Skip to content



SIRIUS is an tool for the computational annotation of fragmentation spectra tool developed by the Böcker Lab at Jena University.

GNPS Workflow

Workflow description

We are working on deploying SIRIUS on GNPS to enable users to perform automated high throughput computational annotation. Various annotations using the SIRIUS modules:

  1. Molecular formula prediction with SIRIUS and ZODIAC.
  2. Structure assignment with CSI:FingerID and COSMIC.
  3. Compound class prediction with CANOPUS.

Checkout the beta workflow here.

For more information on SIRIUS, please refer to the official SIRIUS documentation.

With the Feature-Based Molecular Networking

The MGF file from the Feature-Based Molecular Networking workflow can be used as input for the SIRIUS workflow. The results can then be mapped on the molecular networks. When using MZmine, for best results, uses the SIRIUSExport module from the same aligned peaklist.

With the SIRIUS mzML parser

SIRIUS has now a built-in mzML parser. The SIRIUS workflow on GNPS allows the selection of one mzML file.


Note this support is currently experimental and under development!


Viewing the results

View Molecular Formula Summary: view molecular formula annotations results from SIRIUS and ZODIAC.

View Compound Summary: view structure assignment with CSI:FingerID and COSMIC.

View CANOPUS results: view compound class prediction with CANOPUS.

Mapping the results onto the molecular networks

If you used the Feature-Based Molecular Networking workflow, you can map SIRIUS annotation onto the molecular networks in Cytoscape.

A quick way to do this can be found in this Notebook. Simply enter the FBMN task and Sirius task and you'll get a merged graphml that can be dragged into Cytoscape.

Example Analysis

Checkout this analyis of this E Coli data found here. The steps we took were:

  1. Feature Finding with MZmine2
  2. Feature Based Molecular Networking
  3. Sirius in GNPS

These jobs cannot be cloned and they function as an example. Use this workflow to submit your new job. You will have to provided an .mgf file (exported from MZmine) and select what type of instrument (orbitrap or QToF) have been used for the analysis to set the ppm error.

Let us know what you think!


The GNPS workflow is brought to you Ming Wang and Louis Felix Nothias. If you found this workflow useful for your data analysis, please cite the relevant literature from below.

Main citations

  1. SIRIUS: K. Dührkop, et al., SIRIUS 4: a rapid tool for turning tandem mass spectra into metabolite structure information. Nat. Methods 16, 299–302 (2019).
  2. ZODIAC: M. Ludwig, et al., Database-independent molecular formula annotation using Gibbs sampling through ZODIAC. Nat. Machine Intelligence (2020).
  3. CSI:FingerID K. Dührkop, H. Shen, M. Meusel, J. Rousu, S. Böcker, Searching molecular structure databases with tandem mass spectra using CSI:FingerID. Proc. Natl. Acad. Sci. U. S. A. 112, 12580–12585 (2015).
  4. CANOPUS K. Dührkop, et al., Classes for the masses: Systematic classification of unknowns using fragmentation spectra. bioRxiv, 2020.04.17.046672 (2020).

Additional citations

  1. S. Böcker, M. C. Letzel, Z. Lipták, A. Pervukhin, SIRIUS: decomposing isotope patterns for metabolite identification. Bioinformatics 25, 218–224 (2009).
  2. S. Böcker, K. Dührkop, Fragmentation trees reloaded. J. Cheminform. 8, 5 (2016).

Page Contributors

Ming Wang (48.68%), lfnothias (46.05%), irinakoester (1.32%), Simone Zuffa (3.95%)

Last update: March 23, 2022 00:12:46