Artificial Neural Networks for Classification in Metabolomic Studies of Whole Cells Using H-1 Nuclear Magnetic Resonance

Warning

This publication doesn't include Faculty of Education. It includes Faculty of Science. Official publication website can be found on muni.cz.

Authors	BROUGHAM D F IVANOVA G. GOTTSCHALK M. COLLINS D M EUSTACE A J O CONNOR R. HAVEL Josef
Year of publication	2011
Type	Article in Periodical
Magazine / Source	Journal of Biomedicine and Biotechnology
MU Faculty or unit	Faculty of Science
Citation
Doi	http://dx.doi.org/10.1155/2011/158094
Field	Analytic chemistry
Keywords	NMR SPECTROSCOPIC DATA; HUMAN BRAIN-TUMORS; H-1-NMR SPECTROSCOPY; PATTERN-RECOGNITION; IN-VITRO; EXTRACTS; IDENTIFICATION; LINES; VIVO; RESISTANCE
Description	We report the successful classification, by artificial neural networks (ANNs), of H-1 NMR spectroscopic data recorded on whole-cell culture samples of four different lung carcinoma cell lines, which display different drug resistance patterns. The robustness of the approach was demonstrated by its ability to classify the cell line correctly in 100% of cases, despite the demonstrated presence of operator-induced sources of variation, and irrespective of which spectra are used for training and for validation. The study demonstrates the potential of ANN for lung carcinoma classification in realistic situations.
Related projects:	Centre of Biophysical Chemistry, Bioelectrochemistry and Bioanalysis. New Tools for Genomics, Proteomics and Biomedicine