Biosynthetic Gene Cluster Mining for Natural Product Discovery

Biosynthetic gene cluster mining has emerged as a cornerstone of modern natural product discovery, harnessing advances in genome sequencing, bioinformatics and machine learning to reveal the vast chemical potential encoded in microbial and plant genomes. Biosynthetic gene clusters (BGCs) are contiguous sets of genes that collectively encode the enzymatic machinery for the assembly, tailoring and regulation of specialised metabolites. Mining these clusters enables the prediction of novel chemical scaffolds, dereplication of known compounds and prioritisation of targets for experimental characterisation. Integrative approaches now combine genome mining with metabolomic data—particularly tandem mass spectrometry—to link genetic information with observed metabolites. Large‐scale repositories of predicted BGCs and standardised data schemas facilitate comparative analysis across thousands of genomes, revealing patterns of diversity, evolutionary relationships and ecological distribution. The global significance of this field is underpinned by its applications in antibiotic discovery, agrochemical development and microbiome research, as well as its role in unravelling chemical ecology and microbial interactions.

Research from Nature Portfolio

Recent studies have introduced an Atlas of hypothetical natural product structures by integrating large‐scale genome mining with public mass spectral datasets. A machine‐learning tool for predicting ribosomally synthesised and post‐translationally modified peptides (RiPPs) was employed to generate an extensive catalogue of candidate structures. This resource bridges the gap between genome‐derived predictions and in silico mass‐spectrometry searches, facilitating the identification of novel RiPPs from both microbial and plant sources. Future extensions of this platform are poised to include other natural product classes by incorporating additional biosynthetic rules.

Complementing this, a comprehensive computational platform has been developed to predict the chemical structures of genomically encoded antibiotics across all major antibiotic classes. By applying advanced algorithms to thousands of bacterial genomes—including both cultured isolates and metagenomic assemblies—this tool has charted the secondary metabolite potential of over ten thousand strains. The high accuracy of structure prediction enables downstream machine‐learning models to forecast biological activities, thereby guiding experimental validation and prioritising novel scaffolds for therapeutic development.

Biosynthetic Gene Cluster Mining for Natural Product Discovery publication trend

The graph below shows the total number of articles in biosynthetic gene cluster mining for natural product discovery across all publications each year (not limited to Nature Index journals).

Technical terms

Biosynthetic Gene Cluster (BGC): A contiguous set of genes encoding enzymes and regulatory elements responsible for the biosynthesis of a specific natural product.

Genome Mining: The computational scanning of genome sequences to identify and predict the functions of biosynthetic gene clusters.

Ribosomally Synthesised and Post‐Translationally Modified Peptides (RiPPs): A class of natural products produced from ribosomal precursors that undergo enzymatic modifications to form mature bioactive peptides.

Gene Cluster Family (GCF): A group of homologous biosynthetic gene clusters clustered based on shared domain architecture and sequence features, reflecting related chemical outputs.

Tandem Mass Spectrometry (MS/MS): An analytical technique that fragments molecular ions to generate spectra used for structural elucidation and compound identification.

References

1. HypoRiPPAtlas as an Atlas of hypothetical natural products for mass spectrometry database search. Nature Communications (2023).
2. Comprehensive prediction of secondary metabolite structure and biological activity from microbial genome sequences. Nature Communications (2020).
3. Enhanced correlation-based linking of biosynthetic gene clusters to their metabolic products through chemical class matching. Microbiome (2023).
4. BiG-SLiCE: A highly scalable tool maps the diversity of 1.2 million biosynthetic gene clusters. GigaScience (2021).
5. BiG-FAM: the biosynthetic gene cluster families database. Nucleic Acids Research (2020).