Design of Multifunctional Bioactive Peptides for Potential Application in Food Packaging: A...

Area of Science:

• Biotechnology
• Computational Biology
• Food Science

Background:

• Multifunctional bioactive peptides offer significant potential for active packaging systems to combat foodborne diseases.
• Discovering peptides with antioxidant, antifungal, and antibacterial properties is challenging due to the vast sequence space.

Purpose of the Study:

• To develop a computational framework integrating machine learning and optimization for designing multifunctional bioactive peptides.
• To create a tool for efficient screening and design of peptides with desired bioactivities.

Main Methods:

• Developed three machine-learning-based binary classification models to predict antioxidant, antifungal, and antibacterial activities.
• Integrated these models into an optimization framework to generate novel peptide sequences.
• Analyzed amino acid compositions of designed peptides to confirm characteristic properties.

Main Results:

• Machine learning models achieved high accuracies: 82.37% (antioxidant), 95.49% (antifungal), and 96.75% (antibacterial).
• The computational framework successfully designed novel peptides exhibiting multifunctional bioactivities.
• Designed peptides contained significant proportions of amino acids characteristic of antioxidant, antifungal, and antibacterial activities.

Conclusions:

• The proposed computational framework effectively designs multifunctional bioactive peptides.
• These peptides show potential for application in active food packaging systems.
• A user-friendly tool is available to accelerate the design process for these peptides.