Proceedings of the Symposium on Chemoinformatics
28th Symposium on Chemical Information and Computer Sciences, Osaka
Displaying 1-47 of 47 articles from this issue
Keynote Lecture
Plenary Lecture
  • Yukihiro Ozaki, Hideyuki Shinzawa, Sumaporm Kasemsumran, Yi-ping Du, J ...
    Pages JS1
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    Several methods to improve the performance of Partial Least Squares (PLS) regression have been proposed. These methods can be classified into two kinds, called sample selection and wavelength selection method, respectively. In the sample selection method, multi-objective Genetic Algorithms (GA) were combined with PLS to remove specific samples with systematic errors from data set. It also enables to analyze the factor of the systematic errors in detail. In the wavelength selection method, Moving Window PLS (MWPLS), Changeable Size Moving Window PLS (CSMWPLS) and Searching Combination Moving Window PLS (SCMWPLS) were proposed. These methods are based on the use of moving window which selects local wavelength region in the spectral data. By calculating PLS with the move of the window, it is possible to search the informative region or combination of several informative regions. These methods were applied to near infrared and infrared spectral data to evaluate their performances. The results showed the remarkable improvement of PLS model by these methods.
    Download PDF (128K)
  • [in Japanese]
    Pages KS1
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    Download PDF (105K)
Oral Session
  • Shinsaku Fujita
    Pages J01
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    A new concept "mandala" is proposed in order to comprehend the relationship between molecular symmetry and combinatorics. A regular body of a given symmetry (G) is defined by a three-dimensional skeleton having the same number of symmetry-equivalent vertices as the order of the group G. A mandala is defined as a nested regular body in which each vertex is occupied by a regular body. The regular body represents an intramolecular symmetry, while the mandala represents an intermolecular symmetry (stereoisomerism). Thereby, combinatorial enumeration of stereoisomers can be formulated rationally.
    Download PDF (157K)
  • Hiroko SATOH, Atsushi AOKI, Hiroko ASAOKA
    Pages J02
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    Sharing of common technology with open-source basis software is important for the training of software skill, the development of software technology, and succeeding the skill to the coming generation. The most of source-code of chemical software, however, has not been open to public even if it is common technology. The domination of the market of chemical software by the US and EU companies is also serious problem for other parties. We therefore have been developed an open-source graphics library, ChemoJun for chemical visualization, which is one of the basis technology of chemical software. This paper outlines our concept in the ChemoJun project and the functions and contents of this graphics library.
    Download PDF (435K)
  • Kimito FUNATSU, keisuke OHTA, Yukiharu TSUKAMOTO
    Pages J03
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    We have been developing three different types of organic synthesis design systems that contribute to organic chemist's creative research. The AIPHOS (Artificial Intelligence for Planning and Handling Organic Synthesis) is a superior system to create new synthetic pathways using the advantages of empirical knowledge-based approach and logic-centered approach. The KOSP (Knowledgebase-Oriented Synthesis Planning) uses strategies similar to the AIPHOS. The KOSP has a function of recognizing a strategy site by using a knowledgebase converted from the reaction database (RDB). Thus the KOSP is more empirical system than the AIPHOS. The TOSP (Transform-Oriented Synthesis Planning) is empirical approach type system that uses a transform database built from existing reactions. On the other hand, we have been developing reaction evaluation system that predicts reaction products by using a transform database abstracted from the RDB. The reaction evaluation system can predict reaction products about a synthetic precursor proposed by organic synthesis design systems. From the result, we can check reaction products other than a target product. And this system can evaluate synthetic pathways based on the safety and the generating condition of by-product. We have developed a client/server (C/S) system that integrated three organic synthesis design systems mentioned above and the reaction evaluation system. This C/S system can search synthetic pathways by using each system through a common interface. The purpose of this work is to improve the convenience of reaction prediction of the C/S system by developing a function of reaction prediction between two reactants, reaction formula sorting function and the function of multi-step reaction prediction.
    Download PDF (60K)
  • Junichi Goto, Ryoichi Kataoka, Noriaki Hirayama
    Pages J04
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    Development and validation study of a novel docking program, ASEDock is presented. ASEDock is a fast docking program based on a concave shape alignment method. We have defined ASE Model, a simple but accurate representation of concave shape and ASEScore, which evaluates shape similarity between ASEModel and ligand structure. Initial poses of the ligand were refined based on ASEScore in terms of the maximum overlap to alpha spheres and minimum overlap to receptor atoms, followed by full atomistic energy minimization. Simple function form of ASEScore makes it possible to search all reasonable poses efficiently. Because ASE Model accurately represents concave shape, actual ligand structures of binding mode were found in initial poses with best ASEScore. Since posing algorithm of ASEDock is free from any bias except shape, it is a very robust docking method. Validation study has demonstrated that ASEDock can reproduce experimentally determined docking mode of the complex that has no strong interaction, such as hydrogen bond, between ligand and receptor.
    Download PDF (409K)
  • Soh Koike, Fumitoshi Sato, Naoki Ihara, Naoki Tsunekawa, Noriko Nishin ...
