Key points are not available for this paper at this time.
Metabolomics, similarly to other high-throughput “-omics” techniques, generates large arrays of data, whose analysis and interpretation can be difficult and not always straightforward. Several software for the detailed metabolomics statistical analysis are available, however there is a lack of simple protocols guiding the user through a standard statistical analysis of the data. Herein we present “muma”, an R package providing a simple step-wise pipeline for metabolomics univariate and multivariate statistical analyses. Based on published statistical algorithms and techniques, muma provides user-friendly tools for the whole process of data analysis, ranging from data imputation and preprocessing, to dataset exploration, to data interpretation through unsupervised/supervised multivariate and/or univariate techniques. Of note, specific tools and graphics aiding the explanation of statistical outcomes have been developed. Finally, a section dedicated to metabolomics data interpretation has been implemented, providing specific techniques for molecular assignments and biochemical interpretation of metabolic patterns. muma is a free, user-friendly and versatile tool suite tailored to assist the user in the interpretation of metabolomics data in the identification of biomarkers and in the analysis of metabolic patterns. Keywords: Chemometrics, metabonomics, metabolic pattern, multivariate analysis, R package, statistical analysis, univariate analysis.
Gaude et al. (Fri,) studied this question.