Lightning Talk: Bioinformatics, Molecular Modelling, and AI: Transforming the Enzyme Engineering Landscape

Enzyme engineering has become a critical enabler of sustainable and efficient processes in the pharmaceutical, chemical, and food industries. However, conventional approaches such as directed evolution or structure-based mutagenesis remain resource-intensive and are often limited by the availability of structural or functional data. This presentation outlines a computational framework that integrates molecular modeling, bioinformatics, and artificial intelligence (AI) to accelerate enzyme design and optimisation under these constraints.

We discuss how sequence-based analyses (e.g., multiple sequence alignment, motif detection) and AI-driven sequence-function prediction can be combined with molecular modelling techniques (such as docking, molecular dynamics, and in silico mutagenesis) to construct and evaluate large virtual mutant libraries. This integrated approach enables the prioritisation of high-potential enzyme variants before experimental validation, significantly reducing the number of required wet-lab iterations.

• The methodology will be illustrated through case studies covering:
• Enzyme improvement without available crystal structures.
• Identification of novel enzymes from metagenomic databases.
• Optimisation of selectivity and activity in industrial biocatalysis pathways.