DePP

DePP predicts phage depolymerases from protein sequences using a machine-learning classifier based on amino-acid-derived feature vectors.

Key Features:

• Machine-learning classifier: Uses a supervised machine-learning approach to distinguish phage depolymerases from other proteins.
• Amino-acid-derived feature vectors: Represents proteins by feature vectors derived from amino acid properties.
• Training data: Trained on a dataset of experimentally validated phage depolymerases.
• Performance: Reported model accuracy is approximately 90% on the provided dataset.
• Biological target: Specifically identifies phage depolymerases—enzymes that degrade the extracellular matrix of bacterial biofilms.

Scientific Applications:

• Depolymerase discovery: Identification of candidate phage depolymerases for experimental validation.
• Biofilm research: Facilitates study of enzymes that degrade biofilm extracellular matrix components.
• Antibiotic-resistance research: Supports exploration of adjunctive agents against biofilm-associated and antibiotic-resistant bacterial infections.
• Protein functional annotation: Assists annotation of phage-encoded proteins with depolymerase activity.

Methodology:

Computes amino-acid-derived feature vectors for protein sequences and applies a supervised machine-learning classifier trained on experimentally validated phage depolymerases, with reported accuracy around 90%.