Exploring QSARs for antiviral activity of 4-alkylamino-6-(2-hydroxyethyl)-2-methylthiopyrimidines by support vector machine

The support vector machine, which is a novel algorithm from the machine learning community, was used to develop quantitative structure activity relationship models to predict the antiviral activity of 4-alkylamino-6-(2-hydroxyethyl)-2-methylthiopyrimidines. The genetic algorithm was employed to select the variables that resulted in the best-fitted models. A comparison between the obtained results using support vector machine with those of multiple linear regression revealed that support vector machine model was much better than multiple linear regression. The root mean square errors of the training set and the test set for support vector machine model were calculated to be 0.102 and 0.205, and the correlation coefficients (r2) were 0.956 and 0.852, respectively. Furthermore, the obtained statistical parameter of leave-one-out (LOO) and leave-group-out (LGO) cross-validation test on support vector machine model were 0.893 and 0.881, respectively, which prove the reliability of this model. The results suggest that branching, volume and lipophilicity are the main independent factors contributing to the antiviral activities of the studied compounds.     

Structure/activity investigations of 5-substituted 3-methylisoxazole[5, 4-d]1, 2, 3-triazin-4-one derivatives

The series of 5-substituted 3-methylisoxazole[5, 4-d]1, 2, 3-triazin-4-one derivatives was obtained by diazotization of 5-amino-3-methylisoxazol-4-carboxylic acid hydrazide. The immunological activity of these compounds was investigated experimentally in several in vitro and in vivo assays in mice and human models. In the next step, quantum-chemical investigations were performed using density functional theory with the B3LYP hybrid exchange-correlation energy functional and 6-31G(d, p) basis set. The Polarizable Continuum (SCRF/PCM) solvent model was also taken into account in order to show solvent influence on electron density and electrostatic potential around the exemplary molecules. Correlations between molecular structure and biological properties were found using a stepwise selection of scales for the multiple linear regression (MLR).     

2D, 3D-QSAR,and pharmacophore studies on thiazolidine-4-carboxylic acid derivatives as neuraminidase inhibitors in H3N2 influenza virus

Thiazolidine-4-carboxylic acid derivative compounds with their inhibitory activity against influenza A neuraminidase (NA) were used as a data set for developing the quantitative structure–activity relationship (QSAR) model. The 2D-QSAR model was developed using multiple linear regression analysis with r ² and r ² (CV) value of 0.98 and 0.70, respectively. The generated QSAR model has shown that the electrostatic and steric properties have the predominant influence on biological activity. 3D-QSAR was modeled with r ² value of 0.98 and RMSD of 0.12. The pharmacophore alignments were generated and suggested that the hydrophobic, hydrogen bond donor, and hydrogen bond acceptor features on R¹ and R² site substitutions of the core of thiazolidine are important properties to enhance the activities of molecule against influenza A NA.     

Investigations and design of pyridine-2-carboxylic acid thiazol-2-ylamide analogs as methionine aminopeptidase inhibitors using 3D-QSAR and molecular docking

Methionine amino peptidases (MetAPs) are metalloproteases that remove co-translational N-terminal methionine from nascent polypeptide chains. Due to their essential role in protein synthesis, MetAPs are considered as potential targets for antibacterial drugs. In the present work, three-dimensional quantitative structure–activity relationship (3D-QSAR) studies were carried out on a series of pyridine-2-carboxylic acid thiazol-2-ylamide-based MetAP inhibitors using comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA) techniques. The models were developed using 30 training set molecules. The optimum CoMFA and CoMSIA models obtained for the training set were statistically significant with cross-validated correlation coefficients (q ²) of 0.799 and 0.704 and conventional correlation coefficients (r ²) of 0.989 and 0.954, respectively. These inhibitors were docked into MetAP active site. The CoMFA and CoMSIA field contour maps correlate well with the structural characteristics of the binding pocket of MetAP active site. Using the knowledge of structure–activity relationship and receptor–ligand interactions from 3D-QSAR model and the docked complexes, four new pyridine-2-carboxylic acid thiazol-2-ylamide analogs were designed. These analogs exhibit significantly better predicted activity than the reported molecules. The present work has implications for the development of novel antibiotics as potent MetAP inhibitors.     

