MetaboApps Turns GNPS2 Into a No-Code Downstream Lab for Untargeted Metabolomics
Context and Core Event
On July 24, 2026, Nature Methods published a correspondence titled “Bridging complexity and accessibility in metabolomics with MetaboApps.” The piece, led by researchers including Helena Mannochio-Russo with corresponding authors Mingxun Wang and Pieter C. Dorrestein, is not another end-to-end pipeline claim. It documents a practical architectural move: modular Streamlit applications—MetaboApps—that sit on top of completed GNPS2 jobs and turn heavy molecular-networking outputs into guided, browser-based post-processing.
Untargeted liquid chromatography–mass spectrometry (LC–MS) already produces dense feature tables, tandem spectra, and molecular networks. GNPS and its successor platform GNPS2 made classical and feature-based molecular networking (FBMN) widely available, but the hard part often starts after the job finishes. Analysts still need to filter patterns, attach metadata, run statistics, map chemical classes, query drug or food exposomes, or stitch multi-omics signals. Those steps historically lived in ad hoc notebooks, local R/Python scripts, or specialist desktops—high skill, low reproducibility, and poor handoff across labs.