    Pages J05
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    To elucidate delicate properties of proteins with high accuracy, we have developed a density functional (DF) program, ProteinDF which can treat the whole protein as a molecule. We have developed as a ProteinDF system extending the functions and performances under the "Frontier Simulation Software for Industrial Science" and it was attained to be able to execute all-electron DF calculation on proteins up to the size of the order of the one hundred residues with few trial and errors in practicable time at present. We are now developing the ProteinDF system for the simulation of protein reactions based on the all-electron structure under "Revolutionary Simulation Software" projects of MEXT. We first plan to develop the methods for optimization of the molecular structure of protein, treatment of protein in the solvent and all-electron calculation on very large-sized protein. The database of protein wavefunctions and high performance GUI are also developed in this project. We will argue about the whole system image and its future view with some results.
    Download PDF (624K)
  • Masaya Kawase, Kiyohito Yagi, Manabu Tamura
    Pages J06
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    It has, recently, been recognized that the clinical education is very important for medical and pharmaceutical student. Therefore, this study aimed to establish the virtual clinical education system for medical and pharmaceutical student. In this study, we tried to apply the analytic hierarchy Process (AHP) to the our objective education system. For examining the utility of AHP, it took the case of an inflammation in a chronic disease vice-nasal cavity. The suggested treatment method by AHP agreed with the judgment of the expert doctor. Thus, it was suggested that AHP will be used in the virtual clinical education system.
    Download PDF (112K)
  • Yukino Ochi, Rika Nishikiori, Kousuke Okamoto, Noriyuki Yamashita, Mas ...
    Pages J07
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    DNA microarray data analysis is one of the core technologies promoting genome research. By measuring expression levels of more than 10,000 genes, DNA microarray analysis enables us to analyze the relationshipas between gene expression and behavior in various biological phenomena such as development, differentiation, growth, canceration, aging and so on. The applications of DNA microarray data analysis have increased in the post-genome age in which the function analyses of genes are more important and a number of multivariate DNA microarray data sets have been produced. Since the dimension of the data set is too high and their interpretations as well as comparisons are too difficult, the methods for analyzing the data set have not been established yet.Under these circumstances, the necessity of the method which enables us to classify or interpret the data set is increasing naturally. In this study, we applied nonlinear classification procedures, SOM and kernel PCA, to the DNA microarray data analysis of human lung carcinoma. As a result, we found 82 genes which were estimated to be associated with carcinogenesis from 12,600 genes.
    Download PDF (831K)
  • Mikio Kaihara, Khoichi Inaba
    Pages J08
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    SIMCA (Soft Independent Modeling of Class Analogy) has several good points, for example, the capability of 2Dimensional mappings of multi dimensional samples or the F-tests for classifications. Therefore, if we could select important variables and realize better and clear classification, such sample selection method should be significantly effective and important for the SIMCA. However, we found that just using variables with high modeling powers did not work well. The more, the number of optimal factors should be, the more, such tendency increases. Therefore, decreasing the number of variables for the SIMCA analysis too much means that the misclassification rate should be increasing. The analogy is one of the most important points for classifications, but it could have the limit. Finally, we tried the other classification methods and compared them, each other.
    Download PDF (146K)
  • Hideyuki Shinzawa, Takehiro Nakagawa, Katsuhiko Maruo, Yukihiro Ozaki
    Pages J09
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    A sample selection method in Partial Least Squares (PLS) model building was proposed. This method aims to improve the performance of PLS model by eliminating some uninformative sample(s) with systematic errors from data set. First, data set is divided into three, training, constraint and test set. Then multi-objective GA is performed with training and constraint set to find the samples with systematic errors. The combinations of the samples to be removed from training or constraint set are given as Pareto solutions. Finally, Pareto solutions are evaluated with test set in terms of correlation coefficients and root mean square of standard error of prediction. The proposed method was applied to the near infrared spectra obtained from the surface of a human skin. The result showed one of the Pareto solutions improves the performance of the PLS model remarkably. It was also used to analyze the effect of systematic errosr in detail. It showed the scatter and random noise in the original data associate with the systematic errors. These results indicate the proposed method is a useful tool not only to improve the performance of the PLS model but also to analyze the effect of systematic errors.
    Download PDF (187K)
  • Masamoto Arakawa, Kimito Funatsu
    Pages J10
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    Variable selection is one of the most important procedures for constructing a simple and predictive model in quantitative structure-activity relationship (QSAR) studies. In this study we propose a method for variable selection using the concept of Pareto optimum and genetic programming. As a case study, QSAR model has been developed for a set of inhibitors of the human immunodeficiency virus 1 (HIV-1) reverse transcriptase, derivatives of 1-[(2-hydroxyethoxy)methyl]-6-(phenylthio)thymine (HEPT). Structural descriptors used in this study are Hansch constants for each substituent and topological descriptors. As a result of variable selection, 11 descriptors were selected from 34 original ones. Then counter-propagation (CP) neural network were constructed with selected variables. The obtained network was accurate, predictive and interpretable and the value of explanation variance was 0.875. The CP model could easily be interpreted and the result was quite similar to previous studies of QSAR models of HEPT derivatives. Moreover in order to confirm a predictive ability of the model, validation test was performed.
