A genotype construction platform based on high-throughput flow imaging driven by data has the characteristics of high-throughput, automation, and high sensitivity. By utilizing CRISPR high-throughput genotype construction and droplet microfluidics, ultra high-throughput characterization testing techniques, the relationship between phenotype genes, metabolism, and correlation of HMOs producing strains is deeply explored from the whole genome, such as analyzing the phenotype, genes, metabolism, etc. of HMOs producing strains, and then screening for strains with better expression performance and more stability, achieving efficient strain construction.

The combination of bioinformatics and AI technology plays a synergistic role in enzyme and proteomic research, providing powerful tools and methods for us to gain a deeper understanding of enzyme functions and properties, as well as protein spatial structure simulation. The breakthrough of generative AI technology enables the comprehensive release of the value of biological big data. The complete data loop constructed by "generation feedback optimization" can break through the limits of human computing and bring biological computing into the AI era.

By using gene editing technology, the required functional DNA fragments are introduced into model engineering bacteria, such as Escherichia coli and Pichia pastoris, to change the genetic information of the strains at the genetic level, regulate their production capacity and target product diversity.

Through high-density fermentation and high-purity purification, the target products of human milk oligosaccharides tetrasaccharides and neotetrasaccharides were obtained with a purity of over 92%, which is more than 20% higher than the purity produced by well-known foreign enterprises. The 100L-1000L fermentation tank can meet the sample mass production of some other high value-added biological products, providing conditions for the industrialization of incubating other biological products.