Kimito Funatsu

MassBank: a public repository for sharing mass spectral data for life sciences.

MassBank is the first public repository of mass spectra of small chemical compounds for life sciences (<3000 Da). The database contains 605 electron-ionization mass spectrometry (EI-MS), 137 fast atom bombardment MS and 9276 electrospray ionization (ESI)-MS(n) data of 2337 authentic compounds of metabolites, 11 545 EI-MS and 834 other-MS data of 10,286 volatile natural and synthetic compounds, and 3045 ESI-MS(2) data of 679 synthetic drugs contributed by 16 research groups (January 2010). ESI-MS(2) data were analyzed under nonstandardized, independent experimental conditions. MassBank is a distributed database. Each research group provides data from its own MassBank data servers distributed on the Internet. MassBank users can access either all of the MassBank data or a subset of the data by specifying one or more experimental conditions. In a spectral search to retrieve mass spectra similar to a query mass spectrum, the similarity score is calculated by a weighted cosine correlation in which weighting exponents on peak intensity and the mass-to-charge ratio are optimized to the ESI-MS(2) data. MassBank also provides a merged spectrum for each compound prepared by merging the analyzed ESI-MS(2) data on an identical compound under different collision-induced dissociation conditions. Data merging has significantly improved the precision of the identification of a chemical compound by 21-23% at a similarity score of 0.6. Thus, MassBank is useful for the identification of chemical compounds and the publication of experimental data.

GA Strategy for Variable Selection in QSAR Studies: GA-Based PLS Analysis of Calcium Channel Antagonists⊥

The GAPLS (GA based PLS) program has been developed for variable selection in QSAR studies. The modified GA was employed to obtain a PLS model with high internal predictivity using a small number of variables. In order to show the performance of GAPLS for variable selection, the program was applied to the inhibitor activity of calcium channel antagonists. As a result, variables largely contributing to the inhibitory activity could be selected, and the structural requirements for the inhibitory activity could be estimated in an effective manner.

Genetic algorithm-based wavelength selection method for spectral calibration

In this paper, we propose a genetic algorithm‐based wavelength selection (GAWLS) method for visible and near‐infrared (Vis/NIR) spectral calibration. The objective of GAWLS is to construct robust and predictive regression models by selecting informative wavelength regions. To demonstrate the ability of the proposed method, regression models for soil properties and sugar content of apples are constructed by using GAWLS and other variable selection methods. Copyright

GA strategy for variable selection in QSAR studies: application of GA-based region selection to a 3D-QSAR study of acetylcholinesterase inhibitors.

Comparative molecular field analysis (CoMFA) with partial least squares (PLS) is one of the most frequently used tools in three-dimensional quantitative structure-activity relationships (3D-QSAR) studies. Although many successful CoMFA applications have proved the value of this approach, there are some problems in its proper application. Especially, the inability of PLS to handle the low signal-to-noise ratio (sample-to-variable ratio) has attracted much attention from QSAR researchers as an exciting research target, and several variable selection methods have been proposed. More recently, we have developed a novel variable selection method for CoMFA modeling (GARGS: genetic algorithm-based region selection), and its utility has been demonstrated in the previous paper (Kimura, T., et al. J. Chem. Inf. Comput. Sci. 1998, 38, 276-282). The purpose of this study is to evaluate whether GARGS can pinpoint known molecular interactions in 3D space. We have used a published set of acetylcholinesterase (AChE) inhibitors as a test example. By applying GARGS to a data set of AChE inhibitors, several improved models with high internal prediction and low number of field variables were obtained. External validation was performed to select a final model among them. The coefficient contour maps of the final GARGS model were compared with the properties of the active site in AChE and the consistency between them was evaluated.

GA Strategy for Variable Selection in QSAR Studies: Enhancement of Comparative Molecular Binding Energy Analysis by GA‐Based PLS Method

Comparative molecular binding energy (COMBINE) is a novel approach for estimation of binding affinity in structure-based drug design (SBDD). COMBINE involves an extensive partitioning of binding interaction energy and multivariate regression analysis to derive a model. In COMBINE, partial least squares (PLS) is especially used as a statistical method. Although PLS is robust and stable, it has been shown that its predictive performance drops with the increase of number of variables. Also, from a practical point of view, model becomes complicated and its interpretation is difficult if we use many variables. Therefore, it is expected that PLS coupled with variable selection can produce a more predictive and interpretable model in COMBINE. The purpose of this paper is to examine whether genetic algorithm-based PLS (GAPLS) developed by our group for variable selection can enhance prediction and interpretation of the COMBINE model. The structure-activity data of human immuno-deficiency virus type I (HIV-1) protease inhibitors were used as a test example. By applying GAPLS to this data set, several improved PLS models with a high cross-validated r2 value and low number of variables were obtained. In order to select a best model from them, external validation was performed for each model. The finally selected model was further examined by comparing with the 3D structure of HIV-1 protease in computer graphics and its agreement was confirmed.

