Construction and Validation of an Efficient Screening Model for Fermentation Strains Based on Genetic Algorithm

In fermentation engineering, traditional strain screening methods are inefficient and cannot meet the needs of rapidly developing industries. The purpose of this study is to construct and verify an efficient fermentation strain screening model based on genetic algorithm (GA). First, the growth rate, metabolites and related characteristics of different fermentation strains are collected to construct a feature matrix. Then, the genetic algorithm is used to encode the strains. Based on the fitness function, the strain combination is optimized through selection operations, single-point crossover technology and the introduction of random mutations. After iterative screening, the best strain combination is selected, and then fermentation experiments are carried out to verify that the maximum product yield of the screened strains is 76 g/L. The screening model based on genetic algorithm not only improves the strain screening efficiency but also significantly improves the fermentation production efficiency, and has good application prospects.