Quantitative Structure-retention Relationship Study Research Articles

Quantitative structure-retention relationships study of the retention data of 5,5-disubstituted hydantoins

Summary The behavior of 5,5-disubstituted hydantoin derivatives has been studied using four chromatographic systems, two normal-phase (silica gel stationary phase with ethyl acetate–toluene and acetonitrile–toluene mobile phases) and two reversed-phase (RP-18 systems with methanol–water and acetonitrile–water mobile phases). The effect of the structures of the derivatives on their retention in both normal and reversed-phase modes was investigated by use of QSRR and molecular descriptors. By means of multiple linear regressions results were interpreted in terms of understandable and physically meaningful variables. Among six models proposed, the two best were selected and tested for prediction of retention – one for normal-phase chromatography (NPC) and the other for reversed-phase chromatography (RPC). The most informative properties in the best NPC model, with similar effects on retention, are log P and valence connectivity indices. The informative properties in the best RPC model were log P and molecular weight, but the statistics was much worse than for NPC. Cross-validation indicated the best models are reliable QSRR models.

Quantitative Structure-Retention Relationship Study of Some Phenol Derivatives in Gas Chromatography

A quantitative structure-retention relationship (QSRR) model has been developed for the gas chromatographic retention times of 37 phenolic derivatives in a DB-5 non-polar column (95% dimethyl and 5% diphenyl-polysiloxane). As a first step, multiple linear regression (MLR) was employed to gain informative descriptors that can predict the retention times of these compounds. Descriptors appearing in the MLR model are categorized as topological and geometric parameters that comply with the applied column. Furthermore, each molecular descriptor in this model was examined to unfold the relationship between molecular structures and their retention times. Then, a 4-4-1 neural network was developed using the descriptors selected by the MLR model. The comparison of the standard errors and correlation coefficients reveals the superiority of artificial neural networks (ANN) over the MLR model. This refers to the fact that the retention behaviors of molecules display non-linear characteristics. The consistency and reliability of ANN model was investigated using the L4O cross-validation technique. The obtained results are closely in compliance with the experiment. Moreover, the mean effect of descriptors shows that Kier symmetry index is the most important factor affecting the retention behavior of molecules.

DETERMINATION OF NEW GENERATION PESTICIDES IN COMPLEX PLANT MATRICES BY HPLC USING VARIOUS STATIONARY PHASES

Author interest is focused on method optimization to determine the new compounds using stationary and mobile phases with different physicochemical properties. This work deals with 5 substances composed of nitrogen containing heterocycles, 1,2,4-triazoles. The relationships between the chromatographic retention factor (k) and those physicochemical properties which are relevant in quantitative structure retention relationship (QSRR) studies were investigated. The accuracy of the simple linear regressions between the chromatographic retention and the descriptors for all of the compounds was satisfactory (correlation coefficient, 0.8 ≤ r2 ≤ 1.0). The QSRR models of compounds could be predicted with a multiple linear regression equation having the statistical index, r2 = 1.000. Evaluation of chromatographic properties of the new stationary phases and description of the molecular separation mechanism using the QSRR method including molecular modeling will be suggested.

A quantitative structure–retention relationship study for prediction of chromatographic relative retention time of chlorinated monoterpenes

A novel quantitative structure–retention relationship model has been developed for the gas chromatographic relative retention times (tR) of 67 polychlorinated monoterpene congeners in a non-polar column. Modeling of the relative retention time of these compounds as a function of the theoretically derived descriptors was established by principal component and partial least squares regressions. The choice of optimal training sets is efficiently performed by Kohonen self-organizing map. The genetic algorithm was used for the selection of the variables resulted in the best-fitted models. Appropriate models with low standard errors and high correlation coefficients were obtained. Wiener index, Balaban index, and ideal gas thermal capacity are examples of the descriptors affected by the retention times of polychlorinated monoterpenes.

