Accelerating natural product discovery with linked MS-genomics and language/transformer-based models
Integrated chem-bio characterization of microbial strain libraries can streamline natural product discovery by prioritizing candidate producers. Here, we employ language- and transformer-based models to extract actionable insights from linked mass spectrometry (MS)-genome datasets. Our framework enables ranking of microbial producers to prioritise high-potential candidates for targeted validation. Across three representative case studies, this approach prioritized producers of diverse natural products with 75-100% precision. These findings demonstrate the transformative potential of AI-enabled chem-bio characterization to significantly accelerate natural product discovery and enable access to microbial chemical diversity beyond reference knowledge.
Authors
- Krishnan Adaikkappan
- Fong Tian Wong (ORCID: https://orcid.org/0000-0001-6145-5904)
- Shi Jun Ang (ORCID: https://orcid.org/0000-0002-3569-639X)
- Yee Hwee Lim (ORCID: https://orcid.org/0000-0002-7789-3893)
- Elena Heng
- Winston Koh (ORCID: https://orcid.org/0000-0003-4611-6490)
- Yu Hung Ng
- Zicong Marvin Wong (ORCID: https://orcid.org/0000-0003-1530-0340)
- Yi Wee Lim
- Dillon W P Tay
- Naythan Z X Yeo
Institutions
- Agency for Science, Technology and Research (SG)
- National University of Singapore (SG)
- Institute of High Performance Computing (SG)
- Bioinformatics Institute (SG)
- Institute of Molecular and Cell Biology (SG)
Publication Details
- Journal
- npj Antimicrobials and Resistance
- Published
- 2026-04-30
- DOI
- https://doi.org/10.1038/s44259-026-00206-7
- Primary Topic
- Microbial Natural Products and Biosynthesis
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Agency for Science, Technology and Research
- National Research Foundation Singapore