Multivariate Regression Outperforms Several Robust Architectures of Neural Networks in QSAR Modeling

In the past decade, many authors replaced multivariate regression (MR) by the neural networks (NNs) algorithm because they believed the latter to be superior. To verify this, we have undertaken a comparative investigation of the relationship between biological activities and substituent constants representing physicochemical parameters of the substituent groups of 37 carboquinones and 57 benzodiazepines using MR and NNs. A new method for the selection of descriptors in the best possible MR models is presented. The use of orthogonalization procedure makes the calculation of the statistical parameters (e.g. correlation coefficient, R) for each model much simpler, and the selection of the best MR models is accelerated. Such a procedure is applicable to QSAR modeling even for the selection of the best MR model with six descriptors from a set of 100 descriptors. In case one wants to select, for example, the best 15 out of 100 descriptors, a new procedure is developed for the stepwise selection of descriptors i...

[1]  Milan Randic,et al.  Orthogonal molecular descriptors , 1991 .

[2]  M Karplus,et al.  Evolutionary optimization in quantitative structure-activity relationship: an application of genetic neural networks. , 1996, Journal of medicinal chemistry.

[3]  Igor V. Tetko,et al.  Neural Network Studies, 2. Variable Selection , 1996, J. Chem. Inf. Comput. Sci..

[4]  N. Trinajstić,et al.  New Developments in QSPR/QSAR Modeling Based on Topological Indices , 1997 .

[5]  D. Livingstone,et al.  Structure-activity relationships of antifilarial antimycin analogues: a multivariate pattern recognition study. , 1990, Journal of medicinal chemistry.

[6]  R. Cramer,et al.  Comparative molecular field analysis (CoMFA). 1. Effect of shape on binding of steroids to carrier proteins. , 1988, Journal of the American Chemical Society.

[7]  Lionello Pogliani.  Molecular connectivity descriptors of the physicochemical properties of the .alpha.-amino acids , 1994 .

[8]  H. Kubinyi.  QSAR: Hansch Analysis and Related Approaches: Kubinyi/QSAR , 1993 .

[9]  M. Karplus,et al.  Genetic neural networks for quantitative structure-activity relationships: improvements and application of benzodiazepine affinity for benzodiazepine/GABAA receptors. , 1996, Journal of medicinal chemistry.

[10]  Milan Randić,et al.  Generalized molecular descriptors , 1991 .

[11]  Igor V. Tetko,et al.  Neural Network Studies. 3. Variable Selection in the Cascade-Correlation Learning Architecture , 1998, J. Chem. Inf. Comput. Sci..

[12]  P. Petersen,et al.  Synthesis and structure-activity relationships of new 7-[3-(fluoromethyl)piperazinyl]- and -(fluorohomopiperazinyl)quinolone antibacterials. , 1990, Journal of medicinal chemistry.

[13]  Corwin Hansch,et al.  Comparative QSAR: Toward a Deeper Understanding of Chemicobiological Interactions. , 1996, Chemical reviews.

[14]  George M. Whitesides,et al.  FEED-FORWARD NEURAL NETWORKS IN CHEMISTRY : MATHEMATICAL SYSTEMS FOR CLASSIFICATION AND PATTERN RECOGNITION , 1993 .

[15]  H Ichikawa,et al.  Neural networks applied to quantitative structure-activity relationship analysis. , 1990, Journal of medicinal chemistry.

[16]  T. A. Andrea,et al.  Applications of neural networks in quantitative structure-activity relationships of dihydrofolate reductase inhibitors. , 1991, Journal of medicinal chemistry.

[17]  D. Maddalena,et al.  Prediction of receptor properties and binding affinity of ligands to benzodiazepine/GABAA receptors using artificial neural networks. , 1995, Journal of medicinal chemistry.

[18]  Han van de Waterbeemd,et al.  Chemometric Methods in Molecular Design: van de Waterbeemd/Chemometric , 1995 .

[19]  M. Randic,et al.  ISOMERIC VARIATIONS IN ALKANES : BOILING POINTS OF NONANES , 1994 .

[20]  Davor Juretic,et al.  The Structure-Property Models Can Be Improved Using the Orthogonalized Descriptors , 1995, J. Chem. Inf. Comput. Sci..

[21]  Zlatko Mihalić,et al.  A Structure-Property Study of the Solubility of Aliphatic Alcohols in Water , 1995 .

[22]  C. Hansch.  Quantitative approach to biochemical structure-activity relationships , 1969 .

[23]  M. Randic,et al.  MOLECULAR PROFILES NOVEL GEOMETRY-DEPENDENT MOLECULAR DESCRIPTORS , 1995 .

[24]  M. Karelson,et al.  QSPR: the correlation and quantitative prediction of chemical and physical properties from structure , 1995 .

[25]  Patrick Camilleri,et al.  Chiral chromatography and multivariate quantitative structure-property relationships of benzimidazole sulphoxides , 1993, J. Comput. Aided Mol. Des..

[26]  D. Manallack,et al.  Statistics using neural networks: chance effects. , 1993, Journal of medicinal chemistry.

[27]  Nenad Trinajstić,et al.  A Novel QSPR Approach to Physicochemical Properties of the α-Amino Acids , 1995 .

[28]  Lionello Pogliani,et al.  Molecular connectivity model for determination of physicochemical properties of .alpha.-amino acids , 1993 .

[29]  Nenad Trinajstic,et al.  Structure-Activity Correlation of Flavone Derivatives for Inhibition of cAMP Phosphodiesterase , 1996, J. Chem. Inf. Comput. Sci..

[30]  Alan R. Katritzky,et al.  Quantum-Chemical Descriptors in QSAR/QSPR Studies , 1996 .

[31]  M. Arakawa,et al.  Quantitative structure-activity relationships in 2,5-bis(1-aziridinyl)-p-benzoquinone derivatives against leukemia L-1210. , 1979, Journal of medicinal chemistry.

[32]  Alan R. Katritzky,et al.  Prediction of Gas Chromatographic Retention Times and Response Factors Using a General Quantitative Structure-Property Relationship Treatment , 1994 .

[33]  Hugo Kubinyi,et al.  Evolutionary variable selection in regression and PLS analyses , 1996 .

[34]  G. Drummond,et al.  Deoxyribonucleoside-3′,5′ Cyclic Phosphates. Synthesis and Acid-Catalyzed and Enzymic Hydrolysis , 1964 .

[35]  Qian Liu,et al.  Comparison of the functional-link net and the generalized delta rule net in quantitative structure-activity relationship studies , 1992 .

[36]  Toshio Fujita,et al.  A Method for the Correlation of Biological Activity and Chemical Structure , 2001 .

[37]  Federico G. De las Heras,et al.  Alkylating nucleosides 1. Synthesis and cytostatic activity of N-glycosyl(halomethyl)-1,2,3-triazoles. A new type of alkylating agent. , 1979 .

[38]  Keith L. Peterson.  Quantitative Structure-Activity Relationships in Carboquinones and Benzodiazepines Using Counter-Propagation Neural Networks , 1995, J. Chem. Inf. Comput. Sci..

[39]  S. Smith,et al.  Cyclization-activated prodrugs. Basic esters of 5-bromo-2'-deoxyuridine. , 1990, Journal of medicinal chemistry.