Applications and limitations of AI tools in enzyme design

Bridging Algorithms and Biocatalysis: Perspectives on AI-Supported Enzyme Engineering

The combination of computational and experimental methods has become indispensable for optimization and rational enzyme design. Recently, the development of artificial intelligence (AI)-based tools has further streamlined enzyme engineering pipelines, enabling more accurate designs, while reducing the number of variants required for experimental validation. However, due to the intricate complexity of enzymatic systems, significant challenges must be addressed before we take the next step to fully optimize the use of these AI-guided enzyme design methodologies. These challenges include un-curated datasets, the need to consider both the static and dynamic structure of enzymes, and the requirement for effective interdisciplinary collaborations to ensure the integration of computational and experimental approaches. Here, we present recent advances in AI-based computational enzyme design, discussing the main challenges in the field and how a combination with classical physics-based methods could help overcome them. We further explore novel trends that could completely modulate the future of protein design and provide our outlook on the key concepts and future opportunities that will shape the next steps of enzyme design.

Machine Learning-Assisted Evolution of Broadly Functional Enzyme Libraries

Results indicate that supervised machine learning can help guide the construction of high-value enzyme libraries with expanded catalytic scope, and suggest that supervised machine learning can help guide the construction of high-value enzyme libraries with expanded catalytic scope.

Enzyme Engineering: From Classical Strategies to AI-Driven Biocatalyst Design.

Nowadays, enzyme engineering has moved from traditional structure-based mutagenesis and directed evolution to data-intensive, AI-assisted design paradigms that involve the rapid discovery and optimization of biocatalysts. Whereas classical approaches relied on rational design and experimental screening, advances in high-throughput sequencing, modeling, and machine learning have enabled predictive exploration of sequence-structure-function relationships in enzymes. Importantly, the latest protein language models and deep learning approaches enable accurate prediction of mutational outcomes, stability engineering, and functional annotation at an unprecedented scale. Generative AI models also enable the design of novel enzymes by predicting protein sequences with tailored catalytic functions and broadened substrate specificity. AI combined with design-build-test-learn (DBTL) automation and synthetic biology has enabled the creation of closed-loop engineering workflows for rapid, iterative optimization. This review examines enzyme engineering from classical methods to AI-assisted biocatalyst development, highlighting key advances, challenges, and emerging trends in autonomous laboratories, sustainable biocatalysis, and computational protein design.

Engineering and design of artificial enzymes for non-natural biocatalysis in whole cells.

Artificial enzymes represent transformative biocatalysts that overcome the inherent limitations of natural enzymes in substrate scope and activity specificity, enabling challenging new-to-nature reactions. Despite remarkable advances in designing diverse artificial enzymes through synthetic cofactor or non-canonical amino acid incorporation, their widespread industrial application remains hindered by the high costs and operational complexity of in vitro systems. Whole-cell biocatalysis emerges as a promising alternative by capitalizing on the cellular environment, including cofactor regeneration, multi-enzyme cascades, and enhanced enzyme stability. This review provides a comprehensive introduction to recent progress in artificial enzyme design and their successful applications in whole-cell biocatalysis. We summarize innovative protein engineering tools for creating artificial enzymes, assembly strategies to enhance in vivo catalytic performance, and representative applications of artificial enzymes-containing whole-cell systems in non-natural transformations. We also discuss the challenges and prospects of whole-cell artificial enzyme catalysis in advancing sustainable and scalable biomanufacturing.

Substrate-Specific Differences in Catalytic Strategy and Activity across Ketol-Acid Reductoisomerase Variants

The specificity and catalytic efficiency of enzymes make them attractive for applications ranging from therapeutics to chemical manufacturing. However, it remains challenging to identify specific structural and dynamic mechanisms by which enzymes achieve their catalytic rate enhancements as well as to re-engineer enzymes to improve their catalytic properties. In earlier work we reported the design and selection of KARI mutants with calculated increases in specific activity (i.e., k cat) relative to wild type (WT) for the isomerization step of one of its native substrates: 2-acetolactate (ACL), which leads to the synthesis of the amino acids valine and leucine. Eight mutants were identified with computed improvements in k cat of up to 4 orders of magnitude. In the current study, we investigate the effects of these same mutations on the isomerization of the other native substrate, 2-aceto-2-hydroxybutyrate (AHB, leading to the synthesis of isoleucine). Paralleling our previous work, we use the computational statistical mechanical method transition interface sampling (TIS) to simulate reaction kinetics and compute reaction rate constants. We find that the mutants selected for increased efficiency on ACL had varied levels of activity on AHB–some enhancing reactivity and others diminishing it–with the range in computed AHB rate constants spanning more than 7 orders of magnitude. Analysis of the simulations for WT-AHB revealed that only some of the structural mechanisms associated with mutants’ improved ACL catalysis were expected to transfer to, and thereby improve, AHB catalysis. For two mutants with significantly lower catalytic efficiency on AHB than WT, further analysis identified unique conformational changes that may explain their low activity on AHB.

Microbial lipases: advances in metagenomics and artificial intelligence for enzyme discovery and engineering