Computational data mining method for isotopomer analysis in the quantitative assessment of metabolic reprogramming

Abstract

Measurement of metabolic flux levels using stable isotope labeling has been successfully used to investigate metabolic redirection and reprogramming in living cells or tissues. The metabolic flux ratio between two reactions can be estimated from the 13C-labeling patterns of a few metabolites combined with the knowledge of atom mapping in the complicated metabolic network. However, it remains unclear whether an observed change in the labeling pattern of the metabolites is sufficient evidence of a shift in flux ratio between two metabolic states. In this study, a data analysis method was developed for the quantitative assessment of metabolic reprogramming. The Metropolis-Hastings algorithm was used with an in silico metabolic model to generate a probability distribution of metabolic flux levels under a condition in which the 13C-labeling pattern was observed. Reanalysis of literature data demonstrated that the developed method enables analysis of metabolic redirection using whole 13C-labeling pattern data. Quantitative assessment by Cohen’s effect size (d) enables a more detailed read-out of metabolic reprogramming information. The developed method will enable future applications of the metabolic isotopomer analysis to various targets, including cultured cells, whole tissues, and organs.