    Download PDF (480K)
  • Toshiyuki Mitsui, Shuji Okuyama, Minesada Hida
    Pages J11
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    The different calculation method of regression analysis (DCR) can carry out the determination of each compound correctly without the sample with three or more kinds of compounds mixed also carrying out the separation operation. The DCR calculates the value of the coefficient of heavy regression so that the sum of squares of the difference of the predicted value of the measured value of the unknown sample calculated from the measured value of a known sample and the measured value of an unknown sample and the measured value of an unknown sample may become the minimum. This time, DCR analysis is used to determine the mixing ratio of the forged coin of 500 in which 4 kinds of compounds are mixed. The measured data of X-ray Fluorescence Analysis (XRF) was used for the analysis of the forged coin of 500 yen.
    Download PDF (153K)
  • Takurou Nishimura, Kimito Funatsu
    Pages J12
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    The structure optimization system to which the genetic algorithm was applied for the computer-aided design of a gas separation module was developed. A system consists of three engine and they are creation of the structure of a separation module, calculation of a hydrogen recovery rate, and optimization of a modular structure. It treats the various structure parameters of a separation module as a variable, and makes discreteness ability of a separation module an evaluation value. By extending the package of commercial computer fluid dynamics (CFD), it became possible to evaluate the performance of a gas separation module. Since the gas penetration could be treated reflecting the flow state and concentration polarization of gas, the optimization result which has not been recognized conventionally was able to be obtained. It suggests that this shortens the performance evaluation in a structure design, and can be contributed to development period shortening of manufacture equipment.
    Download PDF (246K)
Poster Session
  • Makoto Ishikawa, Kimito Funatsu
    Pages JP01
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    The sets of molecular descriptors were defined to evaluate reaction applying Chemish. The descriptors are based on the combination of molecular orbital (HOMO, LUMO) and partial charge. All molecular orbitals were caluculated In oder to advance organic molecules, the descriptors are applied to the diastereoselectivity of hydride reduction including cyclic ketones (Adamantan-2-one, Bicyclo[2.2.1]heptan-7-one, 4-tert-Butyl-cyclohexanone). The determination of useful descriptors are analyzed by the combination of Genetic Algorithm(GA) and Partial Least Squares regression (PLS.). As results of the PLS modeling, Q^2 values of each models is more than 0.80. The reaction site (carbonyl group) is recognized in each models.
    Download PDF (127K)
  • Masaki Arisaka, Hiroyuki Koshino, Shungo Koichi, Satoru Iwata, Takeaki ...
    Pages JP02
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    We have developed a system that elucidates candidate partial structures in a project of computer-aided structural elucidation for practical analyses of complicated structures. The system makes elucidation of candidate partial structures with stereochemistry from inputted 13C NMR chemical shift values. The partial structures can be used in constructing whole candidate structures that are inputted to a CAST/CNMR system, which predicts 13C NMR chemical shifts with high accuracy by considering stereochemistry. This paper outlines the concept, algorithms, and the first applications to terpenoids.
    Download PDF (195K)
  • Ikutoshi Matsuura
    Pages JP03
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    The computer system has been developed to convert chemical name of organic compound to the 2D-structural formula. This report describes the process for recognizing stereochemical changes by skeletal modifications of natural product structure. The modifications are described by prefixes 'nor-','homo-','seco-','cyclo-' and 'abeo-'. The stereochemistry of fundamental parent structure on the orientated plane is represented using sequence rule based on locants of atoms connected to a chiral center, by informations of the target atom of alpha-beta, of up or down direction of bondage from the center to the least-prefered connected atom, and of right- or left-handed rotation of three connected atoms in order of preference around the center. Stereochemistry table of atoms presents the three informations of target,direction and rotation and is updated after each skeletal modification. Modification table of a center atom presents information of original connected atom and the replaced one by a modification. After generation of structural diagram, arrangement table of ring atom is generated, presenting location of a ring atom to be entrant or reentrant on each unit ring. Graphic representation of stereochemistry on the structural diagram is achieved by integrating those informations from the tables.
    Download PDF (79K)
  • Ryoichi Kataoka, Junichi Goto, Noriaki Hirayama
    Pages JP04
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    A new docking program, ASEDock, is developed and tested in searching the receptor that could favorably be docked by the query ligand using the testset of the 50 highly resolved complexes selected from PDB. Since it makes good use of Gaussian-type function for evaluating and optimizing the overlap between ligands and site model, it can pose ligands onto the docking site relatively faster and more effectively than using potential energy functions. The fitness after posing is scored by the the overlapping fitness score and pKi derived by fitting the hydrogen-bonds, the hydrophobic contacts, and the entropy changes to the reported pKi's of the complexes in PDB. ASEDock reproduced the original ligand-receptor complexes, and, furthermore, revealed possible pairs that could dock with each other to form stable complexes. Therefore, it could be useful in elucidating the receptors that could be influenced by the binding of query ligands and detect possible side effects by the virtual screening with ASEDock.
