AI-driven biomolecular design review maps advances in peptides, antibodies, and aptamers for therapeutic biomaterials

The design of functional biomaterials is often hampered by the vast, high-dimensional sequence and structural spaces of biomolecular modalities. Traditional empirical screening and iterative experimentation are increasingly inefficient for exploring these complex landscapes. Peptides, antibodies, and aptamers are critical sequence-defined modalities, offering molecular recognition, responsiveness, and biological function to biomaterial platforms. However, their performance relies on preserving folding, assembly, stability, and function through synthesis, conjugation, and integration, posing significant translational challenges.

This comprehensive review synthesizes advances from 2020 to early 2026 in AI-driven biomolecular design, specifically emphasizing peptides, antibodies, and aptamers as programmable components for therapeutic and diagnostic biomaterials. The authors examined the modality-specific constraints and advantages of these systems. They analyzed major predictive, generative, and optimization-based AI models relevant to their design, providing a critical map for rational development. The review also advocates for closed-loop AI workflows as a practical framework to address downstream translational bottlenecks.

The review highlights that AI integration is fundamentally reshaping the design, optimization, and translation of functional biomaterials, particularly for peptides, antibodies, and aptamers. It details how these modalities are crucial for forming material architectures or endowing biomaterial platforms with specific functions. Performance is shown to depend not just on intrinsic molecular activity but also on the preservation of folding, assembly, stability, and function post-synthesis and integration. The analysis of predictive, generative, and optimization-based AI models reveals their distinct advantages and constraints across different biomolecular modalities. > The review strongly makes the case that closed-loop AI workflows, by linking computational design with iterative synthesis, material integration, and biological validation, provide a practical framework for improving translation by embedding downstream constraints directly into the design process.