In silico fragmentation evaluation

In silico fragmentation evaluation evaluates in silico MS/MS fragmentation algorithms by comparing theoretical fragmentations from candidate structures to experimental tandem mass spectra to improve compound identification in untargeted metabolomics.

Key Features:

• Comparative Analysis: Evaluates the performance of publicly available in silico fragmentation algorithms, including MetFragCL, CFM-ID, MAGMa+, and MS-FINDER.
• Optimization Strategies: Optimizes metadata usage, weighting factors, and combinations of methodologies to enhance identification accuracy.
• Integration of Tools: Integrates outcomes from multiple tools—specifically MAGMa+ and CFM-ID—and incorporates compound importance information alongside MS/MS matching.
• Performance Metrics: The combined approach achieved a 93% success rate for training data and 87% for challenge data in the 2016 CASMI challenge, versus 60% correct hits using MS/MS libraries alone.

Scientific Applications:

• Metabolomics: Improves compound identification rates in untargeted metabolomics to facilitate analysis of metabolic pathways and mechanisms.
• Biomarker Discovery: Enables more accurate identification of unknown compounds that may serve as biomarkers of biological states or diseases.
• Pharmacology: Provides fragmentation information for novel compounds to support structural characterization in drug discovery.

Methodology:

The method generates theoretical MS/MS fragmentations from target structures and compares them to experimental tandem mass spectra, leverages multiple in silico fragmentation algorithms (MetFragCL, CFM-ID, MAGMa+, MS-FINDER), and applies metadata weighting and compound-importance information to refine identifications.