Natural products (NPs) encompass various bioactive compounds originating from secondary metabolism in diverse organisms [1]. The structural diversity and inherent bioactivity of NPs make them promising drug leads, and indeed around 60% of newly approved drugs between 1981 and 2019 are structurally related to NPs [1]. Historically, they have been utilized in traditional medicine, with ancient texts offering potential sources for novel NPs [2, 3, 4]. However, the discovery rate of novel and abundant bacterial NPs has decreased in recent decades [5, 6, 7, 8].

The once-fruitful ‘grind and find’ approaches to NP discovery, where culture bioactivity screening guides NP isolation and characterization efforts, now result in high re-discovery rates and contributed to the pharmaceutical industry's shift away from NPs as drug leads in the 1990s [1,9,10]. During the genomic era, it has become evident that biosynthetic enzymes for making microbial NPs are encoded as co-localized biosynthetic gene clusters (BGCs) facilitating their expression in response to environmental stimuli. Indeed, many BGs remain silent (not actively transcribed) without specific triggers, rendering their NPs “invisible” to traditional discovery methods, emphasizing the importance of genome-guided approaches for bacterial NP discovery [8]. As our access to microbial whole-genome sequences increases, it has become apparent that nascent BGCs in microbial genomes are an untapped source of new NPs [8].

Bioinformatic tools can facilitate the organization, analysis, and interpretation of extensive DNA sequence datasets, and are also used to examine gene expression and predict protein structures. Well-established bioinformatic tools such as antiSMASH (the antibiotics and Secondary Metabolite Analysis SHell) can aid NP discovery by analyzing the sequence similarities of environmental bacterial DNA compared to characterized BGCs allowing identification of biosynthesis genes, prediction of their functions, and the resulting NP scaffolds [11]. It is also crucial to validate bioinformatic-driven hypotheses by linking BGCs and the enzymes they encode to NPs using experimental methods, thereby allowing NP researchers to identify BGCs making similar or novel NPs, develop methods for targeted BGC activation, and rationally engineer BGCs to make NP analogues [10,11]. This review aims to highlight recent bioinformatic-driven approaches for the discovery of bacterial NPs. For more comprehensive reviews of available tools for NP discovery in bacteria please see Refs. [12, 13, 14, 15, 16, 17].