Efficient Cell Factory Design by Combining Meta-Heuristic Algorithm with Enzyme Constrained Metabolic Models

The design of high-performance microbial cell factories is essential for advancing sustainable biomanufacturing. However, the intricate nature of metabolic networks complicates the prediction of genetic interventions, making strain optimization a challenging combinatorial problem. Here we present a novel computational strain design algorithm-MetaStrain that integrates enzyme-constrained models (ecModels) with efficient meta-heuristic algorithms to identify non-intuitive combinatorial gene targets for improving product yields. Our algorithm employs a modified enforced objective flux scanning algorithm to reduce the dimensionality of gene candidates and annotate editing strategies. Subsequently, binary and symbolic genetic algorithms and the adaptive differential evolution algorithm with external archives (JADE) are employed for gene target selection, with product yield as the fitness function. We adapted minimization of metabolic adjustment strategy for ecModels to evaluate the productive phenotypes of mutant strains. Computational validation using Saccharomyces cerevisiae as a host demonstrates significant enhancements in 2-phenylethanol production. Further generalization to spermidine biosynthesis confirms the framework’s robustness. The proposed algorithm achieves high computational efficiency, stable convergence, and strong adaptability across diverse metabolic targets. This study provides a powerful computational tool for metabolic engineering, addressing the complexity of strain design and highlighting the potential of meta-heuristic optimization in synthetic biology.