Quantitative structure-retention relationship studies for taxanes including epimers and isomeric metabolites in ultra fast liquid chromatography

Seven pairs of epimers and one pair of isomeric metabolites of taxanes, each pair of which have similar structures but different retention behaviors, together with additional 13 taxanes with different substitutions were chosen to investigate the quantitative structure-retention relationship (QSRR) of taxanes in ultra fast liquid chromatography (UFLC). Monte Carlo variable selection (MCVS) method was adopted to choose descriptors. The selected four descriptors were used to build QSRR model with multi-linear regression (MLR) and artificial neural network (ANN) modeling techniques. Both linear and nonlinear models show good predictive ability, of which ANN model was better with the determination coefficient R 2 for training, validation and test set being 0.9892, 0.9747 and 0.9840, respectively. The results of 100 times’ leave-12-out cross validation showed the robustness of this model. All the isomers can be correctly differentiated by this model. According to the selected descriptors, the three dimensional structural information was critical for recognition of epimers. Hydrophobic interaction was the uppermost factor for retention in UFLC. Molecules’ polarizability and polarity properties were also closely correlated with retention behaviors. This QSRR model will be useful for separation and identification of taxanes including epimers and metabolites from botanical or biological samples.

Determination of Moclobemide and its Metabolites in Human Plasma by SPE-HPLC-UV: Evaluation of Critical Experimental Conditions and QSRR Study

A SPE-HPLC-UV method for determination of moclobemide and its major metabolites, Ro 12-5637 and Ro 12-8095, in human plasma had been developed previously and its selectivity was evaluated against the most frequently coadministered drugs. The objective of the present work was to develop a quantitative structure retention relationship (QSRR) model capable of providing good predictions of chromatographic behaviour of other related potentially interfering drugs, based on the previously generated data, in order to further improve the clinical applicability of the existing method. Moreover, the most critical factors affecting SPE and chromatographic separations of moclobemide and its metabolites were evaluated and discussed, taking into account molecular properties of the analyzed compounds.

Quantitative structure-retention relationship and 3D molecular modeling studies of the unusual chromatographic behavior of triphenylmethane derivatives in RPTLC systems

Dual retention behavior is observed for triphenylmethane (TPM) derivatives in RPTLC on silica gel plates when the composition of acetone-water mobile phases is varied. The physicochemical and molecular properties of the TPM derivatives causing this unusual retention behavior have been investigated by traditional quantitative structure-retention relationship (QSRR) modeling and by 3D molecular modeling. The QSRR studies were performed by using PLS regression analysis based upon use of selected sets of Dragon molecular descriptors and pharmacokinetically relevant VolSurf descriptors to identify the physicochemical properties that govern chromatographic behavior. Comparative molecular similarity indices analysis (CoMSIA) was used to create the 3D isoenergy contours of favored and unfavored contributions of the molecular fields around the molecules in the RPTLC systems used. The dual retention behavior of the TPM derivatives can be attributed to the propensity of these molecules to become oriented with differ...

Quantitative Structure-Retention Relationship Studies with Biopartitioning Micellar Chromatography Systems by Amended Linear Solvation Energy Relationships in Consideration of Electronic Factor

Linear solvation energy relationship (LSER) amended by the introduction of a molecular electronic factor was employed to establish quantitative structure-retention relationship of biopartitioning micellar chromatography (BMC) system. The chromatographic indices, log k, were determined by LC on a C18 column for sixty-five structurally diverse compounds, including neutral (32), acidic (19) and basic (14) compounds. Two micellar mobile phases composed of 0.04 mol L−1 polyoxyethylene (23) lauryl ether (Brij35) were adjusted by phosphate buffer to pH 7.4 and pH 6.5, respectively. When the mean net charge per molecule (δ) was introduced into LSER as the sixth variable, the LSER regression coefficients and predictive capability were significantly improved. However, the δ coefficients of the amended LSER were quite different for acidic and basic compounds, indicating that the molecular electronic factor had a markedly different impact on the retention of acidic and basic compounds in the studied BMC system. This may attribute to the extra interaction for ionized compounds with the free silanol groups in the stationary phase. The comparison of calculated and experimental retention indices suggested that the amended LSER could reproduce adequately the retention of the structurally diverse solutes investigated in BMC.

