Cross-column density functional theory-based quantitative structure-retention relationship model development powered by machine learning

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 0.749 min 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.

Application of salting-out thin layer chromatography in computational prediction of minimum inhibitory concentration and blood-brain barrier penetration of some selected fluoroquinolones

The 2017 FDA safety review regarding the CNS (central nervous system) side effects associated with the systemic use of fluoroquinolones antibacterials (FQs) was the key motivation to carry out this work. The main objective of this study is to investigate lipophilicity and retention parameters of some selected fluoroquinolones antibacterials (FQs) namely; levofloxacin (LEV), ofloxacin (OFL), gatifloxacin (GAT), norfloxacin (NOR), sparfloxacin (SPA), ciprofloxacin (CIP) and lomefloxacin (LOM) using salting-out thin layer chromatography (SOTLC). Statistically significant correlations between the chromatographically-obtained retention parameters and experimental log P values were found and expressed as quantitative structure retention relationship (QSRR) equations. Principal component analysis was carried out to explain the variation between chromatographic and both experimental and computed lipophilicity parameters. In another aspect of this study, a comparison between the chromatographically-determined retention parameters (for five of the drugs under study) obtained using SOTLC (current study) and relative lipophilicity (R_M0) determined using a previously reported RP (reversed-phase)-TLC method was carried out. Statistically significant correlation between the two methods was found, although R_M0 values obtained using SOTLC was lower than those reported using RP-TLC. Multiple linear regression analysis was performed to predict MIC (minimum inhibitory concentration) and blood brain barrier (BBB) penetration of the examined drugs in which efficient QSAR (quantitative structure-activity relationship) and QSPR (quantitative structure-property relationship) models were generated using the calculated chromatographic parameters (R_M0 and C₀). The described models can provide a useful approach to predict MIC and BBB penetration of newly synthesized FQs targeting to increase their activity against Gram-positive organisms and to minimize the associated CNS side effects.