    Download PDF (1042K)
  • Shinya Morishita, Kenshu Kamiya, Yoko Sugawara
    Pages JP05
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    Adenylate kinase is an enzyme which catalyze phosphate transfer amoungst ATP and AMP, and also ADP and ADP. In order to elucidate the nature of the catalytic reaction of adenylate kinase, we carried out theoretical calculations using various level of the theory, Molecular Mechanics, PM3, ab initio, and ONIOM method. With stuructural studies, the reaction is expected to proceed with SN2 like inversion of phosphate group. Two layer ONIOM calculation was performed to acquire structural understanding of phospholylation and reactant, transition state, and product structures were determined with optimization. Our calculation also revealed that the other product which has four terminus ADP oxygen atoms coordinated with Mg2+, was stabler than the product of SN2 like inversion. Also, a few water molecules critical to this reaction is found through molecular dynamics calculation and an explanation for the reaction in which these water molecules participate requires further consideration.
    Download PDF (122K)
  • Keiko Miyamoto, Keiko Takano, Yutaka Fukuda
    Pages JP06
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    Nickel(II) mixed ligand complexes are of interest due to their thermochromic and solvatochromic properties, which will lead to their use as thermosensitive coloring matter, various sensors and indicators, particularly as a Lewis-acid-base indicator. Density functional theory, B3LYP/CEP-31G calculations, were performed on nickel(II) complexes comprising N,N,N',N'-tetramethylethylenediamine (tmen), benzoylacetonate (bzac) and a halide anion (X), Ni(tmen)(bzac)X n(H2O)(n = 1-4, X = Cl, Br, I). Fully optimized geometries of [Ni(bzac)2(tmen)], [NiCl(bzac)(tmen)(H2O)] and [Ni(bzac)(tmen)(H2O)2]+ are all in good agreement with crystallographic data. This series of nickel(II) complexes shows a disproportionation reaction in a non-polar and rather inert solvent, such as 1,2-dichloroethane and acetone, accompanied by highly contrasting color changes of the solutions. Also the presence of an intermediate compound has been confirmed. A possible reaction mechanism of this reaction as well as the structure of the intermediate compound is discussed.
    Download PDF (293K)
  • Naohiro Kobayashi, Hiroaki Kato, Hidetsugu Abe
    Pages JP07
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    We have been accumulating proton NMR spectral data over twenty years. And the database is constructed and various tools that can add, edit, and delete these data are developed. The database is now opened on WWW. In this research, the function for consistency check of the newly added data by the chemical shift prediction using the correlation table and its GUI are developed. Additionally, we will report another function that make it possible to incorporate CI-format data(old exchange format developed in our laboratory)to the present database.
    Download PDF (293K)
  • Nana Kawakami, Hiroaki Kato, Hidetsugu Abe
    Pages JP08
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    We have investigated structure-odor relationships of various organic compounds on the assumption that plural substructures of a compound cause the specific odor of it. For this research, a structure-odor database (OSDB) has been constructed. It contains the chemical structure, several physicochemical parameters, and the odor descriptions of over 3,000 organic compounds. In the present work, the 3D structural information of the compounds are generated by CORINA and stored into the OSDB. Three-dimensional structural features that appear in a particular odor category were found by FragSearch program. Then, we tried to classify compounds into some odor categories by the Support Vector Machine tool using the structural fragments as descriptor. The result suggests that the present approach is successfully classified several odor categories.
    Download PDF (297K)
  • Naofumi Nakayama, Shigeaki Obata, Hitoshi Goto
    Pages JP10
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    Molecular simulations using the conventional force fields have been successful for understanding the nanoscopic behavior of molecules in several chemical environments. However, the classical approaches have some shortcomings due to the electrostatic interaction calculated by fixed atomic charges. To describe the charge reorganization due to nuclear structrural changes of the molecules, recently, we developed the new charge equilibration (NQEq) framework that the parameters are classified to 60 atomic types and optimized to reprduce the experimental dipole moments of several organic compounds. In this study, we presented the new force field, MMFF/NQEq, that the electrostatic energy term evaluated with fixed charges based on the bond-charge increment method in the Merck Molecular Force Field (MMFF) was replaced by the energy expression of NQEq, and that the parameter set of torsion interaction term was refined. MMFF/NQEq can reproduce the relative energies of conformational isomers of the organic compounds obtained by high-level ab initio calculations. We will discuss the accuracy and usefulness of the calculations using MMFF/NQEq.
