The linear free energy relationships, exemplified by the Hammett and the Broensted equations, are of course empirically based and both assumed to be essentially linear. The Yukawa-Tsuno equation is required to ensure the linearity for the region the original Hammett equation otherwise fails. On the other hand, the establishment of the intrinsic non-linearity in the Broensted relationship has lead to the concept of the selectivity reactivity relationship intimately related to the rate-equilibrium relationship-coordinate shifts of transition state. This elegant concept has now been widely used to the quantitative elucidation of the Hammett-type structure-reactivity correlations in benzyl cationic systems. Non-additivity and non-linearity of substituent effects observed in the polyarylcarbenium systems were elucidated rather elegantly in such a way based on the shifts of transition state coordinate. The most of well-known successful results are criticized, demonstrating no change in transition state position.

抄録全体を表示

A role of quantum chemistry for the information chemistry in 21 century and a future aspect of quantum chemistry are discussed. Chemists in 21 century will be drowned in the inundation brought by the exponential increase in the amounts of chemical informations, such as structures and properties of compounds, chemical reaction paths and the mechanisms etc.. Therefore, we must ask to computational chemistry to make useful and convenient data base of enormous amounts of chemical informations. Quantum chemistry will provide useful models, concepts, principles for understanding molecule, molecular interaction, chemical reaction mechanism and will greatly contribute to the evolution chemical informationology. Brief history of japanese quantum chemistry is also presented.

抄録全体を表示

This paper deals with the action of SF6847, which shows scientific fiction-like very potent uncoupling activity of oxidative phosphorylation in mitochondria. SF6847 was found to act as a catalyst due to its protonophoric activity, supporting the establishment of chemi-osomotic energy transduction theory of oxidative phosphorylation. Structural basis of the efficient protonophoric action of SF6847 is discussed.

抄録全体を表示

Personal history of my 30 year research on structure-activity correlation is introduced. The topological index, which was proposed to be defined and coined by myself in 1971, is now accepted as a common terminology in the areas of chemical application of the graph theory and quantitative structure-activity correlation (QSAR) study. This index Z is defined by using the non-adjacent number for the carbon atom skeleton of a saturated hydrocarbon molecule, and has been shown to be well correlated with the boiling point. The first paper of my research of this thema was published in Bulletin of the Chemical Society of Japan in 1977, after several controversial troubles because of the novelty of this work. The original idea came from the QSAR attempt in the author's school age on the correlation between the boiling point and structure of octane isomers. During the 30 years many scientists, abroad and demestic, have influenced and perturbed on the continuation of my research. Many anecdotes of these people are introduced.

抄録全体を表示

We developed a neural network simulator for structure-activity correlation of molecules: Neco. A self-organized network model for high-speed learning was included in Neco, a perceptron type with the three layers. In the hidden layer the neurons is self-organized by using Mahalanobis generalized distance. Using this simulator, hydrophobic parameter, logP, of perillartine derivatives was predicted. We used for inputs the set of six parameters: five STERIMOL (L, Wl, Wu, Wr, and Wd) parameters and the sweet/bitter activity. The 22 sampled data is used for training. Our neural network accurately can predicted hydrophobic parameter, logP. Compared with a normal perceptron network, the learning ability of our network is quite higher and its convergence speed is greatly larger than it.

抄録全体を表示

It becomes difficult to find all of the possible structures of the water cluster when the number of water molecules increases. In this report we construct a systematic way of finding all the possible structures using graph theory. In a water cluster water molecules are held together by hydrogen bonding and taken into the hydrogen bond network. Topological patterns of hydrogen bonds in the water clusters correspond to the directed graphs. We counted up directed graphs which correspond to the topological patterns of hydrogen bonds for (H2O)n,n=3-8. Then, for each topological pattern (n=3-5) we examined whether a structure within the pattern corresponds to a local minimum on the potential energy surface with the level of HF/6-31G*. For (H2O)3, 3 topological patterns were found which correspond to stable structures, and 5 topological patterns for (H2O)4, and 22 for (H2O)5, respectively. We carried out ab intio MD calculations (HF/6-31G*) and examined how the water cluster changes its structure from one stable structure to another.

