Documented workflows
Published software workflows and their original methodological context. Check current software documentation for version-specific changes.
Compound-class prediction with CANOPUS
CANOPUS uses the molecular fingerprint predicted by CSI:FingerID and a deep neural network to assign compound classes and chemical ontologies. It is designed to provide class-level structural information for unknown metabolites, including compounds lacking matching reference spectra or database structures.
- Annotate the precursor formula
Use SIRIUS to obtain a plausible molecular formula and fragmentation-tree representation for each feature.
- Predict the molecular fingerprint
Run CSI:FingerID fingerprint prediction as the structural representation used by CANOPUS.
- Predict compound classes with CANOPUS
Generate class probabilities and ontology assignments without requiring an exact database search for the unknown structure.
- Interpret class-level patterns
Use predicted classes to summarize chemical diversity, compare groups, or prioritize unknown features for deeper structural work.
Requirements
- High-resolution MS/MS suitable for SIRIUS/CSI:FingerID processing; CANOPUS is a webservice-backed SIRIUS feature requiring current account/license access.
Considerations
- CANOPUS predictions are compound-class annotations rather than exact molecular structures.
- Class probabilities and ontology depth vary by spectrum and chemical space; downstream biological interpretation should reflect prediction uncertainty.
Dührkop K, Nothias L-F, Fleischauer M, et al. Systematic classification of unknown metabolites using high-resolution fragmentation mass spectra. Nature Biotechnology 39, 462–471 (2021).
Read the source methodOverview
CANOPUS uses molecular fingerprints predicted from tandem mass spectra to infer hierarchical compound classes. Because it predicts class membership without requiring the exact compound to exist in a spectral or structure database, it is particularly useful for characterizing unknown metabolite features when precise structure identification is not possible. CANOPUS is currently accessed as a webservice-backed component of SIRIUS.
What can this tool do?
Common uses
- Assign hierarchical chemical classes to MS/MS features lacking confident exact structure annotations.
- Summarize chemical diversity across large untargeted metabolomics datasets at class level.
- Provide structural context for features absent from spectral and molecular-structure reference databases.
Scientific fit
Primary function
Metabolite annotation and identification
Other functions
Machine learning and prediction
Analysis types
Untargeted metabolomics, Metabolite profiling
MS level
MS/MS
Acquisition
Data-dependent acquisition (DDA), Targeted MS/MS, General or unspecified acquisition
Ion mobility
Ion mobility not required
Data compatibility
Inputs
Mass spectra
Outputs
Compound list or annotations, Human-readable report
Input formats
mzML, mzXML, MGF, MSP, SIRIUS project with predicted fingerprints
Output formats
CSV or TSV, JSON, SIRIUS class prediction tables
Scope and considerations
Designed for
- Compound-class prediction from accurate-mass tandem mass spectra.
Not designed for
- Confirmation of an exact compound structure or discrimination of stereoisomers.
Important requirements
- SIRIUS-compatible high-resolution MS/MS data and access to the SIRIUS webservice.
Considerations
- CANOPUS predicts compound classes rather than exact structures; results should not be reported as compound-level identifications.
- Class confidence and resolution depend on the information contained in the spectrum and the trained classification hierarchy.
- Inspect predicted class probabilities and hierarchy alongside the spectrum; retain the SIRIUS project and class export.
Access and documentation
CANOPUS is accessed through SIRIUS web services and is currently free for academic use under the SIRIUS licensing terms.
Evidence and review
Documentation-based editorial pass on 6 September 2026: scope, interpretation limits and practical use reviewed against the linked official documentation and cited literature. Software execution, service availability and developer verification were not assessed.
Core publications
Resource papers, direct updates and tool-specific methods. Selection policy.
- Dührkop K, Nothias LF, Fleischauer M, et al. Systematic classification of unknown metabolites using high-resolution fragmentation mass spectra. Nat Biotechnol 39, 462-471 (2021). doi:10.1038/s41587-020-0740-8
Related tools
Entry history
Changes to this catalogue entry, not the tool's software release history. Acceptance and editorial review are separate.
- 2026-09-08 · Entry update
Audit core publications across existing tools
View change · Entry at this revision - 2026-09-07 · Entry update
Expand and document 31 metabolomics tools and resources
View change · Entry at this revision - 2026-08-24 · First committed entry
Complete GNPS SIRIUS and MZmine tool ecosystems
View change · Entry at this revision