Metabolomics Navigator Tools, strategies & workflow guidance for metabolomics

Machine learning and prediction

DeepION

DeepION learns compact representations of MSI ion images for finding co-localized ions or candidate isotope relationships.

Imaging mass spectrometrySpatial metabolomicsCommand lineSource available with restrictions
Access toolSuggest an update

Overview

DeepION learns compact representations of MSI ion images for finding co-localized ions or candidate isotope relationships. Separate COL and ISO modes use augmented images and learned embeddings for downstream similarity analysis.

What can this tool do?

Multivariate statistics

Common uses

  • DeepION learns compact representations of MSI ion images for finding co-localized ions or candidate isotope relationships.

Scientific fit

Primary function

Machine learning and prediction

Other functions

Visualization and reporting

Analysis types

Spatial metabolomics

Data compatibility

Inputs

Imaging data

Outputs

Statistical results

Input formats

Preprocessed MSI matrix, peak CSV, image dimensions and ion mode

Output formats

Learned image representations and co-localization/isotope search results

Scope and considerations

Designed for

  • DeepION learns compact representations of MSI ion images for finding co-localized ions or candidate isotope relationships.

Not designed for

  • Direct raw-imzML processing, isotopologue abundance correction or proof that co-localized ions belong to the same molecule.

Important requirements

  • The documented Python/PyTorch environment, a preprocessed pixel-by-ion matrix, peak list, image dimensions and ionization-mode settings.

Considerations

  • Image shape, preprocessing and missing pixels influence learned similarity. The documented legacy dependency combination needs compatibility checking before a new installation.

Access and documentation

Primary accessDocumentation

Public research source; no repository-wide license grant was established from the reviewed metadata. Check dependency and data terms separately.

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. DeepION: A Deep Learning-Based Low-Dimensional Representation Model of Ion Images for Mass Spectrometry Imaging. Analytical Chemistry (2024). doi:10.1021/acs.analchem.3c05002

Related tools

CRISPSimilar function · CRISP analyzes GC×GC-TOFMS contour images with modules for region selection/deep stacking, synthetic contour generation, image enhancement and classifier training.DeepGCN-RTSimilar function · DeepGCN-RT predicts small-molecule retention times from molecular graphs.IDSL_MINTSimilar function · IDSL_MINT provides transformer-based workflows connecting tandem mass spectra and molecular representations.massNetSimilar function · Learn spectral features and classify tissue regions in mass-spectrometry images.

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