抄録全体を表示

In the methodologies of 3D-QSAR such as Comparative Molecule Field Analysis (CoMFA), appropriate molecular alignment is definitely required for predictive data modeling and correct analysis of the model. In this study, the novel method for molecular alignment using Hopfield neural networks is proposed. A Hopfield neural network is one of artificial neural networks that developed by Hopfield and Tank in 1985, and it has been mainly used as Content-Addressable Memory. Recently, ability of Hopfield neural network to perform combinatorial optimization was reported by Feuilleaubois et al. and several applications were presented. In our method, each molecule is represented by four chemical properties (Hydrophobic, Hydrogen-bonding donor, Hydrogen-bonding acceptor and Hydrogen-bonding donor/acceptor), and then the properties among molecules are corresponded by HNN. In order to validate usefulness of this method, 3D-QSAR analyses of cyclooxygenase-2 (COX-2) inhibitors were performed. All compounds in the data set were superposed by our method, and then QSAR model using CoMFA field variables was calculated by Partial Least Squares (PLS) method. As a result, the robust regression model (R^2=0.93, Q^2=0.70) with five components was obtained by leave-one-out procedure, and it could give meaningful counter map.

抄録全体を表示

The Predictive Toxicology Challenge (PTC) workshop was held in ECML/PKDD conferences this summer. The Challenge was to obtain models that predict the outcome of biological tests for the carcinogenicity of chemicals using information related to chemical structure only. The learning data comes from NTP results, containing 417 compounds for rodent carcinogenicity. The test data consists of 185 compounds provided by FDA. The 7 descriptor sets are submitted for these compounds. 14 groups of researchers proposed more than 26 models to predict the carcinogenicity of test compounds. Their submissions were judged by ROC analysis and by how "toxicologist-friendly" each model was presented. In this paper, we first introduce the PTC workshop outline and gives the overview of the submitted models. The results show the usefulness of various data mining methodology for the analysis of structure activity relationships. The author was one of the submitters and our model was rated as the best. In the latter half of this paper, we show the results derived by the cascade model in our lab. It uses linear fragments and 7 physicochemical descriptors. The derived rules have been able to give us useful insights to the SAR of rodent carcinogenicity.

抄録全体を表示

Congenital malformations or birth defects are a major public concern. However, lack of proper information may panic a pregnant woman who had taken medicines, and thus may result in an unnecessary abortion. During the development of new drugs, the evaluation of reproductive risk using experimental animals is required in preclinical trials. However, the reproductive data of medicinal products have been usually held privileged by pharmaceutical companies once the development of a product has been stopped. We have started a new research / development project in Japan to create an information-community for teratogenic agents, consisting of pregnant women, pharmacists, physicians, research scientists of new drugs and the Japan Pharmaceutical Information Center, a non-profit foundation. Our system will be useful for medical workers who perform the clinical risk-assessment of teratogenic exposures during pregnancy. Thus the pregnant woman and her family will be able to carry out family planning with their minds at ease. In addition, it will be beneficial for the scientists to develop new drugs using the quantitative structure activity relationship based on chemical information and computer science.

抄録全体を表示

Of the macrocyclic benzylisoquinoline alkaloids, cepharanthine and isotetrandrine show the radical scavenging activities, but tetrandrine does not. For understanding the structural requirements for radical trapping of benzylisoquinoline alkaloids, we analyzed conformational flexibilities at the global minimum structures using modified CONFLEX3 algorithm. We found that cepharanthine and isotetrandrine take the flexible cis-configuration, whereas tetrandrine takes the rigid trans-configuration. Hence, it is concluded that flexible structure is responsible for radical scavenging activity of the alkaloids.

抄録全体を表示

The learning system has been exposed to a big change. The systems have changed to the support tool of the knowledge acquisition from the teaching machine of knowledge transmission. The idea of situated learning spread from cognitive science in the '80s. Knowledge is considered to be situation instead of existence. Then, the purpose of learning systems became to support of knowledge construction rather than transmitted knowledge. And the importance of group learning is emphasized. Then, CSCL (Computer Supported Collaborative Learning) systems were developed. In CSCL, new technologies are used and advanced learning environments are improved. The followings can be considered in computer use in chemistry education: Transmission of knowledge, Training, and Support of experience. Transmission of knowledge is an electronic textbook from an affirmative side. Training is an introduction of using software. A novice cannot master the advanced tool of chemistry. Support of experience is a simulation. It is important to make the minute world visible. Learning by experience is important. CSCL can use these methods together. For example, we are developing CSCL with the simulation type. And CSCL that supports an experiment is described briefly. The system supports the process of an experiment planning.

