Genes involved in the biosynthesis of a natural product are usually encoded in close vicinity – or clustered – in the genome of bacteria and core biosynthetic genes of a natural product class are usually characterized by a certain level of homology based on which they can be identified in the genome even though the product of the biosynthetic pathways might not have any apparent similarity. Despite the development of highly sophisticated tools, some natural product biosynthetic pathways remain unrecognized by state-of-the-art bioinformatic platforms. Others are recognized but encode proteins that are responsible for cryptic transformations.
Our goal is to identify novel families of natural product biosynthetic gene clusters in the ever-increasing wealth of publically available bacterial genomes and to characterize the associated products. Based on biosynthetic insights obtained from these studies, we develop bioinformatic tools for the identification of pathways that follow similar biosynthetic principles and predict the structures of the associated metabolites.