AI Methods for Antimicrobial Peptides: Progress and Challenges.

Area of Science:

• Biochemistry and Bioinformatics
• Computational Biology
• Drug Discovery

Background:

• Antimicrobial peptides (AMPs) show promise against multidrug-resistant pathogens.
• Traditional screening methods are costly and time-consuming.
• AI, particularly machine learning (ML), is crucial for accelerating AMP identification and design.

Purpose of the Study:

• To provide a comprehensive overview of AI methods in AMP discovery and design.
• To focus on emerging AI techniques like large language models (LLMs) and graph neural networks (GNNs).
• To address the limitations and future opportunities in AI-driven AMP research.

Main Methods:

• Review of recent advancements in AI for AMP discovery.
• Analysis of classical ML, deep learning (DL), LLMs, and GNNs.
• Exploration of structure-guided AMP design approaches.

Main Results:

• AI has revolutionized the discovery of anti-infective peptides.
• A shift from classical ML to DL models is observed.
• LLMs, GNNs, and structure-guided design represent significant, underexplored potential.

Conclusions:

• AI methods are vital for overcoming challenges in AMP discovery.
• Further research into LLMs, GNNs, and structure-guided design is needed.
• Addressing current limitations will pave the way for future AMP development.