determining the material properties. In this project, a new hybrid algorithm is developed by taking the path integral renormalization group method as a core and by combining with large scale computation algorithm from first principles. The algorithm will be applied to correlated electron systems such as transition metal compounds. This algorithm will enable us to simulate strongly correlated electron systems from the first principles approach with accurate enough treatment of the Coulomb interaction and its fluctuation effects. Back ■Development of the Program Package for the Relativistic Molecular Theory Takahito Nakajima (Associate Professor, The University of Tokyo) Nowadays, scientists treat the materials including a wide variety of atoms, such as superconductors and semiconductors. The object of this proposal is to construct the new accurate theoretical chemical approach that is able to treat truly large molecular systems with the relativistic molecular theory. The next generation of the molecular orbital theory will be developed, and its program will also be published under the open source concept. Compared with the conventional theories or programs, the accuracy of calculation will be improved by 100 or more times in this study. Back ■The Formation of Foundation for Practical Use by an Innovation of Singular Value Decomposition Algorithm Yoshimasa Nakamura (Professor, Kyoto University) Singular value computation (SVC) or singular value decomposition (SVD) of matrices is an important operation in numerical linear algebra. In this project we propose a new SVC-SVD algorithm named an integrable algorithm which is faster and has a higher accuracy than known ones, and then we design a standard package. Singular values of a matrix having one million entries are computed in a computational complexity about 60 percent less that the known standard method. It is expected to apply this new algorithm to a wide area in simulation technology. Back ■Fluid-structure Interaction Resonance Phenomenon Inverse Analysis for Micro Fluid Device Development Junichi　Matsumoto (Research Staff, National Institute of Advanced Industrial Science and Technology) As for a micro fluid device the application to medical treatment and the information industry is expected. This research does the resonance control analysis that perceives to the resonance phenomenon of the fluid-structure interaction problem and moves the solid (the wall) with the actuator resource of a small displacement by well utilizing resonance and make it possible to move the fluid largely and inspect the efficacy with the large-scale parallel three-dimensional analysis. The result of this research is scheduled to disclose as the analysis program and contribute to the commercialization of a near future micro fluid device. Back ■Distributed Optimization Simulation for Discrete-Continuous Systems Kazuo Murota (Professor, The University of Tokyo) Optimization simulation of large distributed systems are getting more important　than before. This project focuses on large discrete-continuous distributed systems　that consist of noncooperative independent subsystems. In particular, it is assumed　that the entire mathematical model is not precisely known. The objective of this　project is to develop optimization algorithms that can be used for simulation of such distributed systems. Back ■Hybrid Molecular Dynamics Simulations for Soft Matters Ryoichi Yamamoto (Associate Professor, Kyoto University) Soft matters, such as colloidal dispersions or emulsions, are important industrially and technologically. It is, however, very difficult to predict physical properties of the soft matters, since they are usually composed of hierarchical meso-scale units. Neither theoretical approaches nor even computer simulations to those matters have been successful, compared with simpler matters, due to the nature of complexity involved. The purpose of this project is to develop a useful computational method “Hybrid Molecular Dynamics Simulation” by using a hybrid description, where large/slow degrees of freedom are modeled as particles but small/fast ones are modeled as fields. It can definitely help to understand the nature of interesting phenomena observed in various soft matters. Back ■Development of a Biology-Inspired Dynamic Flying Simulator Hao Liu (Professor, Chiba University) Till now robotic flapper is responsible for study of insect flapping wing mechanisms as in hovering flight but quick-turn like freely flapping flight keeps unknown. In this project we aim at establishing a biology-inspired dynamic flying simulator, which is capable to mimic the free flights involving hovering, forward flight and quick-turn on a basis of realistically modeling of geometry and kinematics and accurately modeling of dynamics. It is expected that the developed simulator would provide also novel theories and technical innovations for research and development in the Micro Air Vehicles (MAVs). Back
	■Development of Simulation Technique for DNA Nanodevice Production Satoyuki Kawano (Professor, Osaka University) DNA has an electrical conductivity through the base pair and a self-organizing function. It is expected as the material for functional device in bionanoelectronics because of the applicability to the living body and the high possibility of mass production. Ultimate goal of the present research is to realize a computer aided design of DNA nanodevice in biotechnology by the use of novel dynamics simulation technique. DNA is treated as a functional material and the application can be expected in the research areas of electron devices and materials sciences. Back ■Development of Multi-Physics Simulator Based on Quantum Chemical Molecular Dynamics Momoji Kubo (Associate Professor, Tohoku University) For the further miniaturization of the semiconductor devices, it is