Efficient cell factory design by combining meta-heuristic algorithm with enzyme-constrained metabolic models

Area of Science:

• Metabolic Engineering
• Synthetic Biology
• Computational Biology

Background:

• Designing high-performance microbial cell factories for sustainable biomanufacturing is challenging due to complex metabolic networks.
• Predicting synergistic genetic interventions for yield improvement is computationally intensive.

Purpose of the Study:

• To present MetaStrain, a computational framework for identifying non-intuitive combinatorial gene targets to enhance microbial product yields.
• To improve the efficiency and accuracy of metabolic engineering strategies.

Main Methods:

• Integration of enzyme-constrained models (ecModels) with meta-heuristic algorithms.
• Pre-screening using modified enforced objective flux scanning to reduce candidate gene dimensionality.
• Simulation of mutant phenotypes by modulating enzyme bounds within ecModels for efficient design space exploration.

Main Results:

• MetaStrain identified significant enhancements in 2-phenylethanol and spermidine biosynthesis in Saccharomyces cerevisiae.
• Experimental validation in Escherichia coli demonstrated up to a 61.25% increase in L-tryptophan titer with a five-target combination.
• The framework showed high computational efficiency, stable convergence, and adaptability across diverse metabolic targets.

Conclusions:

• MetaStrain provides a powerful computational tool for metabolic engineering, bridging prediction and experimental validation.
• Meta-heuristic optimization holds significant potential for advancing synthetic biology and biomanufacturing.
• The framework facilitates the rational design of microbial cell factories for sustainable bioproduction.