Analysis of stable isotope assisted metabolomics data acquired by high resolution mass spectrometry

Abstract

Stable isotope assisted metabolomics (SIAM) uses stable isotope tracers to support studies of biochemical mechanisms. We report a suite of data analysis algorithms for automatic analysis of SIAM data acquired on a high resolution mass spectrometer. To increase the accuracy of isotopologue assignment, metabolites detected in the unlabeled samples were used as reference metabolites to generate possible isotopologue candidates for analysis of peaks detected in the labeled samples. An iterative linear regression model was developed to deconvolute the overlapping isotopic peaks of isotopologues present in a full MS spectrum, where the threshold for the weight factor was determined by a simulation study assuming different levels of Gaussian white noise contamination. A normalization method enabling isotope ratio-based normalization was implemented to study the difference of isotopologue abundance distribution between sample groups. The developed method can analyze SIAM data acquired by direct infusion MS and LC-MS, and can handle metabolite tracers containing different tracer elements. Analysis of SIAM data acquired from mixtures of known compounds showed that the developed algorithms accurately identify metabolites and quantify stable isotope enrichment. Application of SIAM data acquired from a biological study further demonstrated the effectiveness and accuracy of the developed method for analysis of complex samples.