    Download PDF (146K)
  • Yasuyuki Nishimura, Tamotsu Yoshihiro, Noriko Nishino, Fumitoshi Sato
    Pages JP11
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    We have developed a quantum chemical calculation system for proteins based on the density functional method program ProteinDF. This system is composed of five subsystems: "Automatic computations", "Geometry optimization and ab initio MD calculation (Quantum dynamics)", "Calculations on very large-sized proteins", "Database of protein wavefunctions", and "Integrated environment, ProteinEditor". In this study, we will report the development of ProteinEditor subsystem. In all-electron calculations on proteins, the data size is huge and the simulation becomes complex. Therefore, an interactive and easy control by the user is indispensable. To reduce these problems, ProteinEditor contains a graphical user interface to support the all-electron semi-automatic calculation method for proteins (QCLO method). The high-performance molecular graphics and the structural interaction representation functions can assist the safe quantum chemical calculation. We also implemented the functions of hydrogen addition, the amino acid substitution (point mutation) and checking the structural validity of proteins in PDB format. This work will guarantee the easy and safe achievement of quantum chemical simulation of proteins.
    Download PDF (548K)
  • Yoshihiro Mizukami
    Pages JP12
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    It is well known that frontier orbitals of molecules are important to understand chemical reactivity. We have already published papers on relations between frontier electron densities and toxicities of dioxins and dibenzofurans. In this work, we investigate the relations between frontier electron densities and activities of steroids. Highest occupied molecular orbitals (HOMO) of 31 steroids are calculated by semi-empirical AM1 and ab initio Hartree-Fock molecular orbital method. Electron densities of HOMO of 31 steroids are used as variables for multivariate statistical analysis. 31 steroids are classified according to principal component. There is a correlation between principal component scores of principal components and activity of steroids.
    Download PDF (337K)
  • Noriko Ogawa, Mihoko Ohyama, Miki Yokoyama, Keiko Takano
    Pages JP13
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    Nagatsuka and her coworkers found a new type of glycolipid in the membrane. They isolated the lipid from human cord RBCs and identified it as phosphatidylglucoside (PhGlc). Furthermore, they analyzed the roles of PhGlc and showed that the PhGlc-based microdomain is involved in GL-7 signaling. Their experiments suggest that PhGlc constitutes a previously undescribed lipid signaling domain involved in cellular differentiation of HL 60 cells. However, it is not clear the reason why PhGlcs form functional microdoomains. We expected that the glucose residue of PhGlc is critical for organization of the carbohydrate-dependent signaling domain involved in cellular differentiation. In this study, geometries of the sugar part were focused on and were optimized systematically by semi-empirical PM3 and ab initio RHF MO methods. We further examined stable conformations of the sugar part. It was found that polar hydroxyl groups in the sugar parts faced to the lateral direction in some of the most stable conformers. It suggests that the glucose residue of PhGlc is important in formation of functional lipid domains by hydrogen bonding to glucoses in the neighboring PhGlc.
    Download PDF (470K)
  • Toshiyuki Kamakura, Hitoshi Goto
    Pages JP14
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    For elucidation on the structure and function of protein-ligand complex, it is important to understand both the kinetical and thermodynamical behavior in biological environment. Especially, temperature dependency on their biological activity reveals that the thermal fluctuation plays an important role on the dynamic interaction between protein and ligand. In this study, we propose a new conformational analysis approach to understanding protein-ligand complex functions. Practical mixing of calculated protein-ligand complex by thermodynamic normal mode analysis gives an important technique for finding many plausible structures of protein-ligand complex. Base on this vibrational simulation, the difference of dynamical behaviors between HIV-1 protease and the complex with inhibitor in water will be discussed. As a result, in only protease system, Gly49 fluctuates with neighbor residues. In protease-ligand complexed system and protease-ligand in crystal waters, Gly49 does not fluctuate. Asp25 and Asp25', which are active site, are motionless in any systems. Therefore, I concluded this vibration in this mode is related to activity between protease and inhibitor.
    Download PDF (589K)
  • Takahiro Kosugi, Isao Nakanishi, Kazuo Kitaura
    Pages JP15
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    It is a challenging problem to predict binding free energy between a ligand and a protein with accuracy. If this issue is addressed, precise binding affinity prediction would be considerably helpful for drug design and for understanding protein functions. A lot of methods to predict binding free energy have been reported so far. Here, we propose a new method based on the linear response approximation theory. This method is similar to the linear interaction energy (LIE) method. However, it is completely different in considering internal energy changes of ligand molecules in addition to interaction energies. In the words, it uses all potentials of a ligand molecule. Also, it requires no fitting parameters used by the LIE method. This method was applied to the adenosine deaminase inhibitors and a good correlation between calculated and experimental binding affinities has been obtained. We believe this method would be applicable to all protein systems.
    Download PDF (170K)
  • Noriyuki Mizoguchi
    Pages JP16
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    Monocyclic carbon clusters have two kinds of pi molecular orbitals, out-of-plane and in-plane pi molecular orbitals. By using B3LYP density functional method with 6-31G(d) basis set we optimized monocyclic carbon clusters and studied the numbers of out-of-plane and in-plane pi electrons. It was found from electron-count and geometry index that the carbon rings of size 4m+2 are out-of-plane aromatic and in-plane aromatic (doubly aromatic), those of of size 4m are out-of-plane anti-aromatic and in-plane anti-aromatic (doubly anti-aromatic), and those of size 4m+1 and 4m+3 are out-of-plane aromatic and in-plane anti-aromatic. This result was confirmed also from their NICS values.
