Modeling of thermodynamic and physico-chemical properties of coumarins bioactivity against Candida albicans using a Levenberg–Marquardt neural network

Abstract

Seyyedeh Soghra Mousavi1, Hanieh Bokharaie2, Shadi Rahimi3, Sima Azadi Soror4, Mehrdad Hamidi51Department of Biotechnology, School of Pharmacy, Zanjan University of Medical Science, Zanjan, Iran; 2Genetic Group, Biology Department, Faculty of Basic Sciences, Science and Research Branch, Islamic Azad University, Tehran, Iran; 3Department of Biology, Faculty of Science, Tarbiat Moallem University, Tehran, Iran: 4Plant Protection Department, Faculty of Agriculture, Bu-Ali Sina University, Hamedan, Iran; 5Department of Pharmaceutics, School of Pharmacy, Zanjan University of Medical Science, Zanjan, IranAbstract: In recent years, due to vital need for novel fungicidal agents, investigation on natural antifungal resources has been increased. The special features exhibited by neural ­network classifiers make them suitable for handling complex problems like analyzing ­different properties of candidate compounds in computer-aided drug design. In this study, by using a Levenberg–­Marquardt (LM) neural network (the fastest of the training algorithms), the ­relation between some important thermodynamic and physico-chemical properties of coumarin compounds and their biological activities (tested against Candida albicans) has been evaluated. A set of already reported antifungal bioactive coumarin and some well-known physical descriptors have been selected and using LM training algorithm the best architecture of neural model has been designed for forecasting the new bioactive compounds.Keywords: Levenberg–Marquardt algorithm, coumarin, neural network