QSAR Study of Artemisinin Analogues as Antimalarial Drugs by Neural Network and Replacement Method

Abstract

Quantitative structure-activity relationship (QSAR) models were derived for 179 analogues of artemisinin, a potent antimalarial agent. Molecular descriptors derived solely from molecular structure were used to represent molecular structure. Utilizing replacement method, a subset of 11 descriptors was selected. General regression neural network (GRNN) was used to construct the nonlinear QSAR models in all stages of study. The relative standard error percent in antimalarial activity predictions for the training set by the application of cross-validation (RMSE-CV) was 0.43, and for test set (RMSE) was 0.51. GRNN analysis yielded predicted activities in the excellent agreement with the experimentally obtained values (R2training = 0.967 and R2test = 0.918). The mean absolute error for the test set was computed as 0.4115.

In Silico Rational Design and Virtual Screening of Bioactive Peptides Based on QSAR Modeling

“…The presence of outliers was checked using the Williams plot. This plot utilizes simultaneously the concepts of standardized residual and leverage to show visually the applicability domain of the model Figure depicts the Williams plot for the HLA data set, indicating that there are several outliers.…”

“…Bioinformatic studies have become more and more popular in peptide’s design, particularly the quantitative structure–activity relationship (QSAR) study. QSAR models utilize a mathematical function to summarize the relationship between biological activities of a set of compounds and their structural characteristics. − So far, QSAR models have been successfully established for angiotensin-converting enzyme (ACE)-inhibitory peptides, antioxidant peptides, antimicrobial peptides, bitter peptides, antitumor peptides, etc. − To develop a QSAR model, a set of numerical descriptors is generated to characterize the structure of interest, e.g., amino acids, which serves as independent variables, while the biological activities are the dependent variables. Since the activities of peptides are determined by the amino acid compositions, sequences, and structures, a proper encoding technique should be employed for representing the sequence of amino acids.…”

“…a Maximum R training 2 for the Y-randomization test.Figure 3. Plot of predicted versus experimental activities for the BTT data set.model 6. Figure4depicts the Williams plot for the HLA data set, indicating that there are several outliers.…”

Design of N-11-Azaartemisinins Potentially Active against <i>Plasmodium falciparum</i> by Combined Molecular Electrostatic Potential, Ligand-Receptor Interaction and Models Built with Supervised Machine Learning Methods

“…About the investigated N-11-azaartemisinis) also exhibit the endoperoxide linkage necessary for antimalarial activity. Figure 3 shows the MEP maps for the N-11-azaartemisinins of the training set obtained by the inclusion of substituents in the N atom of the lactam function (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19). As one can see in this figure, the MEP maps are similar to artemisinin and 11-azaartemisinin in the 1, 2, 4 trioxane ring region, with the electron density of some molecules more concentrated in this region, indicating greater biological activity.…”

New structure-based models for the prediction of normal boiling point temperature of ternary azeotropes

“…Variance inflation factors (VIFs) calculated for considered descriptors in the model are all less than 10, indicating the lack of collinearities among them 30 . The external validation of the model was further verified using Q 2 F1 , Q 2 F2 , Q 2 F3 , r 2 m , and CCC (concordance correlation coefficient) 31,32 . For an acceptable model, the Q 2 F1 , Q 2 F2 , and Q 2 F3 values should be greater than 0.6, r 2 m greater than 0.5, and CCC greater than 0.85.…”