抄録全体を表示

With rapid IT innovation, everyone receives the latest information easily in an instant, or the environment where complicated data processing and theoretical calculation can be carried out easily has been ready. However, the practical use is still limited. From the position of the performer belonging to the research and development division of a chemical company, the subject actualized in the trial of practical use of Chemical Information and Computational Chemistry is arranged, and the idea towards effective use is discussed.

抄録全体を表示

It has been possible to practically use computer-assisted synthesis design systems such as AIPHOS and TOSP in order to generate synthetic routes of compounds. However, these systems produce many routes without giving information on which route should be applied first in experiments. It is common knowledge that transition states are keys for how to estimate easiness of chemical reactions and computational chemistry delivers that answer. We are developing a new system which makes it possible to rank the routes easily by using a data base. The data base system, called "Data Base for Transition States, TSDB" combines computational and information chemist ry. In the present paper, we show how to evaluate possibility of synthesis routes from TOSP by use of density functional calculations and to use TSDB for ranking synthesis.

抄録全体を表示

As rapid advancement in processing capability of computers, complicated molecule chemistry calculation chemical calculation is becoming possible to a chemistry informatics. The research on the system which carried out the 3-dimensional solid display of the multi dimensional structure feature with various energy information such as electric potential, and on the system which analysis the feature of the substance with complicated molecular structure is, advanced. On the other hand, in an educational system field, the usage of a computer changes from the stand-alone to the networking at a rapid pace of spreading of networks.Furthermore, it also changes from individual learning to collaborative learning.In collaborative learning , interaction among individuals is emphasized. This paper described the educational system which visualized the multi dimensional structure feature which has been difficult with 3-dimensional solid display. We have proposed the support system to make the learners to understand interactivity molecular structure.

抄録全体を表示

An example of one-semester course for the introduction to information chemistry is presented for the junior students in a bioscience- or biotechnology-related department, together with their curricular background. The course consists of information search for scientific documents and molecular computation. The former includes introduction of the structure and function of abstract databases for scientific documents. An exercise on subject search is afforded with Chemical Abstracts and an instruction on an electronic search is made with Enjoy JOIS. The latter is focused on determination of molecular structures with molecular mechanics and molecular orbital calculation, with Hueckel MO method as an example. Theoretical background of each calculation method is briefly introduced and practical study is also afforded. Planned expansion of the covering area of the course in the future is shown and problems underlying is discussed in relation to the intellectual background of students in bioscience/biotechnology departments.

抄録全体を表示

It is possible to do the reuse of the old BASIC program by Excel. It seems that VBA of Excel can not be used, when you are not versed in the Visual Basic grammar. But, the old N88BASIC grammar is very available in VBA of Excel. It is possible to teach the ingenuity in the Basic programming in which the creative idea in the debug and error countermeasure is cultivated. Within a given function in IT (information and technology), the creative idea can not be cultivated in a computing process.

抄録全体を表示

Isosurface has been used for the display of hydrogen atomic orbital. Isosurface is a surface build by connecting points having a same function value. This method is well suited for the representation of the shape of an atomic orbital, however, it is necessary to use cross sectional contour plot for the presentation of the map of various functional values in an atomic orbital. In this study, we used CAVE to represent shape and function map simultaneously, which is impossible by a conventional plane display system. Isosurfaces having various function values were calculated for an atomic orbital and projected over four 3*3m screens of a CAVE. Isosurfaces are displayed by a half transparent technique so that inner surfaces are to be recognized. As the viewer moves in the CAVE, the system is possible to move on our viewpoints in 4 screens is realize virtual reality in an atomic orbital. We termed the alone displaying method as multi isosurface display.

抄録全体を表示

The inorganic crystal structure database (ICSD) includes about 60,000 crystal data and exports crystallographic information files (CIF) of text format. However, the data of CIF cannot be directly available in C++ programming, because CIF offers only parameters of atomic coordinates and equivalent coordinates of space group. Therefore, we implemented a base class that could export lattice parameters and full atomic coordinates. Although VRML or OpenGL are generally used for a development of 3D graphics software, any nodes or functions that can directly draw a convex polyhedron are not offered. A convex polyhedron should be drawn as an aggregate of surface polygons. In addition, vertex coordinates of the polygons should be sorted in such a way that each surface polygon is drawn with one stroke of the brash. We implemented another base class that could export an appropriate array of vertex coordinates of surface polygons in order to draw a polyhedron in the C++ programming. Arguments of a constructor of this base class are only vertex coordinates of a polyhedron. Our software, which could visualize complicate inorganic crystal structures by way of drawing coordination polyhedrons, was developed without ill effects by using these two base classes.

抄録全体を表示

A chemical equation itself is not going to tell us the reason why a particular reaction proceeds. It is our aim to produce a teaching material to make students marvel at chemical reactions and experience the joy of discovering the why of these reactions. The CG animations of chemical reactions were produced based on quantum chemical calculations. We call these animations Computer Microscope. The QT-VR was used to allow the student to rotate molecules of reactant, transition state, and product. This feature helps students to have the sense of touching the molecule as well as the case of the molecular model. The movies of the rotating HOMO and LUMO combinations help the student to understand what plays an important role in bond formation and scission. The Computer Microscope is an effective tool for better understanding how and why chemical reactions occur.

抄録全体を表示

Multivariate Analysis Modeling like Multiple Regression Analysis has been used for the validation of hypothesis or discovery of new hypothesis. But it is not necessarily a good modeling tool for chemical process because of its strong nonlinearity. Applying Alternate Conditional Expectation to these processes instead of the ordinary least square method sometimes gives informative results to the practitioners. We will discuss two applications : Reactor simulation data to show the nonlinear modeling power of chemical processes and Laboratory data analysis for practical application. This method is a simple extension of ordinary Multiple Regression and practitioners can understand the idea of it. Various applications especially chemical reaction and polymer have been examined.

抄録全体を表示

The polymer database system named "PoLyInfo" has been developed on the basis of a new data model which was constructed by adopting object-oriented analysis to polymer information. In the database, various kinds of data, such as properties, structures, polymerization methods, and monomers, are included and properly associated with each other. Search conditions can be specified from both polymer and monomer. In polymer search, there are two kinds of search options, "Basic search" and "Advanced search". In "Basic search" users can search from popular search items such as polymer name and properties. "Advanced search" is for skilled users and they can specify more detailed conditions. Furthermore, it is also possible to search data with chemical structure of polymer. Users can search the database by selecting atomic groups listed on the template or by drawing a query structure. The function called "Easy browse" is also prepared, where user can get more detailed information by clicking data listing in tables or raphs without setting any search conditions. Data will be continued to update and property prediction module is going to be added to the database. (URL http://triton.tokyo.jst.go.jp /index-e.html)

抄録全体を表示

We have a long-term activities on the NMR spectral database system (SDBS-NMR) in a main-frame computer, which was finished at March 1999. The internet service of SDBS started from 1997 which achieved a great many accesses from the world. To continue the database activities, the database system on a PC was necessary, which was developed by using ISIS/BASE as an input tool. The chemical structures with spectral assignments are treated by molfiles. The spectral patterns displayed by a Java applet can be expanded easily. The DBMS is an ORACLE and since the present system has been developed on a web system in a PC, it is easily used as an intranet system.

抄録全体を表示

This study tries to understand progress and trends of the research in chemical information and computer science in Japan, through a bibliometric analysis of the articles published at the first (1978) to 23rd (2000) Symposia on Chemical Information and Computer Science. The author indexed 539 articles in 11 of the 23 Symposia using 224 index terms selected from Chemical Abstracts General Subject Index Terms. These 224 terms were classified into 15 subject categories. This indexing and classifying unables us to know tendencies and transitions of the subjects having interested Japanese reseachers in this field. Next, the author sorted every author affiliation data from all 1130 articles published until the 23rd Symposium. Those data were used to find the leading Japanese organizations in this field, contribution by academic/governmental/private sectors, the patterns of collaboration and the correlation between the organizations and research subjects.

抄録全体を表示

A new representation method of chemical reactions ReacSim (Reaction Similarity) vector is described. The ReacSim vector has been developed for systematic classification of chemical reaction, about which we are studying for organizing chemical reactions and systematic derivation of knowledge of reactions. The ReacSim vector is defined as highest deviation of physico-chemical parameters from reactants and products. The ReacSim vectors are independent of reaction sites, namely, independent on the number of atoms and bonds of the reaction sites. This benefit of the ReacSim vector makes it possible to classify variety of chemical reactions, generally. 131 reactions that are represented by the ReacSim vectors are classified by a self-organizing neural network method Kohonen neural network. Good correlation between the ReacSim vectors and types of reactions is shown by the classification studies.

抄録全体を表示

We have developed a peak identification system for topological fragment spectra (TFS). The TFS is a kind of quantitative representation of a chemical structure, which is based on enumeration and numerical characterization of all the possible substructures from a chemical structure. It is possible to obtain some variants of the TFS when different weighting methods are applied to the characterization of structural fragments generated. The purpose of the present system is in pursuing the meaning of each peak of TFS and to identify the substructures behind the peaks. The user can identify any peak by specify the peak of TFS on the screen interactively. Then the substructural fragments corresponded to the peak can be displayed with the basic peak information. Moreover, this system can be applied not only for a single structure but also the database that involves a large number of chemical structures. For the latter case, a synthetic spectrum obtained by two or more structures can be generated, and it is possible to identify any peak selected for them. The function of browsing the chemical structures of the database was also implemented for the user who wants to know what chemical structures were included in the database. The function allows the user to know the structures originated for generating the synthetic peak(s). The detail of the system will be described with a couple of illustrative examples.

抄録全体を表示

By using standard coals, we studied simple interpretation methods of infrared and near infrared spectra. In case of interpretation on taken spectra, we would like to know clear methods about how paying attention to where of spectra, what [ how many ] being known, quantitatively or qualitatively. For such purposes, we applied CART(Classifications and Regression trees) to find clear rules. Consequently, it turned out that some of the proximate and ultimate values of coals could be presumed using a comparatively easy rule. For example, although we found that the VM(volatile matter) was closely related to the moisture of coals, we might as well as reconsider the methods for both the examination methods. In this paper, we propse the simple interpretation method of spectra of coals.

抄録全体を表示

We have investigated the audification of molecular information. The musical note information consists of a tone, a velocity and a type of a musical note. The sequence of this musical note information forms a musical score. In the present work, we have developed a computer program for the audification of gene. The DNA nucleotide sequence of a gene was employed for the audification. The program assigns the musical note information to each nucleotide using the transformation algorithm and then outputs the MIDI sequence as the musical score for a MIDI device. Several transformation algorithms were investigated as follows: (1) the simple mapping method that assigns notes to the DNA bases. (2) The unique mapping method that assigns the note information with the base usage on the sequence. (3) The musical-theory-based mapping method that is based on the musical theory to get a better musical score. For the third method, mapping the chord to the codon to form the bass line was employed too. It was observed that it is possible to get some advanced transformation from the DNA sequence to the musical score using the information derived from the base array.

抄録全体を表示

A versatile method for communicating and displaying chemical structural formulas in the Internet has been investigated on the basis of XyMML (XyM Markup Language), where a new DTD (Document Type Definition) for defining XyMML and an XSLT (XSL Transformations) for transforming XyMML (contained in XML (Extensible Markup Language)) into XyM Notation (contained in HTML (Hypertext Markup Language)) have been developed.

抄録全体を表示

With the progress of genome projects, a vast amount of nucleotide sequence data is now available, which makes it possible to study the species-specific gene expression systems involving transcription and translation for a wide range of organisms. To understand species-specific translation system, it is important for comprehending the whole set of tRNAs for individual species. We have been developing a procedure for detecting tRNA genes from DNA sequences. The procedure consists of 2 stages. The first stage finds out cloverleaf secondary structures. The second stage tries to reduce the number of the candidates of the tRNA genes by various restrictions. Some of the tRNA genes in Archaea and Eukaryotes have introns. In the present study, we improve the procedure to find them out. We applied the improved procedure to the complete genome sequence of Aeropyrum pernix (GenBank Accession BA000002) which has tRNA genes with introns. We assumed that the lengths of introns are 8--200 bases, and found 53352 candidates of the tRNA genes in the first stage. In the second stage, we obtained 49 candidates, 17 of which had introns.

抄録全体を表示

We confirmed many introns of human mRNA other than gt//ag (namely, those whose D- and A- ends are gt and ag, respectively), and in order to systematically treat them, we tried to expand the GTAG rule to cover the introns of g//g containing gc//ag (19 cases), gt//gg (11), gt//cg (4), and gg//ag (3) as well as gt//ag (> 5600). Some introns of other kinds, sucu as [ga//ac] also are reported. We found parts of necessary conditions that a new GG rule requires, as follows: (i) A necessity of the second bases at the D-end is carbonyl group, especially on the pyrimidine ring, and especially at the 2-position, (ii) A second base at the A-end requires an amino group especially on the purine ring, and the more carbonyl groups, the more disadvantageous, and (iii) The bases at A-ends appear to be a little more effective than D-ends. Existance of the third kind of splicing pathway is suggested.

抄録全体を表示

We have developed a new computer system CAST/CNMR for accurate13C-NMR chemical shift prediction with considering stereochemical information by using of CAST coding method. CAST is a new canonical coding method for representation of 3D structures. Conformational CAST and configurational CAST are available in CAST/CNMR system, together with CANOST which is a coding method for planar structures. Ring-size information is also available in the system. Most of 13C-NMR chemical shift values are predicted with high accuracy by considering configurational information and solvent effects. Applications of CAST/CNMR to diastereomeric sesterterpenoids are also presented.

抄録全体を表示

We present a new detection method of amino acid residues, which play the role of the protein function. The principle of the method is that proteins with the same function have similar amino acid residues at the similar three dimensional positions. Thus we assume that maximal amino acid residues existing the similar three dimensional positions of proteins are their function site residues. The method first constructs the graph describing relations of three dimensional positions of all possible pairs of amino acid residues in two proteins and then extracts its maximal complete subgraph, which means the maximal amino acid residues at the similar three dimensional positions by the algorithm of Carraghan. The method was tested using electron transport proteins; azurin and plastocyanin, and was obtained 6 reasonable amino acid residues as the function site residues of those proteins. Also the method was applied to acid proteases; porcine pepsin and Rhizopus chinensis acid protease. In this investigation, we restricted the formation of amino acid pairs only the same ones. The result gave 11 amino acid residues, consisting some active site residues, as the function residues of acid proteases. Those results indicate the effectiveness of this new method.

抄録全体を表示

In this study, the integrated system of general purpose molecular dynamics simulators for large scale system of complicated materials such as biopolymer, organic crystal, liquid crystal, polymer, semiconductor, metal and alloy and ceramics are developed. This system has the plug-in software technology for molecular dynamics simulators and analysis programs based on the new theories. Also this system has the software technologies of database control, multiscale high performance graphics, material modeling and graphical user interface for the Beginner user of molecular dynamics simulation. In this presentation, the conceptual design of large-scale general purpose molecular dynamics simulators and integration system for the simulators are described.

抄録全体を表示

The TACPACK2000 system consists of general purpose molecular dynamics simulators for large scale system of complicated materials such as biopolymer, organic crystal, liquid crystal, polymers, semiconductor, metal and alloy and ceramics and integration system for the simulators. In this presentation, the systematization technologies of the integration system are described. In the database management technology, we have developed the new technology using XML(eXtensible Markup Language) to control uniformly the data composed of molecular structure, crystal structure , potential and pseudopotential function etc. Also in the plug-in function, we have developed the software technologies for program restoration / deletion function to this system, compilation linkage functions of the program and automatic generation function of parameters input screen.

