Studies On The Relationship Between Structures And Activities Of N,N-dimethyl-2-bromo-2-phenylethylamine Derivatives And Flavone Derivatives

This dissertation consists of three parts. In the first part, the knowledge of computer aided drug design and fundamentals of our research methods have introduced simply. In the second part, we have improved the traditional method to establish the relationship between biological activities and molecular parameters of N, N-dimethyl-2-bromo-2-phenylethylamine derivatives. The third part is made up of two subsections. Firstly, we have studied the relationship on structures and Cytotoxicitives of 21 flavone derivatives; secondly, additional 41 flavone derivatives have been selected to research the relationship between structures and nitric oxide production inhibitory activities. In chapter one, the development of computer aided drug design has been introduced. Along with the progress of the computer science, quantum chemistry, computational chemistry, computer graph and so on have been combined with computer aided drug design, what make it stronger. Besides, we also introduce the fundamentals of our research methods in this dissertation, such as explanation of the quantitative structure-activity relationship (QSAR), preferences, building the model, regressive analytical method and etc. In chapter two, the traditional method of studying the relationship of structures and activities of N, N-dimethyl-2-bromo-2-phenylethylamine derivatives has been improved by adding some molecular geometrical parameters and shape parameters. Three kind parameters (namely lyophilic parameter of substituents, electronic parameter and steric parameter) have only been used in traditional method, so the prediction ability of the model is not good, and prediction error reach 20%. In order to improve the prediction ability of this model, we have introduced other six main parameters into the study of relationship. The 21 flavonoe derivatives have been optimized by AM1 method of HyperChem6.0 and Other 6 main parameters have been gained by geometry optimizations. Then the relationship between biological activities and molecular parameters has been established by using Partial Least Square method. It has been shown that the QSAR model using 7 components is much better than the traditional method. Leave-one-out Cross-validation has been used to test the stability of QSAR model and to predict the reliability. The prediction error of the model have controlled in less than 10%. The third part is made up of two subsections. In the first part, we make research to describe the quantificational relationship between structures and cytotoxicitives of 21 flavonoe derivatives against HIV. 14 quantum-chemical descriptors have been calculated at B3LYP/6-31G** level. A total of 5 important descriptors have been selected by stepwise regression analysis (SRA). These descriptors are: energy of the lowest occupied molecular orbit (ELUMO), molecular hardness (η), molecular polarizability(α), molecular dipole moment (μ) and net charge on the seventh carbon atom(Q7). A leave-one-out cross-validation method was used to select the number of latent variables for the building of the QSAR models by principal component regression (PCR) and partial least square regression (PLS) method respectively. The cross-validation squared correlation coefficients R cv2 are 0.70 for PCR and 0.74 for PLS, showing that both of these two models have good prediction ability and the PLS model is superior to the PCR model. The qsar models show the effects of five components, and lead to the investigation of the flavonoids with low toxicity. 41 flavonoe derivatives have been selected as samples to study the relationship between the structures and nitric oxide production inhibitory activities. The molecular descriptors include constitutational descriptors, geometrical molecular descriptors, electronic descriptors, charged partial surface area descriptors and topological indices. We have computed 102 descriptors by our designed molecular descriptors program and selected the variable by Monte Carlo simulated annealing. The 47 descriptors of these are chosen to build the model. Then the relationshipbetween nitric oxide production inhibitory activities and molecular parameters has been established by using Partial Least Square method and Leave-one-out Cross-validation has been used to test the stability of QSAR model and to predict the reliability. It shows that the model is effective (R=0.9683) and has good prediction ability (Q=0.7053). According to the model, three kinds of descriptors that consist of the geometrical molecular descriptors, the electronic descriptors and the molecular shape descriptors are proved to affect the activities of 41 flavonoe derivatives more greatly than topological descriptors such as radius of gyration, sum of squares of the charges, moment shape index and relative polar surface area.