Retention Index Prediction Using Quantitative Structure-Retention Relationships for Improving Structure Identification in Nontargeted Metabolomics.

Quantitative structure-retention relationship (QSRR) modeling has emerged as an efficient alternative to predict analyte retention times using molecular descriptors. However, most reported QSRR models are column-specific, requiring separate models for each high-performance liquid chromatography (HPLC) system. This study evaluates the potential of machine learning (ML) algorithms and quantum mechanical (QM) descriptors to develop QSRR models that can predict retention times across three different reversed-phase HPLC columns under varying conditions. Four machine learning methods-partial least squares (PLS) regression, ridge regression (RR), random forest (RF), and gradient boosting (GB)-were compared on a dataset of 360 retention times for 15 aromatic analytes. Molecular descriptors were calculated using density functional theory (DFT). Column characteristics like particle size and pore size and experimental conditions like temperature and gradient time were additionally used as descriptors. Results showed that the GB-QSRR model demonstrated the best predictive performance, with Q2 of 0.989 and root mean square error of prediction (RMSEP) of 0.749min on the test set. Feature analysis revealed that solvation energy (SE), HOMO-LUMO energy gap (∆E HOMO-LUMO), total dipole moment (Mtot), and global hardness (η) are among the most influential predictors for retention time prediction, indicating the significance of electrostatic interactions and hydrophobicity. Our findings underscore the efficiency of ensemble methods, GB and RF models employing non-linear learners, in capturing local variations in retention times across diverse experimental setups. This study emphasizes the potential of cross-column QSRR modeling and highlights the utility of ML models in optimizing chromatographic analysis.

Developing green liquid chromatography (LC) methods is currently a cutting-edge field due to the recognized negative influence of toxic organic solvents on human health and nature. One of the available strategies for “greening” is the addition of cyclodextrin (CD) in the mobile phase. CDs are known for their ability to form inclusion complexes with various analytes, improving their solubility in the aqueous phase, thus reducing the retention times and, simultaneously, the organic solvent consumption. The presented concept could be extended to incorporate sustainability in the field of separation science, from the stage of method development to routine analysis, through the quantitative structure retention relationship (QSRR) modeling approach. This book chapter is aimed at presenting the possibilities of building and employing QSRR models in CD-modified high-performance liquid chromatography (HPLC), followed by investigating their potential utility in different areas of application. The complexity of CD-modified HPLC indicates the associated complexity of the QSRR modeling. This review provides an overview of CD structures and analytical techniques applied for characterization of the formed inclusion complexes. Special focus is placed on the practical and theoretical knowledge regarding the retention mechanisms and equilibria existing in CD-modified HPLC systems as an introduction to retention modeling. Molecular descriptors mostly applicable in modeling retention in HPLC, as well as complex association constants that can properly describe the formed inclusion complexes in CD-modified HPLC, are also presented. Further, machine learning techniques are discussed in terms of their utility and contribution in building good predictive QSRR models. Finally, the key experimental findings about the utility of the obtained QSRR models in retention prediction and evaluation of optimal conditions of the chromatographic method, and then thermodynamic parameters of complexation are presented, followed by theoretical discussion of possible benefits, drawbacks, and future perspectives of the introduced approach.

Previous studies demonstrated that quantitative structure retention relationships (QSRR) combined with the linear solvent strength (LSS) model allow for prediction of gradient reversed phase high performance liquid chromatography (HPLC) retention time for any analyte of a known molecular structure under defined HPLC conditions. The QSRR model derived at the selected gradient time was tested at the same gradient time. Presently, in the first step, experimental retention data for model sets of just 5 analytes were used to derive appropriate QSRR models at two gradient times. Additionally, a new molecular modeling approach based on a more accurate ab initio method was here proposed. Those QSRR models were used to further predict gradient retention times for sets of 16 test analytes belonging to polynuclear aromatic hydrocarbons (PAHs), at two selected gradient times. Then, applying linear solvent strength (LSS) theory, only predicted retention times for PAHs were used to find the optimal gradient conditions to separate them. Satisfactory predictions of gradient retention times for PAHs were obtained. Contrary to the previous achievements, the proposed QSRR provides the chance to predict retention of PAHs with the appropriate selectivity achieving the same sequence of analytes eluted in the experiment and during the simulation performed on the computer screen.

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

Quantitative structure-retention relationship of selected imidazoline derivatives on α1-acid glycoprotein column

Correlation between chemical structure of non-congeneric solutes and their retention on polybutadiene-coated alumina

Reversed-phase high performance liquid chromatography (RP-HPLC) characteristics of some 9,10-anthraquinone derivatives using binary acetonitrile–water mixtures as mobile phase

Systematic studies were performed on an application of chemometrics to molecular modeling and the regularity of the retention index (RI ) of gas chromatography (GC). Molecular structures of hydrocarbons in straight-run gasoline were numerically coded. A set of structural parameters was accordingly obtained for the hydrocarbons and found correlate to their GC retention indices. A quantitative structure-retention relationship (QSRR) model (M1), with a correlation coefficient of R = 0.9901 and a standard deviation (SD) error of SD = 20.24, between the numeric structural codes and GC retention indices of 150 hydrocarbons was developed using multiple linear regression (MLR). If a “leave-one-out” cross-validation procedure was employed to construct a QSRR model for all samples, a second model (M2) with R = 0.9874 and SD = 22.87 was generated. The structural codes of hydrocarbons were tested with MLR for estimation and prediction of the GC RI by models M1 and M2, and the results obtained proved to be satisfactory.