QSRR Prediction of the Chromatographic Retention Behavior of Painkiller Drugs

Quantitative structure-retention relationship (QSRR) analysis is a useful technique capable of relating chromatographic retention time to the chemical structure of a solute. A QSRR study has been carried out on the reversed-phase high-performance liquid chromatography retention times (log tR) of 62 diverse drugs (painkillers) by using molecular descriptors. Multiple linear regression (MLR) is utilized to construct the linear QSRR model. The applied MLR is based on a variety of theoretical molecular descriptors selected by the stepwise variable subset selection procedure. Stepwise regression was employed to develop a regression equation based on 50 training compounds, and predictive ability was tested on 12 compounds reserved for that purpose. The geometry of all drugs was optimized by the semi-empirical method AM1 and used to calculate different molecular descriptors. The regression equation included three parameters: n-octanol-water partition coefficient (log P), molecular surface area, and hydrophilic-lipophilic balance of the drug molecules, all of which could be related to retention time property. Modeling of retention times of these compounds as a function of the theoretically derived descriptors was established by MLR. The results indicate that a strong correlation exists between the log tR and the previously mentioned descriptors for drug compounds. The prediction results are in good agreement with the experimental values.

Comprehensive comparison of eight statistical modelling methods used in quantitative structure–retention relationship studies for liquid chromatographic retention times of peptides generated by protease digestion of the Escherichia coli proteome

In this study, we propose a new peptide characterization method that gives attention to both the amino acid composition and the residue local environment. Using this approach, structural characteristics of peptides derived from Escherichia coli proteome were parameterized and, based upon that, the performance profile of eight statistical modelling methods were validated rigorously and compared comprehensively by applying them to modelling relationship between the sequence structure and retention ability for 816 experimentally measured peptides and to predicting normalized retention times for 121,273 unmeasured peptides in liquid chromatography. Results show that the regression models constructed by nonlinear approaches are more robust and predictable but time-consuming than those by linear ones. In these modelling methods, Gaussian process and back-propagation neural network possess the best stability, unbiased ability and predictive power, thus they can be used to accurately model the peptide structure–retention relationships; multiple linear regression and partial least squares regression perform worse compared to nonlinear modelling techniques but they are computationally efficient, so they are promising candidates for solving the qualitative problems involved in massive data. In addition, by investigating the descriptor importance in different models we found that the amino acid composition presents a significantly linear correlation with the retention time of peptides, whereas the residue environment is mainly correlated nonlinearly with peptide retention. The polar Arg and strongly hydrophobic amino acids such as Leu, Ile, Phe, Trp and Val are the critical factors influencing peptide retention behavior.

Investigation of different linear and nonlinear chemometric methods for modeling of retention index of essential oil components: Concerns to support vector machine

The quantitative structure-retention relationship (QSRR) of the essential oil components against the gas chromatography retention index (RI) was studied. The genetic algorithm (GA) was employed to select the variables that resulted in the best-fitted models. After the variables were selected, the linear multivariate regressions [e.g. the multiple linear regression (MLR), the partial least squares (PLS)] as well as the nonlinear regressions [e.g. the polynomial PLS (poly-PLS), the support vector machine (SVM)] were utilized to construct the linear and nonlinear QSRR models. The obtained results using SVM were compared with those of MLR, PLS and poly-PLS, exhibiting that the SVM model demonstrated a better performance than that of the other models. The relative standard error SE (%) of the training set and the test set for the SVM model was 1.96 and 4.25, and the square correlation coefficients were 0.987 and 0.962 respectively, while the square correlation coefficient of the cross validation ( Q 2) on the SVM model was 0.963, revealing the reliability of this model. The resulting data indicated that SVM could be used as a powerful modeling tool for the QSRR studies. This is the first research on the QSRR of the essential oil compounds against the retention index using the SVM.

