MIDDAS-M: Motif-Independent De Novo Detection of Secondary Metabolite Gene Clusters through the Integration of Genome Sequencing and Transcriptome Data

• DOI: 10.1371/journal.pone.0084028
• Publication Date: 2013-12-31T17:07:51Z
• Authors: Myco Umemura; Hideaki Koike; Nozomi Nagano; Tomoko Ishii; Jin Kawano; Noriko Yamane; Ikuko Kozone; Katsuhisa Horimoto; Kazuo Shin-ya; Kiyoshi Asai; Jiujiang Yu; Joan W. Bennett; Masayuki Machida
• Keywords: 0303 health sciences; Reverse Transcriptase Polymerase Chain Reaction; Science; Gene Expression Profiling; Q; R; Fungi; Real-Time Polymerase Chain Reaction; 03 medical and health sciences; Multigene Family; Medicine; RNA, Messenger; Genome, Fungal; Nucleotide Motifs; Peptide Synthases; Polyketide Synthases; Algorithms; Biomarkers; Software; Research Article; Oligonucleotide Array Sequence Analysis

Many bioactive natural products are produced as "secondary metabolites" by plants, bacteria, and fungi. During the middle of 20th century, several secondary metabolites from fungi revolutionized pharmaceutical industry, for example, penicillin, lovastatin, cyclosporine. They generally biosynthesized enzymes encoded clusters coordinately regulated genes, motif-based methods have been developed to detect metabolite biosynthetic (SMB) gene using sequence information typical SMB core genes such polyketide synthases (PKS) non-ribosomal peptide synthetases (NRPS). However, no detection method exists that functional do not include at present. To advance exploration clusters, especially those without known we MIDDAS-M, a motif-independent de novodetection algorithm clusters. We integrated virtual cluster generation in an annotated genome with highly sensitive scoring cooperative transcriptional regulation member genes. MIDDAS-M accurately predicted 38 experimentally confirmed and/or other 3 fungal strains. further identified new ustiloxin B, which was validated. Sequence analysis indicated novel mechanism biosynthesis independent NRPS. Because it is fully computational empirical knowledge about allows large-scale, comprehensive including mechanisms contain any functionally characterized