Ranking and similarity for quantitative structure–retention relationship models in predicting Lee retention indices of polycyclic aromatic hydrocarbons

Quantitative structure–(chromatographic) retention relationship (QSRR) models for prediction of Lee retention indices for polycyclic aromatic hydrocarbons (PAHs) were gathered from the literature and the predictive performances of models were compared. Numerous Lee retention indices (46) were served as a reliable basis for ranking by a recently developed novel method of ordering based on the sum of ranking differences (SRD) [TrAC, Trends Anal. Chem. 29 (2010) 101–109], by which the best model can be selected easily. Two kinds of references for ranking were accepted, average (consensus) and the experimental retention indices. Leave-many-out cross validation of the SRD procedure provides an easy way to group similar models. Significant differences among models can be revealed by using Wilcoxon's matched pair test.     

Comparison of theoretical and experimental models for characterizing solvent properties using reversed phase liquid chromatography

Theoretical and experimental quantitative structure–retention relationships (QSRR) models are useful for characterizing solvent properties and column selectivity in reversed phase liquid chromatography (RPLC). The chromatographic behavior of a model analyte, the herbicide atrazine, in a system derived from nine organic solvents and three chromatographic columns was used for developing QSRR models. Multiple linear regression (MLR) and partial least squares regression (PLSR) were used as statistical approaches. The similarities and differences between linear solvation energy relationships (LSER), and semi-empirical and theoretical molecular models were demonstrated. QSRR models show high predictive power, and can successfully predict retention factor (log k) for new solvents. The models are useful for solvent optimization and reducing time for method development in RPLC. The herbicide atrazine can be readily analyzed at a low level, and all three columns provided good resolution, high-performance and symmetrical peaks. The method is suitable for analysis of atrazine in water samples.     

Development of Gradient Retention Model in Ion Chromatography. Part III: Fuzzy Logic QSRR Approach

In this paper, the authors tested methodology that overcame the most common limitation of quantitative structure-retention relationship (QSRR) models: their limited applicability at the specific conditions for which models were developed. The modeling was performed on ion chromatographic analysis of “wood sugars”. Adaptive neuro-fuzzy interference system, an advanced artificial intelligence regression tool, was applied in combination with genetic algorithm scanning to obtain good and reliable QSRR models. The obtained QSRR models were applied for predicting data that were required for further development of general isocratic and gradient retention models. All three developed models (QSRR, isocratic, and gradient) indicated good prediction ability with root mean square error of prediction ≤0.1557. The performances of the methodology were compared with those presented in previous research—namely genetic algorithm in combinations with—stepwise multiple linear regression, partial least squares, uninformative variable elimination–partial least squares, and artificial neural network regression.

     

Development of Gradient Retention Model in Ion Chromatography. Part I: Conventional QSRR Approach

New retention methodology that integrates the conventional quantitative structure-retention relationship (QSRR) approach and gradient retention modeling based on isocratic retention data is developed and presented in this paper. Such an integrated approach removes the general QSRR limitation of highly predefined application conditions (i.e., QSRR are generally applicable only under the conditions used during model development) and allows the prediction of retentions over a wide range of different elution conditions (practically for any isocratic or gradient elution profile). At the same time, it retains the ability to predict retention of components unknown to the model, i.e., the components that have not been used in modeling. Ion-exchange chromatography (IC) analysis of carbohydrates was selected as modeling environment. Three regression techniques were applied and compared during QSRR modeling, namely: stepwise multiple linear regression, partial least squares (PLS), and uninformative variable elimination–PLS regression. The obtained prediction results of the best QSRR model (root-mean-square error of prediction = 22.69 %) were similar to those found in the literature. The upgrade from QSRR to the integrated model did not diminish the predictive ability of the model, indicating an excellent potential of the developed methodology not only in IC but also in chromatography in general.     

