A priori knowledge of the retention time of a given analyte simplifies the determination of separation conditions therefore quantitative structure retention relationship (QSRR) modelling might be considered a reasonable selection. The first problem in QSRR modelling is to select the most informative descriptors from among a large number of mutually correlated descriptors, while the second one is to build the core model of isocratic and/or gradient elution retention. A lot of conventional methods have been elaborated that are mainly based on different types of regression and simple variable selection methodology (i.e. principal component analysis), showing rather questionable prediction ability. This work reveals recent results on development of artificial intelligence (AI) hybrid methodology implementing all three AI paradigms: artificial neural networks, genetic algorithms and fuzzy logic. The developed models were fully optimized and validated with external set of compounds showing significant improvement of generalization ability. Furthermore, recent demands for increasing the productivity using the gradients, in combination with ever growing complexity of analyzed samples, are introducing an additional request on the analytical system – beside being fairly separated, the peaks are required be as “smoothly” shaped as possible to ensure their precise quantification. In other words, the analysts are becoming interested in peak shapes and peak shape modelling as well. This work also discusses recent developments is peak shape modelling based on QSRR modelling. The developed models are based on generalized logistic distribution and hybrid AI systems. The external validation results show promising predictive ability, but still indicate that there is much to be done before QSRR based optimization strategy could be efficiently built into a useful commercial software.
Presenting author:
Tomislav Bolanča
University of Zagreb, Faculty of Chemical Engineering and Technology, Marulićev trg 19, 10000 Zagreb, Croatia
This email address is being protected from spambots. You need JavaScript enabled to view it.
Authors:
Tomislav Bolanča - University of Zagreb, Faculty of Chemical Engineering and Technology, Marulićev trg 19, 10000 Zagreb, Croatia
Šime Ukić - University of Zagreb, Faculty of Chemical Engineering and Technology, Marulićev trg 19, 10000 Zagreb, Croatia
Mirjana Novak Stankov - University of Zagreb, Faculty of Chemical Engineering and Technology, Marulićev trg 19, 10000 Zagreb, Croatia
Marko Rogošić - University of Zagreb, Faculty of Chemical Engineering and Technology, Marulićev trg 19, 10000 Zagreb, Croatia