Quantitative Structure-Retention Relationship study of the constituents of saffron aroma in SPME-GC–MS based on the Projection Pursuit Regression method

Quantitative Structure-Retention Relationship (QSRR) studies were performed for predicting the retention times of 43 constituents of saffron aroma, which were analyzed by solid-phase micro-extraction gas chromatography–mass spectrometry (SPME-GC–MS). The chemical descriptors were calculated from the molecular structures of the constituents of saffron aroma alone, and the linear and non-linear QSRR models were constructed using the Best Multi-Linear Regression (BMLR) and Projection Pursuit Regression (PPR) methods. The predicted results of the two approaches were in agreement with the experimental data. The coefficients of determination ( R 2) of the linear BMLR model were 0.9434 and 0.8725 for the training and test sets, respectively. The other non-linear PPR model gave a more accurate prediction with R 2 values of 0.9806 (training set) and 0.9456 (test set). The proposed models could also identify and provide some insights into structural features that may play a role on the retention behaviors of the constituents of saffron aroma in the SPME-GC–MS system. This study affords a simple but efficient approach for studying the retention behaviors of other similar plants and herbs.

QSRR Study of Psychiatric Drugs Using Classification and Regression Trees Combined with Adaptive Neuro‐Fuzzy Inference System

A new Quantitative Structure–Retention Relationship (QSRR) approach was carried out for prediction of gas–liquid retention times of 124 psychiatric drugs in whole blood on fused-silica capillary column coated with crosslinked methylsilicone with nitrogen–phosphorus detection. After screening the descriptors, a total of 699 topological, geometric, and electronic descriptors (zero- to three-dimensional) representing various structural characteristics were calculated for each molecule in the dataset. Combined method of Classification and Regression Tree (CART) as a feature selection method for the extraction of four relevant descriptors and Adaptive Neuro-Fuzzy Inference System (ANFIS) as a modeling technique was used for the prediction of retention times of diverse set of psychiatric drugs. The Root Mean Square Errors (RMSEs) for the calibration and prediction sets are 0.457 and 0.514, respectively.

QSRR Study of GC Retention Indices of Essential-Oil Compounds by Multiple Linear Regression with a Genetic Algorithm

Quantitative structure–retention relationships (QSRR) for components of the essential oil of the plant Bidens pilosa Linn. var. Radiata were studied to enable prediction of their retention indices (I R). A data set was selected consisting of the retention indices of 44 components of the essential oil with a range of more than 635 units. A suitable set of molecular descriptors was then calculated and the best-fitting descriptors were selected by using stepwise multiple linear regression (SW-MLR) and a genetic algorithm (GA-MLR) the selection of variables. Comparison of the results obtained indicated the superiority of the genetic algorithm over the stepwise multiple regression method for feature selection. The predictive quality of the QSRR models was tested for an external prediction set of nine compounds, randomly chosen from the 44 compounds. One GA-MLR model with five selected descriptors was obtained. This model, with high statistical significance (R 2 train = 0.977, SE (%) = 2.33, F = 243.275, R 2 pred = 0.978), could be used to predict the retention indices of the molecules with error <6%.

Quantitative Structure–Retention Relationship Study of Benzodiazepines Using Adaptive Neuro Fuzzy Inference System as Feature Selection Method

