AI-driven biomolecular design: Modalities, models, and translation

Area of Science:

• Biomolecular Science
• Materials Science
• Artificial Intelligence

Background:

• Empirical screening is insufficient for exploring complex biomolecular design spaces.
• Peptides, antibodies, and aptamers are key sequence-defined modalities for biomaterials.
• Biomaterial performance depends on molecular activity and post-synthesis integration.

Purpose of the Study:

• To review advances in AI-driven biomolecular design for therapeutic and diagnostic biomaterials.
• To focus on peptides, antibodies, and aptamers as programmable biomaterial components.
• To explore AI models and closed-loop workflows for optimizing biomaterial development.

Main Methods:

• Synthesis of advances in AI-driven biomolecular design (2020-2026).
• Examination of advantages and constraints of peptides, antibodies, and aptamers.
• Analysis of predictive, generative, and optimization-based AI models.

Main Results:

• AI significantly enhances the design and optimization of functional biomaterials.
• Peptides, antibodies, and aptamers serve as versatile programmable components.
• Closed-loop AI workflows offer a framework for addressing translational bottlenecks.

Conclusions:

• AI is essential for navigating high-dimensional biomolecular design spaces.
• Closed-loop AI workflows integrating design, synthesis, and validation are critical for translation.
• This review provides a roadmap for AI-driven development of next-generation biomolecular biomaterials.