Computer-Aided Molecular Design Using Neural Networks and Genetic Algorithms

Publisher Summary Genetic algorithms (GAs) are general purpose, stochastic, evolutionary search, and optimization strategies based on the Darwinian model of natural selection and evolution. Studies show that the genetic design (GD) approach is able to locate globally optimal designs for many target molecules with multiple specifications. In this chapter, this approach is extended further by describing a computer-aided molecular design (CAMD) framework that uses both neural networks (NNs) and GAs to address the difficulties in both the forward and inverse problems. The NN based property prediction methodology addresses the forward problem while the GA component tackles the inverse problem. This chapter also discusses some recent results on the characterization and analysis of complex search spaces for molecular design to gain further insight into the operation of GAs and describes an interactive CAMD system that is under development.

[1]  S P Levine,et al.  Spectral peak verification and recognition using a multilayered neural network. , 1990, Analytical chemistry.

[2]  John H. Holland,et al.  Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .

[3]  George C. Derringer,et al.  A computer-based methodology for matching polymer structures with required properties , 1985 .

[4]  Kevin G Joback,et al.  Designing molecules possessing desired physical property values , 1989 .

[5]  H Ichikawa,et al.  Neural networks applied to structure-activity relationships. , 1990, Journal of medicinal chemistry.

[6]  Luigi Di Pace,et al.  A machine learning approach to computer-aided molecular design , 1991, J. Comput. Aided Mol. Des..

[7]  Rafiqul Gani,et al.  MOLECULAR DESIGN OF SOLVENTS FOR LIQUID EXTRACTION BASED ON UNIFAC , 1983 .

[8]  L. Hall,et al.  Molecular connectivity in chemistry and drug research , 1976 .

[9]  D. W. Noid,et al.  Neural networks and graph theory as computational tools for predicting polymer properties , 1994 .

[10]  Thomas L. Isenhour,et al.  Chemical applications of pattern recognition , 1975 .

[11]  S. Macchietto,et al.  Design on optimal solvents for liquid-liquid extraction and gas absorption processes , 1990 .

[12]  A. J. Hopfinger,et al.  Molecular modelling of polymers: 5. Inclusion of intermolecular energetics in estimating glass and crystal-melt transition temperatures , 1989 .

[13]  S. Unger.  Molecular Connectivity in Structure–activity Analysis , 1987 .

[14]  D. B. Hibbert.  Genetic algorithms in chemistry , 1993 .

[15]  R. Gani,et al.  A group contribution approach to computer‐aided molecular design , 1991 .

[16]  R. Reid,et al.  The Properties of Gases and Liquids , 1977 .

[17]  Gerald M. Maggiora,et al.  Applications of neural networks in chemistry. 1. Prediction of electrophilic aromatic substitution reactions , 1990, J. Chem. Inf. Comput. Sci..

[18]  Bobby G. Sumpter,et al.  Theory and Applications of Neural Computing in Chemical Science , 1994 .

[19]  Morton E. Munk,et al.  A neural network approach to infrared spectrum interpretation , 1990 .

[20]  Igor I. Baskin,et al.  Inverse problem in QSAR/QSPR studies for the case of topological indexes characterizing molecular shape (Kier indices) , 1993, J. Chem. Inf. Comput. Sci..

[21]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[22]  Venkat Venkatasubramanian,et al.  Internal model control framework using neural networks for the modeling and control of a bioreactor , 1995 .

[23]  Venkat Venkatasubramanian,et al.  Computer-aided molecular design using genetic algorithms , 1994 .

[24]  Lyle H. Ungar,et al.  A hybrid neural network‐first principles approach to process modeling , 1992 .

[25]  Lemont B. Kier,et al.  Design of molecules from quantitative structure-activity relationship models. 1. Information transfer between path and vertex degree counts , 1993, J. Chem. Inf. Comput. Sci..

[26]  George Cybenko,et al.  Approximation by superpositions of a sigmoidal function , 1992, Math. Control. Signals Syst..

[27]  K. Joback,et al.  ESTIMATION OF PURE-COMPONENT PROPERTIES FROM GROUP-CONTRIBUTIONS , 1987 .

[28]  T. Sejnowski,et al.  Predicting the secondary structure of globular proteins using neural network models. , 1988, Journal of molecular biology.

[29]  Venkat Venkatasubramanian,et al.  Evolutionary Design of Molecules with Desired Properties Using the Genetic Algorithm , 1995, J. Chem. Inf. Comput. Sci..

[30]  A. J. Hopfinger,et al.  Molecular modeling of polymers: I. Correct and efficient enumeration of intrachain conformational energetics , 1987 .

[31]  Rafiqul Gani,et al.  A strategy for the design and selection of solvents for separation processes. , 1986 .

[32]  Anton J. Hopfinger,et al.  Molecular modeling of polymers. IV: Estimation of glass transition temperatures , 1988 .

[33]  D. V. Krevelen,et al.  Properties of polymers, their estimation and correlation with chemical structure , 1972 .

[34]  J. Zupan,et al.  Neural networks: A new method for solving chemical problems or just a passing phase? , 1991 .

[35]  Oldřich Štrouf,et al.  Chemical Pattern Recognition , 1986 .