Structure-Guided and AI-Assisted Enzyme Engineering for Sustainable Biomanufacturing

Abstract Enzyme engineering is a key area driving the development of sustainable biomanufacturing through the creation of ultra-efficient and highly selective biocatalysts for a variety of industrial applications. Whereas traditional approaches, such as rational design and directed evolution, have indeed formed the basis of this area, both techniques generally suffer from major limitations due to the huge sequence-function space involved and high experimental requirements. Incorporation of artificial intelligence technologies such as machine learning and deep learning is transforming this area into autonomous design and optimization, thus better exploring this huge space. The breakthroughs in deep learning, such as AlphaFold, have greatly aided in protein structure prediction. This propels the pace of elucidating and optimizing enzyme activities. The integration of structure prediction and AI assistance would be essential in optimizing enzyme stability, activity, and promiscuity. The major practical applications of enzyme technologies could be used to optimize biofuel production, green synthesis of drugs, and waste valorization for a circular economy. Although we have much to incorporate from enzyme technologies, challenges also lie ahead regarding their computational, data, and experimental needs. The problems, however, need to be cracked in order to maximize the potential of enzyme technology to revolutionize sustainable production.