Metabolomics Navigator Tools, strategies & workflow guidance for metabolomics

Metabolite annotation and identification

Mass2SMILES

Mass2SMILES uses deep learning to predict candidate molecular structures and functional groups from tandem mass spectra.

Platform-independentPython packageSource available with restrictions
Access toolSuggest an update

Overview

Mass2SMILES uses deep learning to predict candidate molecular structures and functional groups from tandem mass spectra. The documented inference route reads GNPS-style MGF and uses containerized models with a separate molecular-descriptor component.

What can this tool do?

Structure elucidationSubstructure prediction

Common uses

  • Mass2SMILES uses deep learning to predict candidate molecular structures and functional groups from tandem mass spectra.

Scientific fit

Primary function

Metabolite annotation and identification

Other functions

Machine learning and prediction

Data compatibility

Inputs

Mass spectra

Outputs

Molecular structures

Input formats

GNPS-style MGF tandem spectra

Output formats

Candidate SMILES structures and functional-group predictions

Scope and considerations

Designed for

  • Mass2SMILES uses deep learning to predict candidate molecular structures and functional groups from tandem mass spectra.

Not designed for

  • Confirmed structural identification from a prediction alone or unrestricted redistribution of NIST-derived model resources.

Important requirements

  • Docker and downloaded model/container artifacts, compatible GPU/software settings for the chosen model release and a valid NIST license where required by the supplied artifacts.

Considerations

  • The project describes separate GPU model and CPU descriptor containers because of CUDA compatibility limitations. Prediction reliability depends on spectral preparation and training-domain coverage.

Access and documentation

Primary accessDocumentation

Public source and separately hosted artifacts are available; no repository-wide license grant was established. README states a valid NIST license is required for the supplementary model/container resources.

Evidence and review

Enveda intake rechecked against official repository documentation, package/license metadata and primary publication identity on 2026-10-27. Editorial source review only; no software execution or independent benchmark. See docs/enveda-expansion-review-2026-10-27-batch3.md.

Core publications

Resource papers, direct updates and tool-specific methods. Selection policy.

  1. Preprint · not peer reviewed Mass2SMILES: deep learning based fast prediction of structures and functional groups directly from high-resolution MS/MS spectra. bioRxiv preprint (2023). Preprint; not peer reviewed. doi:10.1101/2023.07.06.547963

Related tools

BAMSimilar function · BAM generates and ranks structural candidates for suspect metabolites by applying biotransformation rules to known anchor structures.BUDDY / msbuddySimilar function · Generate and rank molecular formula annotations for small-molecule mass-spectrometry queries.CANOPUSSimilar function · A SIRIUS-integrated machine-learning method and webservice for predicting compound classes directly from high-resolution MS/MS data without requiring an exact database structure match.CFM-IDSimilar function · Predict tandem mass spectra, annotate fragments and rank candidate structures for small molecules.

Entry history

Changes to this catalogue entry, not the tool's software release history. Acceptance and editorial review are separate.

  1. 2026-10-27 · First committed entry

    Add 45 source-reviewed Enveda tool entries in third expansion batch

    View change · Entry at this revision

Full GitHub history