Improving metabolic engineering design with enzyme-thermo optimization.

Metabolic target and strategy design play a critical role in enhancing the DBTL (Design-Build-Test-Learn) cycle in metabolic engineering. Classical stoichiometric algorithms such as OptForce and FSEOF narrow the experimental search space but fail to account for thermodynamic feasibility and enzyme-usage costs, leaving a space for their predictive performance. In this study, we introduce ET-OptME, a framework integrating two algorithms that systematically incorporate enzyme efficiency and thermodynamic feasibility constraints into genome-scale metabolic models. By mitigating thermodynamic bottlenecks and optimizing enzyme usage through a stepwise constraint-layering approach, ET-OptME delivers more physiologically realistic intervention strategies when compared with experimental records. Quantitative evaluation of five product targets in the Corynebacterium glutamicum model reveals that the algorithm showing at least 292 %, 161 % and 70 % increase in minimal precision and at least 106 %, 97 % and 47 % increase in accuracy when compared with stoichiometric methods, thermodynamic constrained methods, and enzyme constrained algorithms respectively.