DeepRiPP integrates multiomics data to automate discovery of novel ribosomally synthesized natural products

Abstract

Microbial natural products represent a rich resource of evolved chemistry that forms the basis for the majority of pharmacotherapeutics. Ribosomally synthesized and posttranslationally modified peptides (RiPPs) are a particularly interesting class of natural products noted for their unique mode of biosynthesis and biological activities. Analyses of sequenced microbial genomes have revealed an enormous number of biosynthetic loci encoding RiPPs but whose products remain cryptic. In parallel, analyses of bacterial metabolomes typically assign chemical structures to only a minority of detected metabolites. Aligning these 2 disparate sources of data could provide a comprehensive strategy for natural product discovery. Here we present DeepRiPP, an integrated genomic and metabolomic platform that employs machine learning to automate the selective discovery and isolation of novel RiPPs. DeepRiPP includes 3 modules. The first, NLPPrecursor, identifies RiPPs independent of genomic context and neighboring biosynthetic genes. The second module, BARLEY, prioritizes loci that encode novel compounds, while the third, CLAMS, automates the isolation of their corresponding products from complex bacterial extracts. DeepRiPP pinpoints target metabolites using large-scale comparative metabolomics analysis across a database of 10,498 extracts generated from 463 strains. We apply the DeepRiPP platform to expand the landscape of novel RiPPs encoded within sequenced genomes and to discover 3 novel RiPPs, whose structures are exactly as predicted by our platform. By building on advances in machine learning technologies, DeepRiPP integrates genomic and metabolomic data to guide the isolation of novel RiPPs in an automated manner.

Precursor peptide-targeted mining of more than one hundred thousand genomes expands the lanthipeptide natural product family

“…Genome-mining studies based on these enzymes have revealed that lanthipeptide BGCs are distributed widely across bacterial phyla [21][22][23][24][25][26][27][28][29][30][31][32]. Despite the success in bioinformatically identifying likely lanthipeptide BGCs, it has been an outstanding challenge to perform highthroughput analysis of the precursor peptides encoded in these gene clusters.…”

Biosynthesis of Cittilins, Unusual Ribosomally Synthesized and Post-translationally Modified Peptides from Myxococcus xanthus

“…An example for such BGCs is the relatively young but steadily expanding family of ribosomally produced and post-translationally modified peptides (RIPPs), which includes the lantibiotics pinensins (5) and epidermin (6), lassopeptides such as citrocin (7), the antibiotic bottromycins, which feature a unique macoramidine linkage (8,9), and the marine-derived divamides (10), just to name a few. Recently developed bioinformatics tools allow the automated detection of certain classes of RIPPs in bacterial genomes based on signature genes and canonical recognition elements (11)(12)(13)(14)(15)(16). However, the pace of novel RIPPs discovery in recent years suggests that an abundance of yet uncovered RIPP families likely exist.…”

Integration of machine learning and pan-genomics expands the biosynthetic landscape of RiPP natural products

“…Another tool recently described, NeuRiPP, is capable of predicting precursors independent of RiPP subclass, but is limited to precursor analysis 30 . Yet another tool, DeepRiPP, can detect novel RiPP BGCs that are chemically far removed from known examples, but is mainly designed to identify new members of known classes 31 . In the end, an algorithm for the discovery of BGCs encoding novel RiPP classes will need to integrate various sources of information to reliably identify genomic regions that are likely to encode RiPP precursors along with previously undiscovered RTEs.…”