AMPlify

AMPlify predicts and prioritizes antimicrobial peptides (AMPs) from peptide and genomic sequence data to enable discovery of candidate therapeutics against antibiotic-resistant bacteria.

Key Features:

• Deep Learning Architecture: Employs an attentive deep learning framework with an attention mechanism to focus on relevant features within peptide sequences for improved predictive accuracy.
• In Silico Discovery: Performs in silico screening of large volumes of candidate sequences derived from genomes, including Rana [Lithobates] catesbeiana (bullfrog), to prioritize AMP candidates.
• Bioactivity Testing: Predicted peptides are subjected to bioactivity testing against a diverse panel of bacterial species, including World Health Organization priority pathogens such as multi-drug resistant carbapenemase-producing Escherichia coli.
• Validation and Utility: Has identified novel AMPs with antibacterial activity, including four newly predicted peptides that exhibited efficacy against multiple bacterial species in validation experiments.

Scientific Applications:

• AMP discovery and prioritization: Discovery and prioritization of antimicrobial peptides from peptide and genomic data for development of peptide-based therapeutics against resistant bacterial infections.
• Genomic mining: Mining genomic datasets, including Rana [Lithobates] catesbeiana, to identify novel AMP candidates.
• Pathogen-targeted screening: Prioritizing candidates for experimental testing against WHO priority pathogens such as multi-drug resistant carbapenemase-producing Escherichia coli.

Methodology:

Uses attentive deep learning with an attention mechanism for sequence-based prediction and in silico screening of large candidate peptide sets derived from genomic sequences (e.g., Rana [Lithobates] catesbeiana).