DeepRiPP

DeepRiPP integrates genomic and metabolomic data to identify, prioritize, and support the isolation of ribosomally synthesized and posttranslationally modified peptides (RiPPs) from microbial datasets.

Key Features:

• Multiomics integration: Aligns genomic and metabolomic datasets to connect predicted biosynthetic loci with observed metabolites.
• NLPPrecursor: Identifies candidate RiPP precursor peptides independently of genomic context or neighboring biosynthetic genes.
• BARLEY (Biosynthetic Annotation and Ranking for Lantibiotic Enzymes): Uses machine learning algorithms to prioritize genomic loci and assess biosynthetic gene clusters likely to encode novel compounds.
• CLAMS (Comparative Large-scale Automated Metabolomics Screening): Performs large-scale comparative metabolomics to pinpoint predicted metabolites in complex bacterial extracts using a database of 10,498 extracts from 463 strains.
• Machine learning automation: Employs machine learning technologies to automate candidate identification and locus prioritization.

Scientific Applications:

• RiPP discovery: Identification and prioritization of novel ribosomally synthesized and posttranslationally modified peptides from sequenced genomes and metabolomes.
• Genomic–metabolomic linkage: Connecting biosynthetic gene clusters to observed metabolites for validation of predicted RiPP structures.
• Targeted metabolite isolation: Pinpointing and isolating predicted metabolites from complex bacterial extracts for structural characterization.
• Mining microbial diversity: Expanding the repertoire of RiPPs encoded within sequenced microbial genomes for natural product research.

Methodology:

Computational steps include multiomics alignment of genomic and metabolomic data, NLPPrecursor detection of precursor peptides independent of genomic context, BARLEY machine learning–based prioritization of biosynthetic loci and gene clusters, and CLAMS large-scale comparative metabolomics screening across a database of 10,498 extracts from 463 strains to pinpoint target metabolites.