Quantitative structure–mobility relationship study of a diverse set of organic acids using classification and regression trees and adaptive neuro-fuzzy inference systems

A novel similarity search approach for high-performance thin-layer chromatography (HPTLC) fingerprinting of medicinal plants

INTRODUCTION

In addition to the development of analytical equipment, another movement has also appeared in the field of computer assisted techniques for metabolite assessment. Although, some studies can be found in the literature there is still not available reliable and user-friendly software which is coupled with a simple chromatography method for developing a database to identify medicinal plants.

OBJECTIVES

Developing a novel similarity search approach for high-performance thin-layer chromatography (HPTLC) fingerprinting.

METHODS

Combined HPTLC with image analysis approach was used for similarity assessment of 70 standard medicinal plants. Ethyl acetate-ethyl methyl ketone-formic acid 98%-water (50:30:10:10) were chosen among different examined mobile phases. Liebermann-Burchard and anisaldehyde reagents were chosen for HPTLC derivatisation for visualisation. Image analysis based on Cannys' method was used to determine the spot size of each HPTLC image. A similarity search algorithm based on colour (RGB, HSV and Lab) information alone or together with retardation factor (R_f ) and spot size information calculated with the software was built to assess the fingerprinting of medicinal plants.

RESULTS

The software was capable of calculating spots size and R_f values. It authenticated unknown samples based on comparing images information, spots size and/or R_f in the built database. Similarity values were 75-96% for the selected plants chromatograms with those of the same plant in the database. It presents better results than principal components analysis (PCA), classification and regression trees (CART) and partial least squares discriminant analysis (PLS-DA).

CONCLUSION

The procedure paves the way for constructing a database of HPTLC images of medicinal plants.

Shuffling cross-validation-bee algorithm as a new descriptor selection method for retention studies of pesticides in biopartitioning micellar chromatography

Bee algorithm (BA) is an optimization algorithm inspired by the natural foraging behaviour of honey bees to find the optimal solution which can be proposed to feature selection. In this paper, shuffling cross-validation-BA (CV-BA) was applied to select the best descriptors that could describe the retention factor (log k) in the biopartitioning micellar chromatography (BMC) of 79 heterogeneous pesticides. Six descriptors were obtained using BA and then the selected descriptors were applied for model development using multiple linear regression (MLR). The descriptor selection was also performed using stepwise, genetic algorithm and simulated annealing methods and MLR was applied to model development and then the results were compared with those obtained from shuffling CV-BA. The results showed that shuffling CV-BA can be applied as a powerful descriptor selection method. Support vector machine (SVM) was also applied for model development using six selected descriptors by BA. The obtained statistical results using SVM were better than those obtained using MLR, as the root mean square error (RMSE) and correlation coefficient (R) for whole data set (training and test), using shuffling CV-BA-MLR, were obtained as 0.1863 and 0.9426, respectively, while these amounts for the shuffling CV-BA-SVM method were obtained as 0.0704 and 0.9922, respectively.

Bee algorithm and adaptive neuro-fuzzy inference system as tools for QSAR study toxicity of substituted benzenes to Tetrahymena pyriformis

A quantitative structure-activity relationship (QSAR) was developed to predict the toxicity of substituted benzenes to Tetrahymena pyriformis. A set of 1,497 zero- to three-dimensional descriptors were used for each molecule in the data set. A major problem of QSAR is the high dimensionality of the descriptor space; therefore, descriptor selection is one of the most important steps. In this paper, bee algorithm was used to select the best descriptors. Three descriptors were selected and used as inputs for adaptive neuro-fuzzy inference system (ANFIS). Then the model was corrected for unstable compounds (the compounds that can be ionized in the aqueous solutions or can easily metabolize under some conditions). Finally squared correlation coefficients were obtained as 0.8769, 0.8649 and 0.8301 for training, test and validation sets, respectively. The results showed bee-ANFIS can be used as a powerful model for prediction of toxicity of substituted benzenes to T. pyriformis.

Monte Carlo sampling and multivariate adaptive regression splines as tools for QSAR modelling of HIV-1 reverse transcriptase inhibitors

The present work focuses on the development of an interpretable quantitative structure-activity relationship (QSAR) model for predicting the anti-HIV activities of 67 thiazolylthiourea derivatives. This set of molecules has been proposed as potent HIV-1 reverse transcriptase inhibitors (RT-INs). The molecules were encoded to a diverse set of molecular descriptors, spanning different physical and chemical properties. Monte Carlo (MC) sampling and multivariate adaptive regression spline (MARS) techniques were used to select the most important descriptors and to predict the activity of the molecules. The most important descriptor was found to be the aspherisity index. The analysis of variance (ANOVA) and interpretable spline equations showed that the geometrical shape of the molecules has considerable effect on their activities. It seems that the linear molecules are more active than symmetric top compounds. The final MARS model derived displayed a good predictive ability judging from the determination coefficient corresponding to the leave multiple out (LMO) cross-validation technique, i.e. r (2 )= 0.828 (M = 12) and r (2 )= 0.813 (M = 20). The results of this work showed that the developed spline model is robust, has a good predictive power, and can then be used as a reliable tool for designing novel HIV-1 RT-INs.

Classification of anti-HIV compounds using counterpropagation artificial neural networks and decision trees

The main aim of the present work was to collect and categorize anti-HIV molecules in order to identify general structure-activity relationships. In this respect, a total of 5580 drugs and drug-like molecules was collected from 256 different articles published between 1992 and 2010. An algorithm called genetic algorithm-pattern search counterpropagation artificial neural networks (GPS-CPANN) was proposed for the classification of compounds. In addition, the CART (classification and regression trees) method was used for construction of decision trees and finding the best molecular descriptors. The results revealed that the developed CPANN models and decision tree can correctly classify the molecules according to their inhibition mechanisms and activities. Some general parameters such as molecular weight, average molecular weight, number of hydrogen atoms and number of hydroxyl groups were found to be important for describing the inhibition behaviour of anti-HIV agents. The developed classifier models in this work can be used to screen large libraries of compounds to identify those likely to display activity as anti-HIV agents.