What does this strategy enable?
Taxonomically informed scoring adds biological provenance to conventional candidate ranking. Candidate structures are first generated from spectral libraries or in-silico annotation. Known organism–compound associations are then used to calculate taxonomic relevance, allowing candidates reported from related organisms to receive additional support without treating taxonomy as structural proof.
Transferable workflow
Workflow
- 1
Generate candidate structural annotations
Use spectral-library or in-silico annotation to generate ranked candidate structures for MS/MS features.
- 2
Retrieve candidate biological-source information
Collect taxonomic provenance for candidate natural products from curated structure–organism resources.
- 3
Calculate taxonomic relevance
Compare the taxonomy of the analyzed organism with reported producers of each candidate and convert the distance into an annotation-support score.
- 4
Rerank candidates
Combine chemical and taxonomic evidence while retaining the original annotation evidence and uncertainty.
Connected resources
Tools used in this strategy
Strategy profile
Related objectives
Natural-product discovery, Dereplication
Components
Data acquisition, Preprocessing and feature extraction, Annotation or identification, Metadata analysis, Statistical analysis
Biological context
Plant, Microbial, Fungal
Scope and considerations
Designed for
- Improving candidate ranking when reliable organism identity and candidate source-occurrence information are available.
Important requirements
- A taxonomically identified source organism.
- Candidate structures with linked biological-source metadata.
Considerations
- Taxonomy is supporting evidence and cannot establish structure.
- Published occurrence databases are incomplete and biased toward studied taxa.
- The linked contemporary resources illustrate ways to implement the strategy; they are not all software used in the original publication.
Reusable resources
Publication
Rutz A, Dounoue-Kubo M, Ollivier S, et al. Taxonomically Informed Scoring Enhances Confidence in Natural Products Annotation. Frontiers in Plant Science 10, 1329 (2019).