Bioinformatics tool for genome mining of gene clusters encoding structurally interesting metabolites Stanislav Kadlcik 1 , Ondrej Hrebicek 2 , Zdenek Kamenik, Jiri Janata 1 Institute of Microbiology of the Czech Academy of Sciences, Videnska 1083,Prague, Czech Republic and 2 Institute of Microbiology, Czech Republic Search for new structurally interesting microbial bioactive compounds and elucidation of unusual biosynthetic pathways can benefit from gene clusters mining in genome databases e.g. GenBank. However, the current genome mining tools are mainly focused to well recognize frequent types of gene clusters e.g. those involving typical NRPS or PKS genes. Highly desirable types of gene clusters coding for rare and unusual new structures of natural products can be overlooked. We have developed a bioinformatics tool for automated genome mining of gene subclusters i.e. minimal sets of genes indispensable for biosynthesis of building blocks of natural products. The tool uses a Python code and a Blast search in GenBank genomic data to find gene clusters where the searched “marker” subcluster genes are co-localized. The neighbourhood of the marker genes is displayed and analyzed in terms of prediction of function. The tool has a graphical user interface to facilitate user-friendly work to wide range of researchers. We will demonstrate the application of the tool on discovery of about 100 new highly diverse gene clusters for biosynthesis of compounds with a rare 4-alkyl-L-proline structural motif. The tool can nicely complement existing genome mining tools e.g. antiSMASH and PRISM.
P37
© The Author(s), 2022
Made with FlippingBook Learn more on our blog