Maria Fátima Lucas

Zymvol Biomodelling

How Much Physics Do We Really Need to Design an Industrial Enzyme?

Maria Fatima Lucas*

Zymvol Biomodeling SL. C/ Pau Claris, 94, 3B - 08010 Barcelona, Spain

* Presenting author

Computational enzyme design sits at an unusual intersection between molecular accuracy and industrial decision-making. While rigorous descriptions of enzymatic catalysis require consideration of conformational sampling, reaction energetics and dynamics, industrial enzyme development imposes an additional constraint: predictions must be sufficiently fast and scalable to outperform experimental search.

Drawing on more than 150 industrial projects, I will discuss how physics-based approaches can guide enzyme discovery and engineering using comparatively small experimental libraries, including the identification of productive substrate poses, access and egress pathways, and functionally important distal residues. Examples will include computationally guided peroxygenase engineering, discovery of industrial ketoreductases, prediction of distal mutations and their impact on activity, stability and immobilization, and machine-learning-guided engineering.

I will conclude by discussing the emerging opportunity to combine molecular simulations, experimentally validated structure–function data and AI models capable of learning richer representations of physical systems, as well as the remaining challenge of connecting molecular design to economically viable industrial scale-up.

References

Molina-Espeja et al. (2016), Synthesis of 1-Naphthol by a Natural Peroxygenase Engineered by Directed Evolution, ChemBioChem, DOI: https://doi.org/10.1002/cbic.201500493

Sosa et. al., (2024), In silico enzyme screening identifies an SDR ketoreductase from Thermus caliditerrae as an attractive biocatalyst and promising candidate for protein engineering, Frontiers in Chemical Biology, DOI: https://doi.org/10.3389/fchbi.2024.1425501

Casilli et al. (2024), Computation-guided engineering of distal mutations in an artificial enzyme, Faraday Discussions 252, 262–278,DOI: https://doi.org/10.1039/d4fd00069b

De Simone et al. (2024), Distal mutations enhance efficiency of free and immobilized NOV1 dioxygenase for vanillin synthesis, Journal of Biotechnology 391, 92–98, DOI DOI: 10.1016/j.jbiotec.2024.06.012

Hassankalhori et al., (2026), Engineering acid-sensing ion channel 1a with disrupted desensitization: a computer-guided approach, in review.

All Sessions by Maria Fátima Lucas

09:30 - 10:00
FUNCHAL 500 anos

Invited Lecture 31 - Maria Fátima Lucas

How Much Physics Do We Really Need to Design an Industrial Enzyme?

Scroll to Top