抄録全体を表示

We proposed the dual translation (DT) method for the canonical ensemble Monte Carlo (MC) simulation. In this method, two types of particle translation are randomly used in a fixed proportion on the MC process. Although very simple, the method permits the study of molecular configuration at whole area of system consist of liquid/solid, vapor, and their interface regardless of extreme localization of molecules. The method was applied to a plane surface of Kr at the boiling point. Two parameters in Lennard - Jones potential were determined by pilot simulations using the literature data of liquid density and atmospheric pressure. Resulting equilibrium pressures were similar, even if the initial configuration of molecules was drastically changed, showing solidness of the method. Heat of vaporization and surface tension were estimated using the Clapeyron - Clausius equation and the Kelvin equation respectively, and good agreements were found with literature data. It is confirmed for an extremely localized molecular system, that an equilibrium configuration is satisfactorily estimated by the DTMC simulation. It can be utilized for a gas adsorption system instead of conventional GC (grand canonical) MC.

抄録全体を表示

DNA-binding proteins recognize specific base sequences within DNA. To examine the origin of the specificity, we calculated the free energy maps of the base pair and amino acid side chain interactions, as well as the enthalpy change, entropy change and minimum energy maps. These maps are obtained from taking statistic average of the interaction energies. Total number of conformations to obtain these maps for each pair of Asn-AT, Asn-GC, Ser-AT and Ser-GC is 1,556,815,260. For each of the pairs, we found the region around the base pair where the amino acid side chain can form a profound interaction, and it coincides with the tendency found in PDB data. The minimum energy structures obtained from the maps were also in agreement with the interaction patterns found in PDB data.

抄録全体を表示

Silica glasses have anomalous behaviors on mechanical properties. Especially they show several complex deformation patterns upon compression. It is not clear how their structures are transformed without changing their coordination number. Structures of silica glasses upon compression are calculated using the molecular dynamics techniques. Then the calculated structures are analyzed in terms of coordination number, ring size and planarity index. The planarity index expresses the degree of planarity of Si-O rings. It is found that structural changes can be explained in terms of planarity index.

抄録全体を表示

The energies of the conformers of 1,2-dimethoxyethane (DME) in the gas phase are calculated with several levels of ab initio MO calculations. It is found that the energies of the TGT form and the TGG form relative to the most stable isomer (TTT) are 1.1 and 2.9 kcal/mol with the level of HF/6-31G*. They are 0.1 and 1.2 kcal/mol, respectively, with MP2/aug-cc-pVDZ. The energy profile from TTT to TGT is calculated with HF/6-31G* and it is found that the energy barrier is 3.8 kcal/mol. The energy profile from TGT to TGG is calculated with HF/6-31G* and it is found that the energy barrier is 2.5 kcal/mol. The conformations in aqueous solution are investigated with QM/MM method, with HF/6-31G* as QM for DME and TIP3P for the MM water molecules. We take 93 MM water molecules into consideration in the current study. The MD simulation with QM/MM starting from the optimized structure of the TTT form in 93 MM water molecules is performed at 300K, and it is observed that the TTT form changes to the TGT form.

抄録全体を表示

Recently we have developed several enediyne model compounds which posses DNA damaging abilities. In the course of our studies, we often encountered the loss of the DNA damaging abilities of these models in buffer solutions, while the cycloaromatization reaction of them were proceeded via biradical pathway in non-polar media. To overcom this problem, we designed enediyne model compounds which have an electron-withdrawing group on the benzylic radical forming centers (p-radical center).These models cycloaromatized to gave products derived from the biradical, not the zwitterion, in both polar and non-polar media and showed potent DNA cleaving ability. Intuitively, the effect of introducing electron-withdrawing groups may be rationalized that a p-acceptor group destabilizes the benzylic cation to diminish the contribution of the zwitterionic mesomer and therefore enhance the radical character. In this paper, we report the theoretical aspects of this type of bieadicals in detail.

抄録全体を表示

Regioselective photocycloadditions of conjugated cyclic enones are very effective for synthetic applications, but understanding of the selectivity control in the excited states is not clear so far. Contribution of the back-reactions may be not disregarded. 2-Pyridones(1), one of heterocyclic dienones mainly give regio-, site- and stereoselective endo-3,4-[2+2]cycloadducts(3d) and novel azocinones(5) from the excited singlet state, and exo-5,6-[2+2]cycloadducts(4x) from the triplet state by photoreactions with proper electron-dificient alkenes, such as methyl acrylate(2a) or acrylonitrile(2b). In this paper, we quantitatively propose two types of concerted-like two-step reaction mechanisms to products 3d and 5, and also biradical mechanism to product 4x, by performing the transition state(TS) analysis using MOPAC. Each TS for back-reactions of 3d, 4x, and 5 to the reactants 1 and 2 was also estimated.

