FACS

FACS predicts and classifies antimicrobial peptides (AMPs) in genomes and metagenomes to identify high-quality AMP candidates and assess antimicrobial and hemolytic activities.

Key Features:

• End-to-End Pipeline: Provides a comprehensive pipeline for prospection of high-quality AMP candidates from genomic and metagenomic datasets, addressing limitations of standard gene prediction methods on short peptides.
• Novel Feature Set: Utilizes an innovative set of 22 peptide features for classifier construction.
• Enhanced Precision and Recall: Employs classifiers that achieve improved precision and competitive performance with state-of-the-art methods through evaluation on standard benchmarks and a more stringent testing regime.
• Realistic Simulations and Data Validation: Validated using realistic simulations and real-world data to demonstrate recovery of high-quality AMP candidates from diverse datasets.

Scientific Applications:

• Antimicrobial peptide discovery: Supports identification of novel AMPs as potential therapeutic agents against multidrug-resistant pathogens.
• Large-scale genomic and metagenomic screening: Enables processing and analysis of large genomic and metagenomic datasets to prospect for AMP candidates.

Methodology:

Machine learning classifiers trained on a novel 22-feature peptide representation to predict antimicrobial and hemolytic activities, evaluated using standard benchmarks, a more stringent testing regime, realistic simulations, and real-world genomic and metagenomic data, and designed to handle challenges posed by short peptide sequences and limitations of standard gene prediction methods.