What does this strategy enable?
The 2018 method integrates learned pairwise retention preferences with spectral evidence rather than filtering each feature independently by absolute retention time.
Transferable workflow
Workflow
- 1
Prepare candidate lists
Associate measured retention times with spectra and candidate structures.
- 2
Score spectral candidates
The original study uses IOKR scores.
IOKR - 3
Learn and integrate retention preferences
Combine RankSVM order predictions and spectral scores using the published dynamic-programming procedure.
- 4
Review ranked assignments
Inspect candidates and seek independent confirmation.
Connected resources
Tools used in this strategy
Evidence of reuse
Published implementations and adaptations
These papers applied, adapted, extended, benchmarked or formalized the Strategy after its originating publication. They are examples of precedent, not a ranking of software choices.
ROASMI: accelerating small molecule identification by repurposing retention data
Uses learned molecular representations and a neural ranking model to predict retention order within a chromatographic application domain, then integrates this evidence with MS/MS scores. This is a later implementation, not the original 2018 workflow.
Navigator tools used: ROASMI
Joint structural annotation of small molecules using liquid chromatography retention order and tandem mass spectrometry data
LC-MS2Struct learns structured integration of spectral scores and retention order, including stereochemistry-aware fingerprints. This replaces the original integration model rather than reproducing its algorithm.
Navigator tools used: LC-MS2Struct
Strategy profile
Components
Chromatography, Annotation or identification, Statistical analysis
Scope and considerations
Designed for
- Joint interpretation of LC-MS/MS features.
Important requirements
- Candidate structures, spectral scores, measured retention order and suitable training data.
Considerations
- Candidate coverage and chromatographic transferability matter. Rankings do not confirm identities.
Reusable resources
Publication
Liquid-chromatography retention order prediction for metabolite identification