Fragment Ion Abundance-Based Molecular Networking (FIAMN) Enables Skeleton-Oriented Discovery of Diterpenoids from a Soft Coral.

Mass spectral molecular networking has become an indispensable tool for natural product discovery, yet its performance in distinguishing isomeric structures and recognizing unprecedented carbon skeletons remains challenging, especially for low-signal-to-noise ratio mass spectra. Herein, we introduce Fragment Ion Abundance-Based Molecular Networking (FIAMN), an advanced strategy that complements standard molecular networking by integrating mode-based spectral denoising, fragment ion abundance characterization, and cosine-spectral entropy dual similarity fusion. This integrated workflow enables more reliable skeleton-oriented clustering and preferential identification of novel scaffolds from complex mixtures. Application of FIAMN to the soft coral Sinularia australiensis afforded six structurally unique diterpenoids, simaustralenes A-F, representing four unprecedented carbon skeletons. Biomimetic photocyclization studies supported their proposed biosynthetic origins and further highlighted how precursor functional groups govern skeletal diversification. Notably, compound 6 exhibited potent antipulmonary fibrosis activity with efficacy comparable to the clinical drug Nintedanib. By enabling more reliable differentiation of isomeric and low-abundance metabolites, FIAMN provides a practical framework for expanding structurally distinct natural product chemical space through MS-guided discovery.