Quantitative structure–property relationships for pesticides in biopartitioning micellar chromatography

Performance comparison of nonlinear and linear regression algorithms coupled with different attribute selection methods for quantitative structure - retention relationships modelling in micellar liquid chromatography

In micellar liquid chromatography (MLC), the addition of a surfactant to the mobile phase in excess is accompanied by an alteration of its solubilising capacity and a change in the stationary phase's properties. As an implication, the prediction of the analytes' retention in MLC mode becomes a challenging task. Mixed Quantitative Structure - Retention Relationships (QSRR) modelling represents a powerful tool for estimating the analytes' retention. This study compares 48 successfully developed mixed QSRR models with respect to their ability to predict retention of aripiprazole and its five impurities from molecular structures and factors that describe the Brij - acetonitrile system. The development of the models was based on an automatic combining of six attribute (feature) selection methods with eight predictive algorithms and the optimization of hyper-parameters. The feature selection methods included Principal Component Analysis (PCA), Non-negative Matrix Factorization (NMF), ReliefF, Multiple Linear Regression (MLR), Mutual Info and F-Regression. The series of investigated predictive algorithms comprised Linear Regressions (LR), Ridge Regression, Lasso Regression, Artificial Neural Networks (ANN), Support Vector Regression (SVR), Random Forest (RF), Gradient Boosted Trees (GBT) and K-Nearest neighbourhood (k-NN). A sufficient amount of data for building the model (78 cases in total) was provided by conducting 13 experiments for each of the 6 analytes and collecting the target responses afterwards. Different experimental settings were established by varying the values of the concentration of Brij L23, pH of the aqueous phase and acetonitrile content in the mobile phase according to the Box-Behnken design. In addition to the chromatographic parameters, the pool of independent variables was expanded by 27 molecular descriptors from all major groups (physicochemical, quantum chemical, topological and spatial structural descriptors). The best model was chosen by taking into consideration the Root Mean Square Error (RMSE) and cross-validation (CV) correlation coefficient (Q²) values. Interestingly, the comparative analysis indicated that a change in the set of input variables had a minor impact on the performance of the final models. On the other hand, different regression algorithms showed great diversity in the ability to learn patterns conserved in the data. In this regard, testing many regression algorithms is necessary in order to find the most suitable technique for model building. In the specific case, GBT-based models have demonstrated the best ability to predict the retention factor in the MLC mode. Steric factors and dipole-dipole interactions have proven to be relevant to the observed retention behaviour. This study, although being of a smaller scale, is a most promising starting point for comprehensive MLC retention prediction.

Determination of quantitative retention-activity relationships between pharmacokinetic parameters and biological effectiveness fingerprints of Salvia miltiorrhiza constituents using biopartitioning and microemulsion high-performance liquid chromatography

In this study, we analyzed danshen (Salvia miltiorrhiza) constituents using biopartitioning and microemulsion high-performance liquid chromatography (MELC). The quantitative retention-activity relationships (QRARs) of the constituents were established to model their pharmacokinetic (PK) parameters and chromatographic retention data, and generate their biological effectiveness fingerprints. A high-performance liquid chromatography (HPLC) method was established to determine the abundance of the extracted danshen constituents, such as sodium danshensu, rosmarinic acid, salvianolic acid B, protocatechuic aldehyde, cryptotanshinone, and tanshinone IIA. And another HPLC protocol was established to determine the abundance of those constituents in rat plasma samples. An experimental model was built in Sprague Dawley (SD) rats, and calculated the corresponding PK parameterst with 3P97 software package. Thirty-five model drugs were selected to test the PK parameter prediction capacities of the various MELC systems and to optimize the chromatographic protocols. QRARs and generated PK fingerprints were established. The test included water/oil-soluble danshen constituents and the prediction capacity of the regression model was validated. The results showed that the model had good predictability.

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.

