MetaboAnalystR 2.0 contains the R functions and libraries underlying the popular MetaboAnalyst web server, including > 500 functions for metabolomic data analysis, visualization, and functional interpretation. With version 2.0, we aim to address two important gaps left in its previous version. First, raw spectral processing - the previous version offered very limited support for raw spectra processing and peak annotation. Therefore, we have implemented comprehensive support for raw LC-MS spectral data processing including peak picking, peak alignment and peak annotations leveraging the functionality of the xcms (PMIDs: 16448051, 19040729, and 20671148; version 3.4.4) and CAMERA (PMID: 22111785; version 1.38.1) R packages. Second, we have enhanced support for functional interpretation directly from m/z peaks. In addition to an efficient implementation of the mummichog algorithm (PMID: 23861661), we have added a new method to support pathway activity prediction based on the well-established GSEA algorithm (PMID: 16199517). To demonstrate this new functionality, we provide the "MetaboAnalystR 2.0 Workflow: From Raw Spectra to Biological Insights" tutorial. In this tutorial, we perform end-to-end metabolomics data analysis on a subset of clinical IBD samples, showvasing the functionality of this package to infer biological insights directly from m/z features.