|
Type |
Oral Presentation |
Area |
Oral Presentation of Young Analytical Chemists II |
Room No. |
Room 321 |
Time |
FRI 09:00-: |
Code |
ANAL2.O-1 |
Subject |
Prediction of chromatographic elution order using mathematical optimization in QSRR modelling as a means accurate characterization of complex protein mixture |
Authors |
Alham Alipuly, Petar Žuvela1, Jay Liu*, Tomasz Bączek2 Department of Chemical Engineering, Pukyong National University, Korea 1Department of Biomedical Engineering, National University of Singapore, Singapore 2Department of Pharmaceutical Chemistry, Medical University of Gdańsk, Poland |
Abstract |
High-performance liquid chromatography in the reverse-phase separation mode (RP-HPLC) accounts for more than 90 % of separations in modern analytical laboratories. Prediction of liquid chromatography (LC) retention time has become valuable, powerful and routine in method development. While numerous studies have been reported on prediction of retention time, studies on prediction of chromatographic elution order has been very sparse although it is one of the crucial steps in LC retention modelling and prediction. In this work, a first of its kind prediction of elution order is carried out. Elution order prediction is defined as a multi-objective optimization (MOO) problem in quantitative structure-retention relationships (QSRR) modelling with two different objective functions: root mean square error (RMSE) for predicting retention time and sum of ranking difference (SRD) for predicting elution order. For regression modelling, partial least squares (PLS) and artificial neural networks (ANN) are used. Results show that the proposed method as considerably increases computational time but gives more accurate prediction of retention order in case of some columns over a conventional QSRR model. In case of PLS as regression, the proposed method shows mostly incremental improvements of elution order for the Xterra column (tG-20min, T=40℃) with model error up to 10%. for the same column ANN shows considerable improvement over that of PLS. |
E-mail |
alipuy@pukyong.ac.kr |
|