Feature selection by genetic algorithms for mass spectral classifiers

Mass spectral classifiers for 15 substructures have been computed that give discrete present/absent answers. For the development of classifiers, linear discriminant analysis (LDA) and partial least squares discriminant PLS (DPLS) have been used. The low resolution mass spectra were transformed into a set of 400 spectral features. Because each spectrum is described with so many features, some features may not be necessary, and others may contribute only noise. Therefore, the effect of feature selection has been investigated. The methods used were selection by Fisher ratios and selection by a genetic algorithm (GA). The first method is univariate, the second is multivariate; advantages and disadvantages of both are discussed. On the average, feature selection did not significantly change the classification performance compared with results that have been obtained with all features. However, it was possible to reduce the number of features considerably without a loss of classification performance. For a few substructures GA together with LDA resulted in much better classifiers than DPLS with all features. The features selected for classifications of a benzyl substructure and for the presence of chlorine have been interpreted in terms of mass spectrometric fragmentation rules.

Further development of structure generation in the automated structure elucidation system CHEMICS

(1) Present address: National Center for Biomedical Infrared Spectroscopy, Battelldolumbus Laboratories, 505 King Avenue, Columbus, OH 43201. (2) Munk, M. E.; Shelley, C. A.; Woodruff, H. B.; Trulson, M. 0. “Computer-Assisted Structure Elucidation”. Fresenius’ Z . Anal. Chem. 1982, 313, 473-479. (3) Shelley, C. A.; Hays, T. R.; Roman, R. V.; Munk, M. E. “An Approach to Automated Partial Structure Expansion”. Anal. Chim. Acta 1978, 103, 121-132. (4) Carhart, R. E.; Smith, D. H.; Brown, H.; Djerassi, C. ‘Applications of Artificial Intelligence for Chemical Inference. 17. An Approach to Computer-Assisted Elucidation of Molecular Structure”. J . Am. Chem. SOC. 1975,97, 5755-5762. ( 5 ) Carhart, R. E.; Smith, D. H.; Gray, N. A. B.; Nourse, J. G.; Djerassi, C. ‘GENOA: A Computer Program for Structure Elucidation Uti(7) lizing Overlapping and Alternative Substructures”. J . Org. Chem. 1981,

Recent Advances in the Automated Structure Elucidation System, CHEMICS. Utilization of Two-Dimensional NMR Spectral Information and Development of Peripheral Functions for Examination of Candidates

A program for applying two-dimensional NMR spectroscopic data from 1H−13C COSY (one-bond C−H correlations)/1H−1H COSY (three bond H−H correlations) and 2D-INADEQUATE (one-bond C-C correlations) experiments has been developed and introduced into CHEMICS. The main concepts employed in this study consists of making a more accurate assignment of 13C and 1H chemical shifts to carbons and hydrogens in the sample structure at the data analysis step in CHEMICS followed by the generation of more probable structures as candidates using 13C−13C and 1H−1H coupling information. The detailed algorithm are described. In addition, current configuration of CHEMICS including peripheral functions for examination of candidates by mass and 13C-NMR spectral prediction and the brief overview are described.

Development of a New Regression Analysis Method Using Independent Component Analysis

In this paper, independent component analysis (ICA) and regression analysis are combined to extract significant components. ICA is a method that extracts mutually independent components from explanatory variables. A relationship between the independent components and an objective variable is constructed by the least-squares method. This method is named ICA-MLR (MLR = multiple linear regression). We verified the superiority of ICA-MLR over partial least squares (PLS) with simulation data and tried to apply this method to a quantitative structure-property relationship analysis of aqueous solubility. We constructed models between aqueous solubility and 173 molecular descriptors. PLS and genetic algorithm PLS models were constructed for a comparison of ICA-MLR. R2, Q2, and Rpred2 values of the PLS model are 0.836, 0.819, and 0.848, respectively. These values of the ICA-MLR model are 0.937, 0.868, and 0.894, respectively. ICA-MLR achieved higher predictive accuracy than PLS. ICA-MLR could extract effective components from explanatory variables and construct the regression model with high predictive accuracy. In addition, the information of regression coefficients bICA-MLR indicates the magnitude of contribution of each descriptor in the analysis of aqueous solubility.

GA STRATEGY FOR VARIABLE SELECTION IN QSAR STUDIES : GA-BASED REGION SELECTION FOR COMFA MODELING

A novel approach using a genetic algorithm (GA) for variable selection in comparative molecular field analysis (CoMFA) was developed. This approach is named GA-based region selection (GARGS) since the regularly splitting regions in 3D space are used as variables instead of each field variable. GARGS was applied to the data set of polychlorinated dibenzofurans (PCDF) as a test example. The number of field variables was reduced from 1275 to 43, and the values of cross-validated r2(q2) indicating the internal predictivity of the model equation was increased from 0.88 to 0.95 by GARGS. The structural requirements for the PCDF molecules could be easily estimated from the coefficient contour maps of the simplified CoMFA model equation. These structural requirements were consistent with the result from the previous studies, and the utility of GARGS was demonstrated.