Predicting retention times of naturally occurring phenolic compounds in reversed-phase liquid chromatography: a Quantitative Structure-Retention Relationship (QSRR) approach

Quantitative structure-retention relationships (QSRRs) have successfully been developed for naturally occurring phenolic compounds in a reversed-phase liquid chromatographic (RPLC) system. A total of 1519 descriptors were calculated from the optimized structures of the molecules using MOPAC2009 and DRAGON softwares. The data set of 39 molecules was divided into training and external validation sets. For feature selection and mapping we used step-wise multiple linear regression (SMLR), unsupervised forward selection followed by step-wise multiple linear regression (UFS-SMLR) and artificial neural networks (ANN). Stable and robust models with significant predictive abilities in terms of validation statistics were obtained with negation of any chance correlation. ANN models were found better than remaining two approaches. HNar, IDM, Mp, GATS2v, DISP and 3D-MoRSE (signals 22, 28 and 32) descriptors based on van der Waals volume, electronegativity, mass and polarizability, at atomic level, were found to have significant effects on the retention times. The possible implications of these descriptors in RPLC have been discussed. All the models are proven to be quite able to predict the retention times of phenolic compounds and have shown remarkable validation, robustness, stability and predictive performance.

Quantitative and qualitative prediction of corneal permeability for drug-like compounds

A set of 69 drug-like compounds with corneal permeability was studied using quantitative and qualitative modeling techniques. Multiple linear regression (MLR) and multilayer perceptron neural network (MLP-NN) were used to develop quantitative relationships between the corneal permeability and seven molecular descriptors selected by stepwise MLR and sensitivity analysis methods. In order to evaluate the models, a leave many out cross-validation test was performed, which produced the statistic Q(2)=0.584 and SPRESS=0.378 for MLR and Q(2)=0.774 and SPRESS=0.087 for MLP-NN. The obtained results revealed the suitability of MLP-NN for the prediction of corneal permeability. The contribution of each descriptor to MLP-NN model was evaluated. It indicated the importance of the molecular volume and weight. The pattern recognition methods principal component analysis (PCA) and hierarchical clustering analysis (HCA) have been employed in order to investigate the possible qualitative relationships between the molecular descriptors and the corneal permeability. The PCA and HCA results showed that, the data set contains two groups. Then, the same descriptors used in quantitative modeling were considered as inputs of counter propagation neural network (CPNN) to classify the compounds into low permeable (LP) and very low permeable (VLP) categories in supervised manner. The overall classification non error rate was 95.7% and 95.4% for the training and prediction test sets, respectively. The results revealed the ability of CPNN to correctly recognize the compounds belonging to the categories. The proposed models can be successfully used to predict the corneal permeability values and to classify the compounds into LP and VLP ones.

The use of machine learning and nonlinear statistical tools for ADME prediction

Absorption, distribution, metabolism and excretion (ADME)-related failure of drug candidates is a major issue for the pharmaceutical industry today. Prediction of ADME by in silico tools has now become an inevitable paradigm to reduce cost and enhance efficiency in pharmaceutical research. Recently, machine learning as well as nonlinear statistical tools has been widely applied to predict routine ADME end points. To achieve accurate and reliable predictions, it would be a prerequisite to understand the concepts, mechanisms and limitations of these tools. Here, we have devised a small synthetic nonlinear data set to help understand the mechanism of machine learning by 2D-visualisation. We applied six new machine learning methods to four different data sets. The methods include Naive Bayes classifier, classification and regression tree, random forest, Gaussian process, support vector machine and k nearest neighbour. The results demonstrated that ensemble learning and kernel machine displayed greater accuracy of prediction than classical methods irrespective of the data set size. The importance of interaction with the engineering field is also addressed. The results described here provide insights into the mechanism of machine learning, which will enable appropriate usage in the future.

