NRPS-PKS

NRPS-PKS analyzes and predicts domain organization and substrate specificity of nonribosomal peptide synthetases (NRPS) and three types of polyketide synthases (PKS) to elucidate biosynthetic pathways of natural products such as cyclosporin, rifamycin, and erythromycin.

Key Features:

• Database integration: Integrates four comprehensive databases to inform domain organization and substrate specificity predictions based on experimentally characterized biosynthetic gene clusters.
• Domain identification: Identifies NRPS and PKS domains and determines domain/module organization within megasynthases.
• Substrate specificity prediction: Predicts adenylation and acyltransferase domain specificities for NRPS and PKS systems.
• Chemical–domain correlation: Correlates chemical structures of natural products with corresponding NRPS/PKS domains and modules.
• Engineering guidance: Provides guidelines for domain/module swapping and site-directed mutagenesis to support engineering of novel natural products.
• Benchmarking and validation: Validated against a dataset comprising 307 open reading frames, 2,223 functional protein domains, 68 starter/extender precursors with recognition motifs, and chemical structures of 101 natural products from four families.

Scientific Applications:

• Biosynthetic pathway annotation: Annotation and characterization of NRPS and PKS biosynthetic gene clusters and their domain architectures.
• Substrate and product prediction: Prediction of adenylation/acyltransferase substrate specificities to infer starter and extender units in natural product assembly.
• Natural product identification: Linking natural product chemical structures to their corresponding biosynthetic domains and modules to support structure–gene cluster assignments.
• Biosynthetic engineering: Informing domain/module swapping and site-directed mutagenesis strategies for creation of modified or novel natural products.

Methodology:

Integrates four databases and uses sequence and structural features derived from experimentally characterized biosynthetic gene clusters to predict NRPS and three types of PKS domain organization and adenylation/acyltransferase specificities, with benchmarking on the stated dataset.