Development of Gradient Retention Model in Ion Chromatography. Part II: Artificial Intelligence QSRR Approach
- 262 Downloads
Methodology that integrates traditional QSRR modeling with transfer of information from isocratic to gradient environment was presented in the previous paper as an efficient new chromatographic approach. The previous research included application of conventional regression techniques that resulted in relatively high prediction errors. Also, it was shown that prediction error of the integrated model was mostly caused by prediction of the QSRR models. Therefore, artificial intelligence was applied in this work in order to improve the prediction ability of QSRR-based model: Artificial neural networks were selected for QSRR modeling, while genetic algorithm was used for the selection of optimal descriptors. Both artificial neural networks and genetic algorithm were optimized in order to build accurate and reliable QSRR models. Selection function, crossover function, and percentage of genes’ mutations were varied in case of genetic algorithm, while artificial neural networks were optimized by means of different network type, training algorithm and number of neurons in hidden layer. During retention modeling, basic QSRR models developed for specific eluent strength were upgraded to isocratic, and thereafter to gradient retention model. None of the three developed models showed systematic errors, and the obtained predictions (RMSEP 11.66, 10.67, and 7.10 %, respectively) indicated significant improvement from the results presented in previous paper.
KeywordsIon chromatography QSRR Artificial intelligence Gradient retention model
Generous support and help from Thermo Fisher Scientific Corporation is gratefully acknowledged.
- 1.Kaliszan R (1997) Structure and retention in chromatography. A chemometric approach. Harwood Academic Publishers, AmsterdamGoogle Scholar
- 11.Nasal A, Payer K, Haber P, Forgacs E, Cserhati T, Kaliszan R (1998) LC GC Int 11:240–252Google Scholar
- 18.Ukić Š, Novak M, Žuvela P, Avdalović N, Liu Y, Buszewski B, Bolanča T (2014) Chromatographia. doi: 10.1007/s10337-014-2653-5
- 24.Haykin S (1994) Neural networks: a comprehensive foundation. Macmillan, New YorkGoogle Scholar
- 26.Kröse B, van der Smagt P (1996) An introduction to neural networks, 5th edn. The University of Amsterdam, AmsterdamGoogle Scholar
- 27.Unay D (2006) Multispectral image processing and pattern recognition techniques for quality inspection of apple fruits. Presses universitaires de Louvain, Louvain-la-NeuveGoogle Scholar
- 29.Thermo Fisher Scientific (2011) Product manual CarboPac PA20. http://www.dionex.com/en-us/webdocs/4378-Man-031884-05-CarboPac-PA20-Jul11.pdf. Accessed 14 Jan 2014
- 30.Basumallick L, Rohrer J (2012) Thermo Fisher Scientific Application Note 282. http://www.dionex.com/en-us/webdocs/113489-AN282-IC-Biofuel-Sugars-03May2012-LPN2876-R2.pdf. Accessed 14 Jan 2014
- 32.Ukić Š, Rogošić M, Novak M, Šimović E, Tišler V, Bolanča T (2013) J Anal Methods Chem 2013. doi: 10.1155/2013/549729