Ai-driven discovery of novel antifungal non-ribosomal peptides from microbial genomes
An artificial intelligence platform that links microbial biosynthetic gene clusters to LC-MS/MS features, enabling rapid discovery of novel antifungal non-ribosomal peptides. Two candidates, CHM-1504 and CHM-1460, have been isolated and validated as predicted antifungal compounds.
Overview
This solution is an AI-powered discovery platform that accelerates the identification of novel antifungal non-ribosomal peptides (NRPs) from microbial sources. By integrating genomic data with LC-MS/MS (liquid chromatography tandem mass spectrometry) profiling of microbial fermentation extracts, the platform connects biosynthetic gene clusters to specific molecular features and prioritizes compounds with predicted antifungal activity. Applied to a library of 400 Streptomyces strains, the method identified 20 novel antifungal NRPs, two of which (CHM-1504 and CHM-1460) have already been isolated in milligram quantities for further evaluation. The approach addresses the urgent need for new antifungal agents in an era of rising drug-resistant fungal infections.
Technical specifications
Key features:
- AI algorithm that correlates biosynthetic gene cluster predictions with LC-MS/MS molecular features from fermentation extracts
- Application to 400 Streptomyces strains, yielding 20 novel antifungal NRP candidates
- Isolation of CHM-1504 (26 mg) and CHM-1460 (11 mg) as lead compounds with predicted antifungal activity
- Continued isolation efforts for the remaining 18 identified candidates
- Integration of genomics and metabolomics data streams for natural product discovery
Technology readiness level
Two lead compounds have been isolated in sufficient quantities for preliminary characterization. The remaining 18 predicted antifungal NRPs are progressing through isolation workflows. The AI discovery pipeline has been validated on a 400-strain Streptomyces library, demonstrating reproducible identification of bioactive candidates. Further biological validation and optimization are required to advance the leads toward preclinical development.
About Carnegie Mellon University
Carnegie Mellon University is a private, global research university in Pittsburgh with a strong applied-research culture and an emphasis on translational impact. Industry collaborates through co-located labs, specialized testbeds, and a federally funded software engineering center that de-risks complex systems. Proximity to a regional robotics and advanced manufacturing cluster enables rapid prototyping, field trials, and access to a skilled talent pipeline, with flexible sponsored-research and affiliate models. Research is supported by competitive federal funding from agencies such as NSF, DoD, DOE, NIH, and NASA. A dedicated tech transfer office and entrepreneurship programs provide IP strategy, licensing, and startup acceleration.
delivered to your inbox.