Quantitative structure-retention relationship (QSRR) models were used to predict the retention time (t(R)) of mycotoxins and fungal metabolites. Heuristic method 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 5-21-1 RBFNN architecture using molecular descriptors calculated from molecular structure alone. The RBFNN model gave a square of correlation coefficient (R(2)) of 0.8709 and root mean square error of 1.2892 for the test set. This article provided a useful tool for predicting the t(R) of other mycotoxins when experiment data are unknown.

As a potent zinc chelator, hydroxamic acid has been applied in the design of inhibitors of zinc metalloenzyme, such as histone deacetylases (HDACs). A series of hydroxamic acids with HDAC inhibitory activities were subjected to the QSRR (Quantitative Structure-Retention Relationships) study. Experimental data in combination with calculated molecular descriptors were used for the development of the QSRR model. Specially, we employed PCA (principal component analysis) to accomplish dimension reduction of descriptors and utilized the principal components of compounds (16 training compounds, 4 validation compounds and 7 test compounds) to execute GA (genetic algorithm)-BP (error backpropagation) algorithm. We performed double cross-validation approach for obtaining a more convincing model. Moreover, we introduced molecular interaction-based features (molecular docking scores) as a new type of molecular descriptor to represent the interactions between analytes and the mobile phase. Our results indicated that the incorporation of molecular interaction-based features significantly improved the accuracy of the QSRR model, (R2 value is 0.842, RMSEP value is 0.440, and MAE value is 0.573). Our study not only developed QSRR model for the prediction of the retention time of hydroxamic acid in HPLC but also proved the feasibility of using molecular interaction-based features as molecular descriptors.

The hazardous psychoactive designer drugs are compounds in which part of the molecular structure of a stimulant or narcotic has been modified. A quantitative structure-retention relationship (QSRR) study based on a Levenberg-Marquardt artificial neural network (L-M ANN) was carried out for the prediction of the capacity factor (k′) of hazardous psychoactive designer drugs that contain Tryptamine, Phenylethylamine and Piperazine. The genetic algorithm-partial least squares (GA-PLS) method was used as a variable selection tool. A PLS method was used to select the best descriptors and the selected descriptors were used as input neurons in neural network model. For choosing the best predictive model from among comparable models, square correlation coefficient (R2) for the whole set is suggested to be a good criterion. Finally, to improve the results, structure-retention relationships were followed by nonlinear approach using artificial neural networks and consequently better results were obtained. Also this demonstrates the advantages of L-M ANN. This is the first research on the QSRR of the designer drugs using the GA-PLS and L-M ANN.

Quantitative Structure–Retention Relationship Modeling of Morphine and Its Derivatives on OV-1 Column in Gas–Liquid Chromatography Using Genetic Algorithm

In the study, 18 antiepileptic hydantoin analogues were investigated by means of reversed-phase HPLC on C-18 stationary phase and eluent acetonitrile-water. Quantitative structure-retention relationship (QSRR) study has been applied in order to understand factors that affect the retention which is closely correlated to the activity (ED₅₀ values). To overview the compounds for similarities and dissimilarities principal component analysis (PCA) has been applied. Six multiple linear regression models based on the most relevant descriptors were developed. Descriptors for MLR were selected according to variable importance calculated by partial least squares (PLS) analysis. Besides ALOGP the most important is aromatic ratio for mobile phases with more than 45% of acetonitrile, as well as electrotopological states when the % of acetonitrile is less than 40%. High agreement between experimental and predicted retention, obtained in the validation procedure, indicated the good quality of the derived QSRR models. For individual linear models, crossvalidation squared correlation coefficients (Q²) ranging from 0.697 to 0.837 were obtained. The residual values (difference between observed and calculated) agreed well within experimental error. Additionally, models were compared in terms of the smallest residual value by recently developed method of ranking based on the sum of ranking differences (SRD).

Liquid chromatographic retention time prediction models to secure and improve the feature annotation process in high-resolution mass spectrometry

Accurate prediction of the separation conditions for a set of target analytes with no retention data available is fundamental for routine analytical assays but remains a very challenging task. In this paper, a quality by design (QbD) optimisation workflow capable of discovering the optimal chromatographic conditions for separation of new compounds in hydrophilic-interaction liquid chromatography (HILIC) is introduced. This workflow features the application of quantitative structure−retention relationship (QSRR) methodology in conjunction with design of experiments (DoE) principles and was used to carry out a two-level full factorial DoE optimisation for a mixture of pharmaceutical analytes on zwitterionic, amide, amine, and bare silica HILIC stationary phases, with mobile phases containing varying acetonitrile content, mobile phase pH, and salt concentration. A dual-filtering approach that considers both retention time (tR) and structural similarity was used to identify the optimal set of analytes to train the QSRR in order to maximise prediction accuracy. Highly predictive retention models (average R2 of 0.98) were obtained and statistical analysis of the prediction performance of the QSRR models demonstrated their ability to predict the retention times of new compounds based solely on their molecular structures, with root-mean-square errors of prediction in the range 7.6–11.0 %. Further, the obtained retention data for pharmaceutical test compounds were used to compute their separation selectivity, which was used as input into a DoE optimiser in order to select the optimal separation conditions. Experimental separations performed under the chosen optimal working conditions showed good agreement with the theoretical predictions. To the best of our knowledge, this is the first study of a QbD optimisation workflow assisted with dual-filtering-based retention modelling to facilitate the method development process in HILIC.