    Download PDF (189K)
  • Noriko Nishino, Toshiyuki Hirano, Tamio Nishimura, Sou Koike, Fumitosh ...
    Pages JP17
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    We have developed a density functional(DF) program, ProteinDF for all-electron calculation on proteins. In order to overcome difficulties of a convergence problem in large-scale molecules, we already proposed a new localized orbital (LO), which named Quasi-Canonical LO (QCLO). The previous convergence process using QCLOs paid attention to the primary structure and adopted linear elongation by the arithmetric series. In this study, a new convergence process is proposed by taking the secondary structure into consideration such as hydrogen bonds and salt bridges, and those linkages are calculated in pairs here. The effectiveness of the new process was discussed by comparing with the values of the original method, i.e. the total energy, convergence history and difference of Mulliken populations between initial and converged states. As a result, we proved the new convergence processes gave significantly better initial guess than previous one. This method could be useful for safe and automatic all-electron DF calculations on very large-scale proteins.
    Download PDF (175K)
  • Naoki Ihara, Fumitoshi Sato
    Pages JP18
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    To reveal functions of proteins, it is important to understand dynamical features of proteins. Especially in biocatalyst processes, quantum chemical calculations are indispensable. We developed a Gaussian-based density functional method program, ProteinDF, and achieved the all-electron calculations on proteins. However, the protein dynamics simulation based on all-electron calculation is still far from a practical idea in a present computer resource. To overcome this problem, we also developed a new method to calculate localized molecular orbitals on fragments. We named it quasi-canonical localized orbital (QCLO) method. By using this method, we can perform partial simulations of proteins based on all-electron calculations to divide into the dynamically and statically calculated areas, which greatly decrease the computational task. In this study, we implemented the partial simulation program using the QCLO method. We will discuss the details of the computational method and show the effectiveness of this method.
    Download PDF (76K)
  • Chika Nishimura, Keiko Takano
    Pages JP19
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    It is well known that gold can be dissolved in aqua regia. Recently, reversible dissolution/deposition of gold was reported in organic solvent with excessive iodine. In both reactions, gold exists in the form of gold halides (AuX2-, AuX4-, X = Cl, Br, I). To examine the geometries and electronic structures of gold halides, which is closely related to the dissolution/deposition reactions, MP2, CCSD(T) and B3LYP calculations were carried out using relativistic effective core potentials (RECPs). Optimized geometric parameters of linear triatomic anions of gold halides and polyhalides are compared with the experimental data. There is not much difference in the calculated bond lengths between MP2, CCSD(T) and B3LYP. Atoms at the edge have negative charge in the gold halides as well as in the polyhalides. Pentatomic anions of gold halides and polyhalides had three local minima on the potential energy surface, with V-shaped, T-shaped and square-planar geometries. Studies on electronic structures and reaction mechanisms involving pentatomic anions are in progress.
    Download PDF (24K)
  • Yasuhiro Shigemitsu, Naoko Mizuyama, Yoshinori Tominaga
    Pages JP20
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    The absorption and fluorescent properties of newly synthesized 5-aryl[2,2']bipyridyls were examined both in solution and in the solid state. 6-methoxy derivatives 1a-c have weak fluorescence whereas 6-amino derivatives 2a-f showed strong fluorescence. There seems no notable differences in their structural characteristics indicated by X-ray crystallographic and MO-optimized geometries. In aiming to find rational explanations for the fluorescent differences from the theoretical viewpoints, their spectroscopic properties were computationally analyzed using semi-empirical molecular orbital calculations. The geometries of the ground (S0) and the first singlet excited state (S1) were optimized and employed to estimate the vertical transition energies. The drastic differences in the fluorescent intensities between 1a-c and 2a-f can be qualitatively explained due to the relative energy levels S1 and the neighboring triplet level; the energy gaps supposedly regulate the non-radiative intersystem crossing (ISC) probabilities. On the relaxed S1 geometry, 1a-c have quite small S1-triplet energy gaps whereas 2a-f have no such quasi-degeneracies, leading to their quantitatively different quantum yields.
    Download PDF (262K)
  • Toshiaki Taura, Nakafu Itakura
    Pages JP21
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    Some conformational isomers of amino acids like phenylalanine and aspartic acid play a significant role in the reaction and the function of the amino acid in the biological system. The population of the conformational isomers has been so far determined by a use of infrared spectroscopy and proton magnetic resonance. However, recently, computer chemistry is making great advances according to the progress of personal computers (e.g. windows machine) and the appearance of high performance software comercially available. Now we can use these computers and software for our research. In this work, we optimized structures of a neutral molecule, an anion, a cation and a cobalt(III) complex of phenylalanine using ab-initio calculation and also DFT calculation. Stability and population of each conformational isomer were discussed by estimating the energy value of the optimized structure. The population of the conformational isomers was consistent with that obtained by the experimental method (NMR). Both methods indicated that the isomer having a benzyl group trans to a carboxyl group is the most stable.