AbstractA Quantitative Structure–Retention Relationship (QSRR) study of 32 benzodiazepines is performed in this work. Two feature selection methods of Adaptive Neuro Fuzzy Inference System (ANFIS) and a stepwise regression approach adopted for the Multiple Linear Regressions (MLR) were used to predict the Liquid Chromatography‐Mass Spectrometry (LC‐MS) Retention Time (RT) of these compounds on a Xterra MS C‐18 stationary phase. ANFIS and MLR methods were used as variable selection tools and a neural network was employed for predicting the RTs. Three descriptors of 3D‐MoRSE‐signal 06/weighted by atomic polarizabilities (Mor06p), Radial Distribution Function‐1.0/weighted by atomic van der Waals volumes (RDF010v), and number of functional groups of RCoN&lt;and &gt;NCN (N‐072) reveal the importance of dispersion interactions between benzodiazepines and C‐18 stationary phases and also electrostatic and hydrogen bond interactions of these compounds with the polar mobile phase. The superiority of 3‐4‐1 ANFIS‐Artificial Neural Networks (ANN) model over MLR‐ANN with a linear feature selection method was illustrated by Leave‐Multiple‐Out (LMO) cross‐validation method. Values of R2L6O=0.771 and RMSEL6O=4.792 for ANFIS‐ANN model should be compared with the values of R2L6O=0.632 and RMSEL6O=6.481 for the MLR‐ANN model. LMO‐CV and Y‐randomization results indicate the robustness and predictive ability of the generated model of ANFIS‐ANN and show the ability of ANFIS in selecting the features for phenomena with nonlinear characteristics. Sequential Zeroing of Weights (SZW) as a sensitivity analysis method showed the importance of dispersion interaction in the retention mechanism of benzodiazepines in LC method.

Quantitative Structure–Retention Relationship Study of a Variety of Compounds in Reversed‐Phase Liquid Chromatography: A PLS‐MLR‐STANN Approach

AbstractA quantitative structure–retention relationships model has been developed to study the retention behavior of 87 aliphatic and aromatic compounds in Reversed‐Phase Liquid Chromatography (RPLC) on five bonded‐phase columns differing in silanol group acidity. Six numerical descriptors of Molecular Mass (M), partial charge of the most negative atom (NPCH), partial charge of the most positive hydrogen (PCHH), van der Waals volume (VOLUME), Dipole Moment (DIMO), and Highest Occupied Molecular Orbital (HOMO) have been calculated for each compound. A separate Multiple Linear Regression (MLR) model has been developed using the six descriptors for each column. Partial Least Square (PLS) combined with MLR mean effects has been used for the mechanism interpretation. The descriptors of M, HOMO, and DIMO showed the largest PLS regression coefficients. It is found in the present work that solute size, n–π donor–acceptor and dipole–dipole interactions play a major role in the RPLC retention mechanism. Also, a self‐training artificial neural network was developed using the six descriptors as its inputs. The results obtained using this model are in good agreement with the experimental results of all five columns. Superiority of this model with low standard errors and high correlation coefficients for all five stationary phases reveals nonlinear characteristics of retention in RPLC. Successful application of the six descriptors in modeling of the retention behavior of different data sets using gas chromatography (previous work) and RPLC (present work) with different stationary phases reveals the robustness of the parameters.

Semi-empirical topological method for prediction of the gas chromatographic relative retention times of Polybrominated Diphenyl Ethers (PBDEs)

Quantitative structure-retention relationship (QSRR) studies have proved to be a valuable approach in the prediction of the gas chromatographic relative retention time (GC-RRT) of organic chemicals. Polybrominated diphenyl ether (PBDE) congeners are now ubiquitous environmental pollutants. Of the 209 possible PBDE congeners, 126 have been synthesized and their retention-time data on seven different stationary phases has been determined [Korytár et al.:J Chromatography A 1065:239-249, (2005)]. To estimate and predict the GC-RRT values of all 209 PBDEs on different stationary phases, 17 molecular descriptors from the semi-experience algorithm in MOPAC program and the topological structures of PBDE molecules were calculated. By means of the VSMP (variable selection and modeling based on prediction) program [Liu et al.:J Chem Inf Comput Sci 43:964-969, (2003)], six optimal descriptors were selected to develop a QSRR model for the prediction of GC-RRT of PBDE. The descriptors contain some energy information (such as the energy of the lowest unoccupied molecular orbital and highest occupied molecular orbital) and topological information (the number of ortho-, meta-, and para- substituted bromine atoms) as well as the molecular weight (lnM (W)). All the models developed were cross-validated using leave-one-out (LOO). For seven GC stationary phases, the estimated correlation coefficients (r(2)) are all more than 0.985 but for the column CP-Sil 19 (r(2) = 0.9392) and LOO-validated correlation coefficients (q(2)) all more than 0.985 but for the column CP-Sil 19 (q(2) = 0.9345).