Antimycobacterial activities of novel 1-(cyclopropyl/tert-butyl/4-fluorophenyl)-1,4-dihydro- 6-nitro-4-oxo-7-(substituted secondary amino)-1,8-naphthyridine-3-carboxylic acid

Fifty-one 1-(cyclopropyl/tert-butyl/4-fluorophenyl)-1,4-dihydro-6-nitro-4-oxo-7-(substituted secondary amino)-1,8-naphthyridine-3-carboxylic acids were synthesized and evaluated for antimycobacterial in vitro and in vivo against Mycobacterium tuberculosis H37Rv (MTB), multi-drug-resistant Mycobacterium tuberculosis (MDR-TB) and Mycobacterium smegmatis (MC2) and also tested for the ability to inhibit the supercoiling activity of DNA gyrase from M. smegmatis. Among the synthesized compounds, 1-tert-butyl-1,4-dihydro-7-(4,4-dimethyloxazolidin-3-yl)-6-nitro-4-oxo-1,8-naphthyridine-3-carboxylic acid (10q) was found to be the most active compound in vitro with an MIC of 0.1 microM against MTB and MDR-TB and was 3 and 455 times more potent than isoniazid against MTB and MDR-TB, respectively. In the in vivo animal model 10q decreased the bacterial load in lung and spleen tissues with 2.39 and 3.89-log10protections respectively at the dose of 50 mg/kg body weight.     

Ligand Efficiency Outperforms pIC50 on Both 2D MLR and 3D CoMFA Models: A Case Study on AR Antagonists

The concept of ligand efficiency is defined as biological activity in each molecular size and is widely accepted throughout the drug design community. Among different LE indices, surface efficiency index (SEI) was reported to be the best one in support vector machine modeling, much better than the generally and traditionally used end‐point pIC₅₀. In this study, 2D multiple linear regression and 3D comparative molecular field analysis methods are employed to investigate the structure–activity relationships of a series of androgen receptor antagonists, using pIC₅₀ and SEI as dependent variables to verify the influence of using different kinds of end‐points. The obtained results suggest that SEI outperforms pIC₅₀ on both MLR and CoMFA models with higher stability and predictive ability. After analyzing the characteristics of the two dependent variables SEI and pIC₅₀, we deduce that the superiority of SEI maybe lie in that SEI could reflect the relationship between molecular structures and corresponding bioactivities, in nature, better than pIC₅₀. This study indicates that SEI could be a more rational parameter to be optimized in the drug discovery process than pIC₅₀.     

Quantitative Structure–Activity Relationship Analysis of a Series of Human Renal Organic Anion Transporter Inhibitors

Organic anion transporters (OATs) have been proved to play important roles in the membrane transport of numerous potentially toxic xenobiotics, drugs, and endogenous metabolites. In general, OATs substrates can compete with one another for the transporter to mutually decrease renal secretion and thus delay the clearance and prolong the duration of action of each compound. Such interactions have the potential to bring about adverse outcomes for clinical cases. Therefore, it is very important to assess the molecular bioactivity to inhibit OATs during the development of new drugs and co‐administration. In this work, the relationships between 45 chemicals and their corresponding hOAT1 and hOAT3 inhibitory activities were analyzed. The quantitative structure–activity relationship (QSAR) model was developed by genetic algorithm and multiple linear regression method. The predictive power of the proposed model was strictly evaluated, and the applicability domain was also defined. The proposed models were robust and satisfactory and could provide a feasible and effective tool for hOAT1 or hOAT3 inhibitor screening.