essential to clarify the multi-physics phenomena including chemical reactions on atomic levels. This project aims the development of new electronic- and atomic-scale simulator, which can elucidate various multi-physics phenomena including chemical reactions, by the integration of the quantum chemical molecular dynamics and non-equilibrium classical molecular dynamics methods. This new simulator enables new process and material design for semiconductor devices, which cannot be realized by the other simulation methods. This project also contributes to the realization of the multi-scale simulation. Back ■Development of AB Initio Simulation Methods for Quantum Hydrogen Systems Masanori Tachikawa (Professor, Yokohama City University) A hydrogen atom is the most ubiquitous species in Nature and plays a central role in most part of science phenomenon. The purpose of the project is to develop new quantum simulation methods, where the hydrogen nuclei as well as electrons are treated quantum mechanically. This method will be combined with the conventional schemes to elucidate the universal concept over the wide variety of systems ranging from material science to biochemistry. Back ■Development of Dynamic Bond Type Large-scale Molecular Dynamics Method Takanobu Watanabe (Associate Professor, Waseda University) Future nano and bio-technologies require a large-scale atomistic simulation technique applicable to various systems of composition involving chemical reactions. This research is aiming at the development of a novel molecular dynamics (MD) simulation method that treats not only the motion of nuclei, but also the motion of additional degrees of freedom corresponding to dangling bonds. In this scheme, an attractive interatomic interaction is caused only if two atoms overlap their bonds. The coordination number of an atom is controlled by the number of the dynamic bonds, whereas, in conventional empirical potentials, it has been controlled by unnatural steric hindrances caused with many-body interaction terms, or by fixing the binding topology. The new method drastically simplifies the parameter fitting procedure so that it will become easy to design new potential fields for complicated systems consisting of many types of elements. The dynamic bond type MD method can be expected as a leading technology of next generation large-scale MD to broaden its application to a wide range of chemical reactions. Back
	■Scale-free Metabolic Simulation Masanori Arita (Associate Professor, The University of Tokyo) Our research realizes an on-line metabolic map that can seamlessly represent and trace metabolic flows including lipids, sugars, and other secondary metabolites. The integration of molecular flows in different scales will shape the foundation of biological simulation and functions as the core of future bioinformatics. Back ■Development of the Bilateral Multiscale Neural Simulator Mihoko Otake (Associate Professor, The University of Tokyo) There is an urgent need for the technological development which is applicable to diagnosis, treatment and prevention of nervous diseases which cause movement disorders because of the rapid aging of this country. In this study, the multiscale simulator is developed which computes both macroscopic whole body motions and microscopic neuronal activities bilaterally. Four-tiered detailed neural model is built comprising molecular, cellular, tissular and individual levels based on anatomy and physiology, whose activities are calculated in parallel. The movement and the internal state of the nervous system will be estimated, which are altered by the chemicals and exercises. The research is promising for the novel diagnosis of neurological disorders and their treatments through medication and movement therapy. Back ■Tailor-made Tissue Regeneration Therapy 　Supported by a Novel Bone Remodeling Simulation Ken-ichi Tezuka (Associate Professor, Gifu University) Bone is a complex system with functions including those of adaptation and repair. To understand how bone cells create a structure adapted to the mechanical environment, we proposed a bone remodeling model based on a reaction-diffusion system. By using this model, we are going to simulate abnormal state of bone tissues affected by various diseases. Next, we will try to use this simulation to design and produce artificial bones, that can be used for a new generation of tissue regeneration therapy. Back ■Computer-assisted Design of Human Antibodies Highly Specific for Targeting Molecules Tyuji Hoshino (Associate Professor, Chiba University) When a disease-causing substance (antigen) has entered in a body, some adequate molecule (antibody) that has a strong binding ability to the antigen is selected through self protect mechanism; immune system. We are planning to develop an innovative methodology to determine the highly specific antibodies for a targeting antigen, taking full advantage of computer simulation. This computational technology will enable us to design and produce the effective medical antibodies, especially for immune-therapy on oncogenesis. Back ■Precise Electronic Structure Calculations on Biochemical Molecules Including Transition Metal Ions Ryo Maezono (Research Staff, National Institute for Materials Science) Ab-initio quantum Monte Carlo method using modern parallel computing is a promising approach to perform accurate quantum simulations for biochemical issues which require high resolution in energy. Biomolecules including transition metal ions are the systems of great interests possessing useful biological properties. To deal with such systems developing and improving the model potentials to describe transition metal ions is highly in demand, which is the aim of the present study along with the objective to construct the basis for practical applications of simulation technics. Back