Quantitative structure–retention relationship studies with immobilized artificial membrane chromatography: II: Partial least squares regression

We aimed to establish quantitative structure–retention relationship (QSRR) with immobilized artificial membrane (IAM) chromatography using easily understood and obtained physicochemical molecular descriptors and to elucidate which descriptors are critical to affect the interaction process between solutes and immobilized phospholipid membranes. The retention indices (log k IAM) of 55 structurally diverse drugs were determined on an immobilized artificial membrane column (IAM.PC.DD2) directly or obtained by extrapolation method for highly hydrophobic compounds. Ten simple physicochemical property descriptors (clog P, rings, rotatory bond, hydro-bond counting, etc.) of these drugs were collected and used to establish QSRR and predict the retention data by partial least squares regression (PLSR). Five descriptors, clog P, rotatory bond (RotB), rings, molecular weight (MW) and total surface area (TSA), were reserved by using the Variable Importance for Projection (VIP) values as criterion to build the final PLSR model. An external test set was employed to verify the QSRR based on the training set with the five variables, and QSRR by PLSR exhibited a satisfying predictive ability with R p = 0.902 and RMSE p = 0.400. Comparison of coefficients of centered and scaled variables by PLSR demonstrated that, for the descriptors studied, clog P and TSA have the most significant positive effect but the rotatable bond has significant negative effect on drug IAM chromatographic retention.

Quantitative structure-retention relationship studies using immobilized artificial membrane chromatography I: Amended linear solvation energy relationships with the introduction of a molecular electronic factor

Linear solvation energy relationships (LSERs) amended by the introduction of a molecular electronic factor were employed to establish quantitative structure-retention relationships using immobilized artificial membrane (IAM) chromatography, in particular ionizable solutes. The chromatographic indices, log k IAM, were determined by HPLC on an IAM.PC.DD2 column for 53 structurally diverse compounds, including neutral, acidic and basic compounds. Unlike neutral compounds, the IAM chromatographic retention of ionizable compounds was affected by their molecular charge state. When the mean net charge per molecule ( δ) was introduced into the amended LSER as the sixth variable, the LSER regression coefficient was significantly improved for the test set including ionizable solutes. The δ coefficients of acidic and basic compounds were quite different indicating that the molecular electronic factor had a markedly different impact on the retention of acidic and basic compounds on IAM column. Ionization of acidic compounds containing a carboxylic group tended to impair their retention on IAM, while the ionization of basic compounds did not have such a marked effect. In addition, the extra-interaction with the polar head of phospholipids might cause a certain change in the retention of basic compounds. A comparison of calculated and experimental retention indices suggested that the semi-empirical LSER amended by the addition of a molecular electronic factor was able to reproduce adequately the experimental retention factors of the structurally diverse solutes investigated.

Quantitative structure-retention relationship study of the gas chromatographic retention indices of saturated esters on different stationary phases using novel topological indices

The gas chromatographic retention data of 90 saturated esters obtained isothermally at 150 °C on seven stationary phases (SE-30, OV-7, DC-710, OV-25, XE-60, OV-225 and Silar-5CP) are well-correlated with recently introduced Lu index and novel distance-based atom-type DAI topological indices using multiple linear regression (MLR) method. For the retention indices (RI) based on each stationary phase, the correlation coefficient r of the final models is larger than 0.994, and particularly, the decrease in the standard error is within the range of 53.0–64.8% as compared with the simple linear models with Lu index alone. The results indicate the efficiency of these indices in the structure–retention index correlations of complex compounds. The role of each of the molecular size and individual groups in the molecules is illustrated by analyzing the relative or fraction contributions of individual indices. The results also indicate that the molecular size makes a dominant contribution to retention behaviors, while some atomic type or groups have smaller influence dependent on the polarity of the column. The leave-one-out cross-validation demonstrates the final models to be statistically significant and reliable.