MetaboAnalyst

What MetaboAnalyst is

MetaboAnalyst is a web-based suite of tools designed for comprehensive metabolomics data analysis, interpretation, and integration with other omics layers. Originally created by members of the Wishart Research Group at the University of Alberta and maintained by the Xia Lab at McGill University, it supports both targeted and untargeted metabolomics workflows across quantitative and high-resolution mass spectrometry experiments.

This database maps metabolite names and common synonyms onto standard cross-database identifiers, including the Human Metabolome Database (HMDB), PubChem, ChEBI, KEGG, and METLIN, enabling consistent cross-referencing across diverse experimental formats.

MetaboAnalyst addresses the challenge of moving from raw spectral files or compound concentration tables to biological conclusions. It incorporates modules for raw liquid chromatography-mass spectrometry (LC-MS) data processing, univariate and multivariate statistics, biomarker discovery, metabolic pathway analysis across more than 120 species, dose-response modeling, and causal inference.

How it is organised

The compound dictionary underlying MetaboAnalyst cross-indexes metabolite nomenclature, synonyms, and vendor-specific naming conventions from platforms such as RefMet, Metabolon, Biocrates, and The Metabolomics Innovation Centre (TMIC) against reference chemical databases including HMDB, PubChem, ChEBI, KEGG, and METLIN. Duplicate feature entries can be systematically handled through averaging or configurable user-selected rules including sum, median, maximum, or retaining the initial entry.

The analytical platform processes raw centroid spectra in open formats like mzML, mzXML, and mzData via MetaboAnalystR and the asari algorithm to perform peak picking, alignment, and annotation. For downstream functional exploration, it applies algorithms such as mummichog and Gene Set Enrichment Analysis (GSEA) to map mass-to-charge ratios and retention times directly to pathways, and leverages general linear models, multivariate projections, and two-sample Mendelian randomization for statistical and causal evaluation.

Key facts

What you can use it for

Metabolite identification and synonym mapping

Resolving compound lists or vendor-specific nomenclature from platforms such as Biocrates, Metabolon, and TMIC to standard HMDB, PubChem, ChEBI, KEGG, and METLIN identifiers.

Untargeted LC-MS spectral processing and pathway mapping

Processing centroid LC-MS1 and tandem MS2 spectra using algorithms like asari and mummichog to perform peak annotation and infer metabolic pathway activity without prior manual compound identification.

Multivariate biomarker evaluation

Assessing candidate biomarkers using receiver operating characteristic (ROC) curves, Elastic Net classification, partial least squares-discriminant analysis (PLS-DA), or Random Forests.

Dose-response modeling for risk assessment

Calculating benchmark doses from metabolomics features across repeated or continuous exposure curves to evaluate chemical risk.

Causal inference using mGWAS

Evaluating potential causal associations between genetically influenced metabolites and phenotypes using two-sample Mendelian randomization based on public genome-wide association study data.

Query it here

The research agent queries MetaboAnalyst directly — no account, no API key — and every answer links back to the record it came from. Ask in plain language and it writes the query.

Strengths and limits

Questions