Network Topology Evaluation and Transitive Alignments for Molecular Networking

Area of Science:

• Analytical Chemistry
• Computational Chemistry
• Bioinformatics

Background:

• Untargeted tandem mass spectrometry (MS/MS) is vital for analyzing complex chemical samples.
• High-throughput MS/MS generates large datasets, posing interpretation challenges.
• Molecular Networks (MNs) organize and visualize MS/MS data but their construction needs optimization.

Purpose of the Study:

• To evaluate network topology's influence on Molecular Network construction.
• To introduce novel metrics for assessing MN quality.
• To propose and validate the Transitive Alignments approach for enhanced MN construction.

Main Methods:

• Developed metrics to evaluate Molecular Network construction.
• Benchmarked existing state-of-the-art MN approaches.
• Introduced the Transitive Alignments technique, leveraging network topology.
• Combined Transitive Alignments with pseudoclique finding for improved molecular families.
• Developed induced transitive alignments for targeted network construction.

Main Results:

• Transitive Alignments effectively realigns MS/MS spectra of related compounds with multiple structural differences.
• The combination of Transitive Alignments and pseudoclique finding yielded more complete and higher-quality molecular families.
• Induced transitive alignments demonstrated effectiveness in a natural product discovery application.

Conclusions:

• The Transitive Alignments approach significantly enhances Molecular Network construction for untargeted MS/MS data.
• This method improves the identification of related compounds, metabolic pathways, and potential biomarkers.
• The developed workflow is released as a high-throughput tool for the research community.