In short
- Natural products and their derivatives remain a major source of new drugs, especially antibacterials.
- A survey of about 170,000 bacterial genomes estimated that only 3% of the natural products they could encode have been experimentally characterised.
- antiSMASH 8.0 detects 101 types of biosynthetic gene cluster, up from 81 in the previous version.
- A predicted gene cluster is a hypothesis. Novelty and production have to be confirmed in the lab.
Why mine genomes for natural products?
Bacteria and fungi make small molecules that are not needed for basic growth but help them compete, communicate and survive. Many of these specialised metabolites became antibiotics, anticancer agents and immunosuppressants. A review of drugs approved over nearly four decades, up to 2019, showed that natural products and compounds derived from them remain a major source of new medicines [1].
The genes that make each metabolite usually sit together in a biosynthetic gene cluster (BGC), so they can be found by reading the genome. The potential is large. An analysis of about 170,000 bacterial genomes and 47,000 metagenome-assembled genomes estimated that only 3% of the natural products these genomes could encode have been characterised experimentally [2]. Streptomyces encoded by far the greatest biosynthetic diversity, followed by Amycolatopsis, Kutzneria and Micromonospora. Less-studied groups such as Myxococcaceae and some cyanobacteria also showed high potential [2].
What antiSMASH does
antiSMASH is the most widely used tool for detecting and characterising BGCs in bacteria and fungi. It searches the genome for signature biosynthetic enzymes, such as polyketide synthases (PKS) and nonribosomal peptide synthetases (NRPS), and groups the surrounding genes into a predicted cluster region. Version 7.0 improved predictions of regulation and chemical structure [3]. Version 8.0 raised the number of detectable cluster types from 81 to 101 and improved support for terpenoids, tailoring enzymes and modular enzymes [4].
For each region, antiSMASH reports the cluster type, the core biosynthetic genes, and domain-level predictions for modular PKS and NRPS enzymes. It also compares the region with known clusters. The key reference is MIBiG, a community-curated repository of experimentally validated BGCs linked to their products [5, 6].
Reading antiSMASH results
| Output | What it means | How to use it |
|---|---|---|
| Region and cluster type | Location and predicted class (NRPS, type I PKS, RiPP, terpene, etc.) | Builds the biosynthetic inventory of the strain |
| KnownClusterBlast similarity | Share of genes in a known MIBiG cluster that have a match in the region | High similarity suggests a known compound; low similarity suggests possible novelty |
| Contig edge flag | The cluster runs off the end of a contig | The cluster may be incomplete; treat its size and similarity with caution |
| Domain predictions | Predicted building blocks of PKS and NRPS products | Gives a first idea of the chemical scaffold |
When many strains are analysed together, BiG-SCAPE groups related clusters into gene cluster families. This shows which clusters are shared across a collection and which are unique to one strain, which helps prioritise candidates for further work [7].
What genome mining cannot tell you
- Whether the cluster is active. Many BGCs are silent under laboratory conditions. Producing the compound may require different media, co-culture or heterologous expression.
- The exact structure of the product. Domain predictions give a scaffold, but tailoring reactions and unusual chemistry often need metabolomics or NMR to resolve.
- Whether low similarity means a new compound. A cluster may look novel simply because its relatives are not yet in MIBiG.
- The full cluster in a fragmented assembly. Large modular PKS and NRPS genes are repetitive and often break across contigs in short-read assemblies. Long-read sequencing helps.
Uses beyond drug discovery
BGC analysis is also valuable for probiotic, food and agricultural strains. It can reveal bacteriocins and other ribosomally synthesised peptides (RiPPs) behind antimicrobial activity, siderophores involved in iron acquisition, and metabolites that matter for plant–microbe interactions. For a strain being developed as a product, a full biosynthetic profile is part of understanding what it can do.
In the BacSeq Analytics Premium package, secondary metabolite analysis with antiSMASH reports every predicted region with its cluster type, closest known cluster and similarity, and flags clusters that are incomplete or potentially novel. The interpretation explains what each finding means for the research question.
References
- Newman DJ, Cragg GM. Natural products as sources of new drugs over the nearly four decades from 01/1981 to 09/2019. J Nat Prod. 2020;83:770–803. doi:10.1021/acs.jnatprod.9b01285
- Gavriilidou A, Kautsar SA, Zaburannyi N, et al. Compendium of specialized metabolite biosynthetic diversity encoded in bacterial genomes. Nat Microbiol. 2022;7:726–735. doi:10.1038/s41564-022-01110-2
- Blin K, Shaw S, Augustijn HE, et al. antiSMASH 7.0: new and improved predictions for detection, regulation, chemical structures and visualisation. Nucleic Acids Res. 2023;51:W46–W50. doi:10.1093/nar/gkad344
- Blin K, Shaw S, Vader L, et al. antiSMASH 8.0: extended gene cluster detection capabilities and analyses of chemistry, enzymology, and regulation. Nucleic Acids Res. 2025;53:W32–W38. doi:10.1093/nar/gkaf334
- Terlouw BR, Blin K, Navarro-Muñoz JC, et al. MIBiG 3.0: a community-driven effort to annotate experimentally validated biosynthetic gene clusters. Nucleic Acids Res. 2023;51:D603–D610. doi:10.1093/nar/gkac1049
- Zdouc MM, Blin K, Louwen NLL, et al. MIBiG 4.0: advancing biosynthetic gene cluster curation through global collaboration. Nucleic Acids Res. 2025;53:D678–D690. doi:10.1093/nar/gkae1115
- Navarro-Muñoz JC, Selem-Mojica N, Mullowney MW, et al. A computational framework to explore large-scale biosynthetic diversity. Nat Chem Biol. 2020;16:60–68. doi:10.1038/s41589-019-0400-9