Beyond SELEX: AI-driven aptamers from sequence to structure to function.

AI-driven aptamer design methods from sequence to function have advanced rapidly between 2024 and 2026, enabling de novo generation and optimization of aptamers.

• Why it matters: Traditional SELEX-based aptamer discovery faces limitations such as library capacity, PCR bias, and multi-round screening, which hinder rapid development of aptamers for clinical and biosensing applications.
• What they did: The review covers bioinformatics databases, structure prediction tools like AlphaFold 3, machine learning models for interaction prediction, and recent AI generative approaches including diffusion models, language models, and reinforcement learning, with over 2024-2026 advances highlighted.
• The result: These AI-driven strategies facilitate precise, efficient aptamer development for diagnostics, therapy, and biosensing, and pave the way for future closed-loop, interpretable frameworks in translational research.