Artificial intelligence-driven anticancer peptide discovery.

Area of Science:

• Biotechnology
• Computational Biology
• Drug Discovery

Background:

• Cancer remains a significant global health challenge with limitations in current therapies.
• Anticancer peptides (ACPs) offer a promising avenue for novel drug development due to their selectivity and low toxicity.
• Traditional ACP screening methods are inefficient, costly, and complex, hindering high-throughput applications.

Purpose of the Study:

• To systematically review the advantages and limitations of 68 AI models for ACP screening.
• To propose a comprehensive AI-based framework for ACP screening to enhance efficiency and clinical translation.
• To integrate multi-omics and biotechnologies into the AI framework for improved ACP development.

Main Methods:

• Systematic review and summarization of 68 AI models for ACP screening.
• Development of a novel AI-based ACP screening framework encompassing data organization, feature extraction, model construction, interpretability analysis, and experimental validation.
• Integration of multi-omics and biotechnologies with the AI screening framework.

Main Results:

• AI models demonstrate significant potential in improving ACP screening efficiency and accuracy.
• Existing AI models for ACP screening often lack interpretability and experimental validation, impacting their practical utility.
• The proposed AI framework addresses these limitations by incorporating interpretability analysis and experimental validation.

Conclusions:

• The developed AI-based ACP screening framework accelerates the identification and development of novel anticancer peptides.
• This framework enhances screening efficiency and promotes the clinical application of AI-discovered ACPs.
• Integrating AI with multi-omics and biotechnologies is crucial for translating AI-selected ACPs into effective clinical therapies.