QSAR Modeling of Genotoxicity on Non-congeneric Sets of Organic Compounds

摘要

A multi-linear (ML) and artificial neural network (ANN) approaches have been used to derive quantitativestructure-activity relationships (QSAR) between the genotoxicity (mutagenicity) and molecular structure of compounds by using large initial pools of descriptors. All derived models involve descriptors that describe possible structural factors influencing the mutagenicbehavior of organic compounds. Different quantum chemical characteristics of compounds have been successfully used together with conventional molecular descriptors. The connection between descriptors represented in the models and the mutagenic behavior ofcompounds is also discussed.