烯烃气相色谱保留指数的QSRR研究

通过对184个烯烃类化合物在不同固定相不同柱温下的617个样本的气相色谱保留指数值(RI)与其部分参数:拓扑指数(mQ)、偶极矩(DPL)、固定液极性值(CP)及柱温(T)建立定量-色谱保留相关(QSRR)模型.分别利用多元线性回归(MLR)、偏最小二乘回归(PLSR)、人工神经网络(ANN)建模,同时采用内部及外部双重验证的办法对所得模型稳定性能进行深入分析和检验,建模计算值、留一法(LOO)交互检验(CV)预测值和外部样本的复相关系数Rcum,QLOO和Rext分别为0.999 2,0.998 4和0.999 2(MLR);0.999 0,0.998 0和0.999 1(PLSR);0.999 4,0.998 7和0.999 2(ANN).结果表明:所建定量结构保留关系(QSRR)模型具有良好的稳定性和预测能力,较好地揭示了烯烃类化合物在不同固定相不同柱温上气相色谱保留指数的变化规律.     

含氧化合物气相色谱保留指数的预测

通过对部分含氧化合物(醇、酯、醛、酮)在不同固定相不同柱温下的849个样本的气相色谱保留指数值(RI)与其部分参数:拓扑指数(mQ)、定位基参数(Sox)、固定液极性值(CP)及柱温(T)建立定量结构-色谱保留相关(QSRR)模型。分别利用多元线性回归(MLR)、偏最小二乘回归(PLSR)、人工神经网络(ANN)建模,同时采用内部及外部双重验证的办法对所得模型稳定性能进行深入分析和检验,建模计算值、留一法(LOO)交互检验(CV)预测值和外部样本预测值的复相关系数Rcum、QLOO和Rext分别为0.9832、0.9829和0.9836(MLR);0.9832、0.9830和0.9836(PLSR);0.9910、0.9909和0.9900(ANN)。结果表明:所建定量结构保留关系(QSRR)模型具有良好的稳定性和预测能力,较好地揭示了含氧化合物(醇、酯、醛、酮)在不同色谱条件下气相色谱保留指数的变化规律。     

Structural Characterization of Volatile Components of Rosa Banksiae Ait for Estimation and Prediction of Their Linear Retention Indices and Retention Times

The molecular electronegativity-distance vector （MEDV） was used to describe the molecular structure of volatile components of Rosa banksiae Ait, and QSRR model was built up by use of multiple linear regression （MLR）. Furthermore, in virtue of variable screening by the stepwise multiple regression technique, the QSRR models of 10 and 6 variables and linear retention index （LRI） 10, 7 and 6 varieables were built up by combinating MEDV with the Ultra2 column GC retention time （tR） of 53 volatile components of Rosa Banksiae Air. The multiple correlation coefficients （R） of modeling calculation values of QSRR model were 0.906, 0.906, 0.949, 0.943 and 0.949, respectively. The cross-verification multiple correlation coefficients （RCV） were 0.903, 0.904, 0.867, 0.901 and 0.904, respectively. The results show that the models constructed could provide estimation stability and favorable predictive ability.     

Boosted regression trees, multivariate adaptive regression splines and their two-step combinations with multiple linear regression or partial least squares to predict blood-brain barrier passage: a case study

The use of some unconventional non-linear modeling techniques, i.e. classification and regression trees and multivariate adaptive regression splines-based methods, was explored to model the blood-brain barrier (BBB) passage of drugs and drug-like molecules. The data set contains BBB passage values for 299 structural and pharmacological diverse drugs, originating from a structured knowledge-based database. Models were built using boosted regression trees (BRT) and multivariate adaptive regression splines (MARS), as well as their respective combinations with stepwise multiple linear regression (MLR) and partial least squares (PLS) regression in two-step approaches. The best models were obtained using combinations of MARS with either stepwise MLR or PLS. It could be concluded that the use of combinations of a linear with a non-linear modeling technique results in some improved properties compared to the individual linear and non-linear models and that, when the use of such a combination is appropriate, combinations using MARS as non-linear technique should be preferred over those with BRT, due to some serious drawbacks of the BRT approaches.