2024/08/12 by Xianghu Wang, Michael Strobel, Allegra T Aron +16
Biochemistry, Genetics and Molecular Biology · Chemistry · Computer Science · Mathematics · #Artificial intelligence #Bioinformatics #Bioinformatics and Genomic Networks #Biology #Chemical space #Chemistry #Computational Drug Discovery Methods #Computational biology #Computer network #Computer science #Data mining #Drug discovery #Mass spectrometry #Mathematics #Metabolomics and Mass Spectrometry Studies #Network topology #Pairwise comparison #Snapshot (computer storage) #Tandem mass spectrometry #Topology (electrical circuits) #Transitive relation #Visualization
paper · doi:10.1021/jasms.4c00208
openalex publication_date 2024/08/12 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/06
Untargeted tandem mass spectrometry (MS/MS) is an essential technique in modern analytical chemistry, providing a comprehensive snapshot of chemical entities in complex samples and identifying unknowns through their fragmentation patterns. This high-throughput approach generates large data sets that can be challenging to interpret. Molecular Networks (MNs) have been developed as a computational tool to aid in the organization and visualization of complex chemical space in untargeted mass spectrometry data, thereby supporting comprehensive data analysis and interpretation. MNs group related compounds with potentially similar structures from MS/MS data by calculating all pairwise MS/MS similarities and filtering these connections to produce a MN. Such networks are instrumental in metabolomics for identifying novel metabolites, elucidating metabolic pathways, and even discovering biomarkers for disease. While MS/MS similarity metrics have been explored in the literature, the influence of network topology approaches on MN construction remains unexplored. This manuscript introduces metrics for evaluating MN construction, benchmarks state-of-the-art approaches, and proposes the Transitive Alignments approach to improve MN construction. The Transitive Alignment technique leverages the MN topology to realign MS/MS spectra of related compounds that differ by multiple structural modifications. Combining this Transitive Alignments approach with pseudoclique finding, a method for identifying highly connected groups of nodes in a network, resulted in more complete and higher-quality molecular families. Finally, we also introduce a targeted network construction technique called induced transitive alignments where we demonstrate effectiveness on a real world natural product discovery application. We release this transitive alignment technique as a high-throughput workflow that can be used by the wider research community.