Core Viewpoint - The article discusses the development of a novel artificial intelligence-based protein engineering computational simulation method called AiCE, which integrates structural and evolutionary constraints to enhance protein evolution and function design without the need for specialized AI model training [4][12]. Group 1: Protein Engineering Overview - Protein engineering involves modifying amino acid sequences to alter protein structure and function, offering significant potential in both basic research and industrial applications, with market size expected to exceed hundreds of billions [2]. - Current strategies in protein engineering, such as rational design and directed evolution, face challenges including high costs and long experimental cycles, limiting their scalability [2]. Group 2: AI in Protein Engineering - The rapid advancement of artificial intelligence has led to its application in life sciences, particularly in simulating mutations and functional modifications of proteins [3]. - Existing AI models struggle with generalizability across various proteins and require substantial computational and experimental resources, necessitating the development of more efficient and universal protein engineering strategies [3]. Group 3: AiCE Method Development - The AiCE method allows for efficient protein evolution simulation and function design without the need for training dedicated AI models, significantly reducing computational costs [4][12]. - AiCE utilizes existing universal inverse folding models to predict amino acid sequences based on given protein structures, enhancing the accuracy of predictions [5][6]. Group 4: Performance and Applications - AiCE single module achieved a 16% prediction accuracy using 60 deep mutational scanning datasets, with a 37% performance improvement over unrestricted methods [6]. - AiCE multi module predicts mutation combinations effectively while maintaining low computational costs, demonstrating comparable predictive capabilities to larger models [7]. Group 5: Experimental Validation - The research team validated AiCE's functionality across eight diverse proteins, including deaminases and nucleases, confirming its simplicity, efficiency, and versatility [9][10]. - The development of new base editors with enhanced precision and activity, such as enABE8e and enDdd1-DdCBE, showcases AiCE's practical applications in precision medicine and molecular breeding [9][10]. Group 6: Significance and Future Directions - The study highlights the importance of developing efficient bioinformatics tools to reduce computational burdens, making AI-driven protein engineering accessible to more researchers [12]. - The advancements presented in this research mark a significant step forward in the field of protein evolution, elevating AI-based approaches to a new level [12].