    Download PDF (385K)
  • Susumu TSUCHIYA, Takanori ODA, Akio OKAMURA, Hideo SEKIYAMA
    Pages JP22
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    The study of intermolecular interactions is a useful step toward gaining a basic understanding of the dynamic behaviour of molecules in various environments. Especially, interactions of the metylamine (CH3NH2) molecule with other molecules provide valuable information on protein-protein interactions or protein-solvent interactions in biological systems. From this point of view, high-level ab initio calculations [MP2] using correlation consistent basis sets of Dunning [aug-cc-pVTZ] were carried out to study the structures and stability of CH3NH2/H2O, CH3NH2/HF, and other clusters related to CH3NH2 derivatives. For CH3NH2/H2O, several different orientations were studied. From the result of full optimization, the bond angle N(CH3NH2) .H(H2O)-O(H2O) is 162 in the most stable structure of CH3NH2/H2O. Furthermore, for CH3NH2/HF, N(CH3NH2) .H(HF)-F(HF) is linear(180 ). The details of the portential energy surfaces and the effects of the basis set superposion error (BSSE) have been discussed. In the case of other clusters related to CH3NH2 derivatives (especially CH3NH2 dimer, trimer, etc) further investigations have been in progress.
    Download PDF (96K)
  • Hideo SEKIYAMA, Hirokazu HIRAYAMA, Yuusuke TAKAHASHI, Shinnosuke KATO
    Pages JP23
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    Recently, the experimental detection and characterization of Van der Waals trimers is a subject of growing interest. In the present study, The potential energy surface for the Ar/(NH3)2 trimer has been explored by using second-order M?ller-Plesset perturbation theory (MP2). Dunning and coworker's correlated consistent aug-cc-pVTZ basis set was employed for the ab initio calculation. Intermolecular potential energy surfaces were calculated at the several geometries for the Ar-(NH3)2 trimer. Two stable isomers were located and the geometries of the two isomers were fully optimized. In the two optimized structures, the NH3-NH3 interaction play an important role in the Ar/(NH3)2 trimer. The structural parameters of (NH3)2 unit in the Ar/(NH3)2 trimer are very similar to the values in the (NH3)2 dimer. This indicate that the structure of the (NH3)2 dimer is nearly unaffected by the weak interaction with the Ar atom. In addition, the local structure of the Ar atom in the global minimum structure for the Ar/(NH3)2 trimer is almost the same as that of the Ar/NH3 dimer.
    Download PDF (91K)
  • Takahiro Takahashi, Masamoto Arakawa, Kimito Funatsu, Yoshinori Ema
    Pages JP24
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    Chemical Vapor Deposition (CVD) is one of the most important ultra fine manufacturing processes for semiconductor devices. Although process simulators are very helpful for developing CVD processes, the huge calculation costs of the simulators limit their applications to various CVD processes. In our previous work, we showed that using models of the correlation between calculating conditions and calculated results with simulators (that is, modeling the calculation processes of the simulators) drastically decreases the calculation costs and extends the applicable area of the simulators. Therefore, we have developed an automatic modeling system for the calculation processes of a process simulator for CVD. The system consists of a software agent, generalized modeling software, Chemish and the simulator. The agent autonomously makes the models by operating the simulator and the modeling software. The models showed good predictability with the calculated results of the simulator. Because the simulator can be replaced with the models, the models will be used for the rapid research and development of the CVD processes.
    Download PDF (211K)
  • Masakazu Makino
    Pages JP25
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    A quantitative structure and property relationships (QSPR) analysis on gas chromatograph relative retention times (GCRT) of polychlorinated biphenyls(PCBs), and their molecular geometrical and topological parameters was carried out. It was found that there was a different change in GCRT depending on solvent accessible surface area(SAS) between the congeners and the isomers. To elucidate the retention characteristics of the isomers, the GCRT were classified in terms of the number of the pair of proximitive Cl and H substituents bonding to the biphenyl skeleton (Npx-Cl and Npx-H). The correlation coefficients were markedly improved between SAS and GCRT ranging in the same category. Next an analysis on the basis of the connectivity indices, which had been derived by Randic, Kier, and Hall, was performed to clarify the topological characteristics for the proximitive positions. It was found that there was a close correlation between the second-order connectivity descriptor, 2 kai , and GCRT for the isomers. On the other hand, from analyses on the correlation between 2 kai and some electrical descriptors, it was revealed that there was a significant correlation between 2 kai and sum of net charge at H substituents (QH).
    Download PDF (189K)
  • Yuuichi Tokuda, Koichi Takahashi
    Pages JP26
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    Genetic algorithm(GA) has been used widely to inference the globally stable conformation(s) of molecules. GA is essentially a parallel algorithm and has the feature to search not only for the globally stable conformation but also for locally stable states which are similar to the globally stable conformation. In the case of condensed phase such as solutions or solids, molecules are under the interactions with other molecules. As intermolecular interactions are weak compared to the covalent forces of molecules, locally stable conformations in isolated state may still have similar structures in condensed media. It is known that flexible molecules like enkephalin analogs take two or more conformations even in the crystal states. This facts means some of locally stable conformations of isolated flexible molecules may be identical to the conformations in crystal states. To get the variety of stable conformations, we developed the GA to comprise immune system and built immune algorithm(IA). IA suppresses the conformations in high concentration to get the variety of conformations. GA and IA were applied to enkephalin analogs and collagen like peptides. We found (1) IA searched more variety of conformations than GA did, (2) enkephalins gave more conformational variety than collagen peptides.
