DEVELOPMENT OF FUZZY ADAPTIVE LEAST-SQUAR ES AND ITS USES IN QUANTITATIVE STRUCTURE-ACTIVITY-RELATIONSHIPS

Fuzzy adaptive least-squares (FALS), a pattern recognition method of t he correlating chemical structure with activity rating, has been devel oped to generate quantitative structure-activity relationship (QSAR) m odels for the drug design and screening hazardous chemicals. It is a n ovel feature of FALS that the degree to which each sample belongs to i ts activity rating is given by a fuzzy membership function. The method and its application contributed by the author and his coworkers were briefly reviewed. Using FALS, congeneric QSAR analyses of 29 allergeni c alpha-methylene-gamma-butyrolactones and 31 calmodulin inhibitors, a nd non-congeneric QSAR analyses of aquatic toxicity of 394 organic che micals and human acute toxicity of 504 compounds were performed to con struct predictive QSAR models. Furthermore, predictive accuracies of r odent carcinogenicity of 25 organic compounds issued from U.S. Nationa l Institute of Environmental Health Sciences were compared using sever al methods including a QSAR model with FALS. Finally a revised method of calculationg log P (partition coefficient in octanol/water) was pro posed for the QSAR studies. The revised method is not only simple and convenient but also reliable compared with the Rekker method and the H ansch-Leo method.