Phenyl Alkyl Amine Hallucinogenic Agents And Diaryl Pyrrole Ketones 5-ht <sub> 2c </ Sub> Inhibitors Qsar Study

Quantitative structure-activity relationship is an internationally active research field. It is a frontal task in agricultural chemistry, environmental chemistry and medicinal chemistry. Especially in environmental toxicology estimation and computer-aided drug design, QSAR is gaining popularity and wide applications. This thesis introduced artificial neural network, support vector machine, comparative molecular field analysis and comparative molecular similarity indices analysis to study the structure-activity relationships of hallucinogenic phenylalkylamines and diaryl- substituted pyrrolones as 5-HT_2C inhibitors, respectively. Based on our perspective work, some important factors determining the activity were found, which can be helpful for structure modification and discovery of new active compounds.In the first part of the paper, a review of QSAR on its studying methods, progress and applications is presented.In chapter 2 and 3 of the thesis, quantum, topological, and hydrophobic descriptors were combined to study the quantitative structure- activity relationship of 88 phenylalkylamines. Stepwise multiple regression was chosen to find out the most influential descriptors, and seven descriptors, accounting for distribution of atomic charges, molecular orbital energy, molecular branching degree and hydrophobic property were selected. Three methods that is Multiple Linear Regression (MLR), Generalized Regression Neural Network (GRNN) and Support Vector Machine (SVM), were used to build QSAR models respectively. Comparison of the results obtained from the three models showed that the SVM and GRNN methods exhibited better performance than MLR method. SVR method got better results for the prediction of the test set, which showed the better generalization of the SVR method.Chapter four discussed the 3D-QSAR of 90 phenylalkylamines by CoMFA method. Two conformations were compared during the modeling. Conformation I referred to the amino group close to ring position 6 and conformation II related to the amino group trans to the phenyl ring. Satisfactory results were obtained by using both conformations. Comparison of the statistic results showed that the model based on conformation I was better than conformation II. And this may suggest that conformation I be preponderant when the hallucinogenic phenylalkylamines interact with the receptor. In the process of modeling, the influences of some important parameters were investigated and discussed. The CoMFA coefficient contour maps suggested that both steric and electrostatic interactions play an important role. The contributions from the steric and electrostatic fields were 0.450 and 0.550, respectively. It is helpful to study the mechanism and identify new potential new hallucinogens.In chapter five, 34 compounds of diaryl substituted pyrrolones as 5-HT_2C inhibitors were studied. Two 3D-QSAR methods were utilized. For the CoMFA method, only the steric and electrostatic fields were not complete to explain the interactions between the compounds and the 5-HT_2C receptor. When adding the hydrophobic field acquired from the CoMSIA method, a much better model was established. The q² value was 0.563 with principal components of 6, the noncross- validated squared coefficient was 0.924 and standard error of estimation was 0.229. For the CoMSIA method, the best model was established also by the three fields with the optimal number compotent of 6,q² of 0.571 and R²of 0.929, standard error of estimation of0.225. And the contributions of the steric, electrostatic and the hydrophobic fields were 0.114,0.422 and 0.463 respectively. It can be concluded that the activity of these inhibitors was mainly dependent on the steric, electrostatic and hydrophobic interactions. And comparison of the contributions of the three fields suggests that the hydrophobic interactions play a much important role in the interactions between the ligands and the 5-HT_2C recptor.