FeatureMAP Tutorial

Author

Yuting Wang, Xiaotao Shen

Published

June 6, 2026

Preface

FeatureMAP: A Feature-based Metabolite Functional Module Identification and Annotation Pipeline

FeatureMAP (featuremap) is a feature-centric framework for untargeted metabolomics that identifies biologically meaningful metabolic modules directly from metabolomic data, rather than relying only on predefined pathways or fully confirmed metabolite identities. It is developed by the Shen Lab at Nanyang Technological University, Singapore.

This tutorial presents a five-step workflow: data preparation followed by the four FeatureMAP analysis functions shown when the package loads:

  1. Data preparation: prepare a valid feature table and optional MS2 annotation table for FeatureMAP.
  2. Feature annotation: annotate raw LC-MS features to metabolite candidates using MS1 (and optional MS2) information.
  3. Subnetwork extraction: construct a robust prize-collecting Steiner forest (PCSF)-based metabolic subnetwork.
  4. Module identification: detect functional modules by community detection and evaluate module significance.
  5. Module annotation: interpret significant modules with database knowledge, optional PubMed evidence, and LLM reasoning, either in a fast two-call mode or in an advanced multi-agent debate mode that can also synthesize the annotated modules into one study-level picture.

This workflow helps convert feature-level metabolomics signals into interpretable biological modules in a reproducible and analysis-ready format.

Aim of This Tutorial

This guide walks you through the two faces of FeatureMAP: the R package for command-line (code-based) workflow and the Shiny App/Web version for point-and-click exploration. In the end, you will be able to:

  1. Prepare a valid feature table for FeatureMAP.
  2. Run the complete five-step FeatureMAP workflow with R code.
  3. Complete the same analysis with the FeatureMAP Shiny App/Web interface.
  4. Understand and inspect key result objects and module-level interpretations.
  5. Export interactive HTML reports for the module debate traces and the cross-module synthesis network.

Whether you favour the command line or a browser, this tutorial gives you everything you need to apply FeatureMAP confidently in your metabolomics projects.

📥 Contact Us