Antibacterial Activities of Carbapenem Derivatives and Quantitative Structure - Activity Relationship for Drug Design

作者：Liu Jian^*; Zhou Lu; Zuo Zhili

来源：QSAR and Combinatorial Science, 2008, 27(10): 1216-1226.

DOI：10.1002/qsar.200710104

• Carbapenem
• Drug design
• GA-ANN
• QSAR

beta-Lactams, a class of antibiotics including penicillins, cephems. monobactams, and carbapenems, is widely used. In this work, the Quantitative Structure-Activity Relationship (QSAR) models have been built by Partial Least Squares (PLS), Artificial Neural Network (ANN), and Genetic Algorithm optimized ANN (GA-ANN) to study the antibacterial activities of carbapenem derivatives. Of the three methods, GA-ANN is potentially useful in predicting QSAR properties of chemical agents. Furthermore. some novel penem derivatives are designed using classical bioisosterism strategy and the antibacterial activities are predicted by the built GA-ANN model without any chemical or biological experiments. In conclusion. the niapenems and thiapenems show good antibiotic activities.

• 出版日期2008-10
• 单位四川大学