Prediction of the retention of β-diketonato complexes in TLC systems on silica gel by quantitative structure-retention relationships

Quantitative structure-retention relationships for a series of 30 mixed ?-diketonato complexes of cobalt(III), chromium(III) and ruthenium(III) were derived by multiple linear regression analyses using molecular descriptors obtained by quantum chemical calculations. The retention parameters were obtained by thin layer chromatography on silica gel using mono and two-component solvent systems. The molecular descriptors included in the multiple linear regression analysis were molecular weight, molecular volume, surface area, hydrophilic-lipophilic balance, percent hydrophilic surface area, dipole moment, polarizability, refractivity, energy of the highest occupied molecular orbital and energy of the lowest unoccupied molecular orbital. High agreement between the experimental and predicted retention parameters was obtained when polarizability and the hydrophilic-lipophilic balance were used as the molecular descriptors. Comparison of the models with those established on polyacrylonitrile showed that the structure of the sorbent is responsible for the chromatographic behaviour of the same compounds. The presented models can be used for the prediction of the retention of new solutes in screening chromatographic systems.

Quantitative structure-retention relationships for organic pollutants in biopartitioning micellar chromatography

Quantitative structure-retention relationship (QSRR) models have been successfully developed for the prediction of the retention factor (log k) in the biopartitioning micellar chromatography (BMC) of 66 organic pollutants. Heuristic method (HM) and radial basis function neural networks (RBFNN) were utilized to construct the linear and non-linear QSRR models, respectively. The optimal QSRR model was developed based on a 6-17-1 radial basis function neural network architecture using molecular descriptors calculated from molecular structure alone. The RBFNN model gave a correlation coefficient (R2) of 0.8464 and root-mean-square error (RMSE) of 0.1925 for the test set. This paper provided a useful model for the predicting the log k of other organic compounds when experiment data are unknown.

Quantitative structure-(chromatographic) retention relationships

Since the pioneering works of Kaliszan (R. Kaliszan, Quantitative Structure-Chromatographic Retention Relationships, Wiley, New York, 1987; and R. Kaliszan, Structure and Retention in Chromatography. A Chemometric Approach, Harwood Academic, Amsterdam, 1997) no comprehensive summary is available in the field. Present review covers the period of 1996-August 2006. The sources are grouped according to the special properties of kinds of chromatography: Quantitative structure-retention relationship in gas chromatography, in planar chromatography, in column liquid chromatography, in micellar liquid chromatography, affinity chromatography and quantitative structure enantioselective retention relationships. General tendencies, misleading practice and conclusions, validation of the models, suggestions for future works are summarized for each sub-field. Some straightforward applications are emphasized but standard ones. The sources and the model compounds, descriptors, predicted retention data, modeling methods and indicators of their performance, validation of models, and stationary phases are collected in the tables. Some important conclusions are: Not all physicochemical descriptors correlate with the retention data strongly; the heat of formation is not related to the chromatographic retention. It is not appropriate to give the errors of Kovats indices in percentages. The apparently low values (1-3%) can disorient the reviewers and readers. Contemporary mean interlaboratory reproducibility of Kovats indices are about 5-10 i.u. for standard non polar phases and 10-25 i.u. for standard polar phases. The predictive performance of QSRR models deteriorates as the polarity of GC stationary phase increases. The correlation coefficient alone is not a particularly good indicator for the model performance. Residuals are more useful than plots of measured and calculated values. There is no need to give the retention data in a form of an equation if the numbers of compounds are small. The domain of model applicability of models should be given in all cases.

Micellar liquid chromatography with dodecyl dimethyl betaine as an in vitro method for prediction of protein-drug binding

With the accelerating development of new drugs, there is a high need for rapid and simple screening technologies. In this paper, a new in vitro method, dodecyl dimethyl betaine (BS-12) micellar liquid chromatography (MLC) was presented for prediction of protein-drug binding based on the similar property of BS-12 micelles to protein. The predictive possibility of this method was validated by comparing the retention factors of drugs (antidiabetic and antibacterial drugs) on C18 modified by different surfactants with those on the protein column. Through the investigation of the concentration and pH effect on the retention of the drugs in BS-12 MLC, quantitative retention-protein binding relationships were established according to the retention factors in 0.2 M BS-12 (pH 7.4) MLC and those on the protein column. According to the relationships established, the protein binding of seven drugs for psychiatric disorders, six potential drugs for antibiotics and four commercial antibiotics were predicted. The results were consistent with those on the BSA column very well. This indicated, BS-12 MLC was a simple, fast and reproducible method to predict protein-drug binding.