Molecular Structure Discovery for Untargeted Metabolomics Using Biotransformation Rules and Global Molecular...

Area of Science:

• Metabolomics
• Mass Spectrometry
• Computational Chemistry

Background:

• Untargeted mass spectrometry is vital for metabolomics but suffers from low tandem mass spectra annotation rates.
• Accurate metabolite identification is crucial for understanding biological pathways and molecular underpinnings of life.

Purpose of the Study:

• To develop a novel data-driven approach to enhance metabolite annotation in untargeted mass spectrometry.
• To improve the identification of molecular structures from tandem mass spectra using biochemical reaction principles.

Main Methods:

• Introduction of the Biotransformation-based Annotation Method (BAM), a novel computational approach.
• Leveraging molecular structural similarities and biotransformation rules inherent in biochemical reactions.
• Applying biotransformation rules to known "anchor" molecules with high spectral similarity to hypothesize and rank structures for unknown "suspect" molecules.

Main Results:

• BAM demonstrated success in annotating query spectra within a large-scale molecular network.
• The method achieved a 24.2% success rate in assigning correct molecular structures to anchor-suspect cases.
• Significant advancement in metabolite annotation accuracy and throughput was observed.

Conclusions:

• BAM represents a significant advancement in metabolite annotation for untargeted mass spectrometry-based metabolomics.
• The data-driven, biotransformation-based approach effectively addresses the challenge of low spectral annotation rates.
• This method holds promise for accelerating discoveries in biological and biomedical research through improved metabolomic data interpretation.