    Download PDF (538K)
  • Hiroko Satoh, Tomoo Nukada, Katsuhito Akahane, Makoto Satoh
    Pages JP27
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    The goal of our project is to develop a new type of chemical educational systems by using a haptic device to virtually feel attractive and repulsive force in the van der Waals molecular interaction. We focus on the interaction between rare-gas atoms (He, Ne, Ar, Kr, and Xe) and construct a robust basic system. Two visually different systems are developed. One is E-HaptiChem, which uses a 3D-multimedia open-source graphics library, JUN as a general basis and a chemical open-source graphics library, ChemoJun as a chemical tool. Another one uses an AVS graphics library. In the current phase, basic functions are developed and implemented to the both systems. This paper outlines the project, methods, and the first results from the systems.
    Download PDF (387K)
  • Tetsuya Maeshiro, Shin-ichi Nakayama
    Pages JP28
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    Conceptual network in the field of Molecular Biology of Cell was used to analyze how the conceptual network evolves. Conceptual network is generated by extraction of terms and by linking terms according to semantic relationships among terms. Such network represents knowledge of Molecular Biology of Cell, where a node represents a concept and a link indicates two concepts are semantically related. Years of 1989, 1994 and 2001 were selected to analyze the structural change in the conceptual network. Only the hierarchical relationship was used for the analysis. The depth is six for all three instances, and the number of concepts increases for every hierarchical level. However, proportion of concepts belonging to both consecutive years is small than expected, and the number of concepts belonging to the previous year is very high. The analysis suggests that concepts become more general and thus less used with the advance in the field, climbing the hierarchical "mountain" and ultimately disappearing from the conceptual network.
    Download PDF (593K)
  • Shin-ichi Nakayama, Tetsuya Maeshiro
    Pages JP29
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    It is important problem in a protein science to detect active sites on protein structures. Some automatic detection methods has been presented as follows; finding a similar structure to known protein active sites, finding pockets and/or cavities on protein surfaces, and so on. We already proposed a method to detect amino acid residues, which construct a function site, using the similarity in 3-D structure of same function proteins. By the last report we clarified the scope and limits of this method. In this report, we applied Herfindahl index, which generally uses in economics parameter, as an indicator of variety of extracted amino acid residues as function site. The index values are obtained in all pairs of 10 serine proteases. The pair, which extracts reasonable amino acid residues, gives high index score and the other pair, which does not give those residues, gives low one. By comparing the index scores and detection validity, which found in previous study, it is found that the method could be detect the true active site residues in the case of more than 450 in index score.
    Download PDF (253K)
  • Yoshimasa Itoh, Takashi Nakayama, Joji Otaki, Haruhiko Yamamoto
    Pages JP30
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    We developed programs for classifying G-protein coupled receptors (GPCRs) to subfamilies, using several methods. The methods are self-organizing map (SOM), nearest neighbor method, and a feedforward artificial neural network. These programs were trained using GPCR dataset and predicted classification of given GPCR sequences as one of fifteen conventional subfamilies. All of three methods employ the same representation of GPCR sequence, that is, auto-cross covariance derived from z-scales that were computed as principal components of 26 physicochemical properties of amino acids. All of them gave results with high precision for non-aligned GPCR amino acid sequences. Specifically, for test datasets, the nearest neighbor gave the highest precision (99.68%), SOM also gave high precision (97.96%), and the feedforward network gave 92.50%. The output plane of the SOM consists of 50x50 neurons, and learn to respond individually to distinct input patterns so that nearby neurons respond to similar inputs. The feedforward network consists of three layers with 12 hidden layer units and 15 output layer units. Two-fold cross-validation were carried out by selecting 30 training data sets from GPCR dataset. The results suggest that the methods could become useful tool for non-aligned sequence classification.
    Download PDF (382K)
  • Koichi Hiromatsu, Hiromi Yamashita, Tsutomu Nishihara
    Pages JP31
    Published: 2005
    Released on J-STAGE: November 16, 2005
    CONFERENCE PROCEEDINGS FREE ACCESS
    The prediction system of the biodegradation and bioaccumulation of the existing chemical using the structure activity relationships as a part of the existing chemicals for the check of their safety are developed. The prediction system evaluates substructure information automatically from the structural formula inputted on on-line, is classified into similar structure and outputs the prediction value of qualitative biodegradation and bioaccumulation from the flow of the experimental knowledge, or regression analysis. Furthermore, the structure, examination result of biodegradation and bioaccumulation and the prediction result in this system of the training set (the substances used for development of this prediction system) can be referred to. Acquiring the information on a training set is one effective means to evaluate the reliability of a prediction value.
    Download PDF (262K)
feedback
Top