Breakthrough advances in artificial intelligence (AI) are propelling de novo protein design past the boundaries of natural evolution, making it possible to engineer proteins with entirely novel structures and functions. Benefiting from iterative improvements in machine learning algorithms, AI-driven de novo strategies have overcome traditional reliance on natural templates. These approaches autonomously optimize catalytic sites and overall stability, significantly enhancing enzyme performance and applicability. Generative models, including large language models and diffusion models, can rapidly produce novel protein structures with specialized functions, offering innovative technological paths for biomolecule development. This review systematically discusses recent key developments and representative examples of AI applications in enzyme engineering and design. We highlight a fundamental shift from traditional “structure-based function analysis” to a new paradigm of “function-driven structural innovation.” Furthermore, we comprehensively evaluate current challenges in AI-driven protein engineering and suggest promising future directions.
Jin et al. (Tue,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: