جستجوی مقالات مرتبط با کلیدواژه « ارتباط کمی ساختار- فعالیت » در نشریات گروه « علوم پایه »

Tuberculosis drug resistance is still one of the most important challenges in the treatment of this infectious disease, and therefore the discovery and development of new effective anti-tuberculosis drugs are always of interest to researchers. In this study, Quantitative structure – activity relationship (QSAR) analysis was applied on a series of imidazole[1,2-a] pyridinecarboxamide derivatives as anti-tuberculosis agents. The biological activity of the 18 derivatives were estimated by multiple linear regression and artificial neural network approaches. The four molecular descriptors (nCl, MATS8m, BELe4 and GATS8e) were selected by using stepwise multiple linear regression. The best results of artificial neural network were obtained with a 5-5-1 architecture trained with the feed forward backpropagation algorithm. An external test set containing 5 compounds for evaluating the model's predictive ability was used. The results showed that the artificial neural network approach provides better predictive power compared with multiple linear regression. According to the results of this study, electronegativity, atomic masses and molecular geometry have been found to be important factors controlling the anti-tuberculosis activity.

In this study, prediction of average lethal dose of important drugs for children using molecular descriptors and application of Quantitative Structure-Activity Relationship (QSAR) models by Multiple Linear Regression (MLR) and Artificial Neural Network (ANN) models have been investigated separately. After getting a lot of descriptors the Stepwise regression method was used to reduce the number of descriptors (variables) and the best results were obtained with 8 descriptors. Multivariate linear regression model was then used to predict the lethal dose of the drugs, which yielded almost good results and the parameters R2, Q2 and RMSE for this model were calculated and reported as 0.894, 12.15 and 0.882, respectively. Also by using artificial neural network a better model with correlation coefficients of training, test, validation and total groups were calculated 0.984, 0.994, 0.999 and 0.983, respectively, indicating good validity of this method for Predicting the lethal dose of other similar drugs for children.

Quantitative Structure-Activity Relationship (QSAR) was developed for modeling and predicting of the PIM inhibitory activities a data set containing 39 structures of triazolopyridine derivatives with known biological activities. Segmentation the whole dataset into a training set and test set was performed randomly. StepWise (SW) and Genetic Algorithm (GA) techniques with Multiple Linear Regression (MLR) were used to select the most important descriptors and to create the best prediction model. Comparison of the results obtained for SW-MLR and GA-MLR models was showed that GA-MLR model is superior to the SW-MLR model. The robustness and the predictive ability of the final GA-MLR model validated by internal and external statistical validations including Leave-One-Out (LOO) cross-validation, Leave-Group-Out (LGO) cross-validation, Y-randomization and external test set. High agreement between experimental and predicted activity values indicated that GA-MLR model with five variables has good quality and it could be used in design novel compounds with higher PIM inhibitor activity.