抄録全体を表示

The concept of aromaticity, usually associated with Huckel's 4n + 2 rule, has proven to be extremely fruitfuiil in organic chemistry. Schleyer and co-workers proposed the nucleus-independent chemical shifts (NICS) for the index of aromaticity. The values of NICS indicate the nature of aromaticity and/or anti-aromaticity for compounds. However, the values of NICS are not sensitive to the same nature of aromaticity. We defined the aromaticity property from a different point of view. The two structures, Kekule's (D3h symmetry) and aromatic (D6h symmetry) structures, were analyzed by a CiLC ( CI/LMO/CASSCF) method. The CiLC analysis showed the alternation of C-C bonds for Kekule's structure and the same electronic states of six C-C bonds for aromatic structure. Especially for aromatic structure, the weights of singlet coupling term are same to those of polarization (ionic) term for each C-C bond. These hypothesis for aromaticity were applied for other compounds.

抄録全体を表示

We (F.S., T.Y., H.K) have developed a density functional method program, ProteinDF, and carried out an all-electron calculation on a metalloprotein. In this article, we will report the method and its integrated system that help us achieve all-electron calculations on proteins automatically.

抄録全体を表示

We have developed a molecular orbital (MO) program ProteinDF based on the density functional theory with Gaussian functions for all-electron calculations on proteins. By using it, we recently performed an all-electron calculation on a typical hemoprotein, cytochrome c. The program was coded by the object-oriented language C++ to construct the complex MO program for calculations on proteins with over thousand orbitals. This program was parallelized for distributed parallel processing systems such as workstation clusters. In this study, we report parallelization methods by using the object-oriented technology in ProteinDF and their efficiency.

抄録全体を表示

The improvement of the Research and Development (R&D) for material formation processes both in the speed, and the cost is strongly demanded in the semiconductor industries. The automation of the R&D will be the candidate for it. However, intellectual works, for example, analyzing the experimental data, and modeling the processes, are formidable obstacles for the automation. In this paper, we showed the novel system, which automatically infers the reaction models from the experimental data using Genetic Algorithms (GA). In the system, the reaction models were presented by the hierarchical chromosomes, and determined chemical-kinetically using the experimental data. We applied the system to the Tetra EthOxy Silane (TEOS) thermal Chemical Vapor Deposition (CVD) processes in order to investigate the potential of the system. The system presented the simple models, which successfully reproduced the experimental data. In addition, the proposed models were in good agreement with the ones, which were manually estimated using another authorized procedure. Therefore, we concluded that the system has enough ability to contribute to the automation of the R&D for the material formation processes.

抄録全体を表示

In order to internationally facilitate useful and effective debates on biopolymers by reducing indistinguishability between different reports of an indentical phenomenon and of different materials, we have designed a database and constructed a prototype system. The database is design to treat any data as if it is an independent journal article by giving a registry number. A data is a question, an opinion, or a comment on a journal article on biopolymer. It may be a counterargument on one of the registered opinions. We have collected about 200 cases of relevant data on coding redions, intron nucleotide sequences, tRNA sequences which seems not to form cloverleaf structures. As one of the byproducts, we found a cluster of putative tRNA genes in a ceratin Gram-positive bacterium. The database will be open near future.

抄録全体を表示

In order to systematically detect candidates of tRNA genes from a bacterial genomic DNA, we tried to establish a set of consensus nucleotides for every 61 anticodons. In addition to purine (r) and pyrimidine (y) for couples of nucleotides consisting of a & g, and t & c, respectively, we uses k(t & g), m(a & c), s(c & g), and w(a & t). We analyzed 30 bacteria whose complete genomes have been known. The result of our analysis shows that there seems not to be a comprehensive set of consensus nucleotides, but some patterns of such sets. Using such patterns may be useful for effective detection of cloverleaf structures of tRNA genes. Classification may be better with such patterns than with anticodon.

抄録全体を表示