Author/Editor		Pečnik, Klemen; Todorović, Vesna; Bošnjak, Maša; Čemažar, Maja; Kononenko, Igor; Serša, Gregor; Plavec, Janez
Title		The general explanation method with NMR spectroscopy enables the identification of metabolite profiles specific for normal and tumor cell lines
Type		članek
Vol. and No.		Letnik 19, št. 19
Publication year		2018
Volume		str. 2066-2071
ISSN		1439-7633 - Chembiochem : a European journal of chemical biology
Language		eng
Abstract		Machine learning models in metabolomics, despite their great prediction accuracy, are still not widely adopted due to lack of efficient explanation of their predictions. In this study we propose using the general explanation method to explain machine learning model's predictions in order to gain detailed insights into metabolic differences between biological systems. The method was tested on a dataset of 1H NMR spectra acquired on normal lung and mesothelial cell lines and their tumor counterparts. Initially, the random forests and artificial neural network models were applied to the dataset and achieved excellent prediction accuracy. The models' predictions were explained with the general explanation method which enabled identification of discriminating metabolic concentration differences between individual cell lines and enabled the construction of their specific metabolic concentration profiles. This intuitive and robust method holds great promises for in-depth understanding of the mechanisms that underline phenotypes as well as biomarker discovery in complex diseases.
Keywords		NMR machine learning the general explanation method NMR spectroscopy