Waste to resource: Mining antimicrobial peptides in sludge from metagenomes using machine learning

Area of Science:

• Microbiology
• Biotechnology
• Environmental Science

Background:

• Antibiotic resistance is a critical global health threat.
• Antimicrobial peptides (AMPs) are natural compounds with potential as antibiotic alternatives.
• Environmental sources offer vast, untapped reservoirs for novel AMP discovery.

Purpose of the Study:

• To identify and characterize novel antimicrobial peptides (AMPs) from environmental samples.
• To evaluate the potential of engineered environments, specifically sludge, for AMP discovery.
• To validate the antibacterial activity of predicted AMPs.

Main Methods:

• Metagenomic sequencing and machine learning for large-scale AMP prediction.
• Metaproteomic analysis and correlation studies for targeted AMP identification in sludge.
• Chemical synthesis and experimental validation of candidate AMPs.

Main Results:

• Predicted over 16 million antimicrobial peptides (AMPs) from environmental metagenomes.
• Identified 27 candidate AMPs from sludge, with 21 demonstrating antibacterial activity against tested strains.
• Confirmed engineered environments, particularly sludge, as a viable source for novel AMPs.

Conclusions:

• Engineered environments are a promising source for discovering new antimicrobial peptides (AMPs).
• Metagenomic and metaproteomic approaches coupled with machine learning are effective for mining AMPs from complex environments like sludge.
• This study validates the potential of AMPs as a future strategy against antibiotic-resistant bacteria.