Quantitative Structure–Retention Relationships with Non-Linear Programming for Prediction of Chromatographic Elution Order

Liu Liu,Žuvela Žuvela,Wong Wong,Alipuly Alipuly,Bączek Bączek

doi:10.3390/ijms20143443

Abstract

In this work, we employed a non-linear programming (NLP) approach via quantitative structure–retention relationships (QSRRs) modelling for prediction of elution order in reversed phase-liquid chromatography. With our rapid and efficient approach, error in prediction of retention time is sacrificed in favor of decreasing the error in elution order. Two case studies were evaluated: (i) analysis of 62 organic molecules on the Supelcosil LC-18 column; and (ii) analysis of 98 synthetic peptides on seven reversed phase-liquid chromatography (RP-LC) columns with varied gradients and column temperatures. On average across all the columns, all the chromatographic conditions and all the case studies, percentage root mean square error (%RMSE) of retention time exhibited a relative increase of 29.13%, while the %RMSE of elution order a relative decrease of 37.29%. Therefore, sacrificing %RMSE(tR) led to a considerable increase in the elution order predictive ability of the QSRR models across all the case studies. Results of our preliminary study show that the real value of the developed NLP-based method lies in its ability to easily obtain better-performing QSRR models that can accurately predict both retention time and elution order, even for complex mixtures, such as proteomics and metabolomics mixtures.

Highlights

Quantitative structure–retention relationships (QSRRs) [1,2] modelling has become a de-facto standard for the prediction of retention time in reversed-phase liquid chromatography analysis, which accounts for >90% of separations in modern laboratories [3]
There are only a few studies in literature dealing with the problem of elution order prediction in reversed phase-liquid chromatography (RP-LC) including our previous work where we presented a multi-objective-optimization (MOO)-based method [4,5,6,7]
In our previous work [4], we have presented an MOO-based elution order prediction method using genetic algorithms (GA) [8,9] for optimization employing two QSRR models with a priori selected molecular descriptors related to the RP-LC retention mechanism

Summary

Introduction

Quantitative structure–retention relationships (QSRRs) [1,2] modelling has become a de-facto standard for the prediction of retention time in reversed-phase liquid chromatography analysis, which accounts for >90% of separations in modern laboratories [3]. In our previous work [4], we have presented an MOO-based elution order prediction method using genetic algorithms (GA) [8,9] for optimization employing two QSRR models with a priori selected molecular descriptors related to the RP-LC retention mechanism. The developed NLP-based method is directly implemented within the QSRR modelling process and was used for prediction of elution order of two (more simple) analytical mixtures: (i) analysis of 62 organic molecules on the Supelcosil LC-18 column; and (ii) analysis of 98 synthetic peptides on seven RP-LC columns with varied gradients and column temperatures. All the analytes predicted using the NLP-based QSRR elution order prediction method fall within their respective chemical domains of applicability In the first case study, the Supelcosil LC-18 column was used, whereas for the second case study: Xterra MS C18, LiChrospher RP-18, LiChrospher CN, Discovery HS F5-3, Discovery RP Amide C16, PLRP-S and Chromolith columns were used

QSRR Model Development

External Validation

Applicability Domain

Elution Order Prediction

Findings

Conclusions