Use of MetaboAnalyst as a tool to study metabolomics data for a dietary intervention study

Background Metabolomics is the comprehensive analysis of metabolites (<1500 Da). MetaboAnalyst (MA) is web‐based, publicly available software for analysis of quantitative metabolomics data. Objective To demonstrate the use of MA in a pilot study evaluating the effect of diet change on the human plasma metabolome. Methods Plasma collected from healthy adult females before and after 5 days of receiving a high‐fat, low‐carbohydrate, and adequate protein diet were analyzed using HPLC‐ESI‐TOF‐MS. Features (mass/retention time) and abundances were uploaded to MA for analysis. PCA and PLS‐DA were used to differentiate between groups. Features were selected based on significance, using a volcano plot (p of 0.1 and fold change of 2). The remaining features were filtered based on presence in samples. A heatmap was used to identify clusters of features. Results The volcano plot showed that less than 5% of the features were significant. The number of features was reduced from 2225 to 81 and 48 of these features were present in 50% of the subjects. Using these targeted 48 features, the PCA, PLS‐DA, and heatmap showed improved clustering of subjects based on diet group compared to the global analysis. Conclusions MA can be used to evaluate the effects of a diet change on the plasma metabolome. Supported in part by the National Institutes of Health (NIH) and National Center for Research Resources (NCRR) CTSA grant 1UL 1RR029890.