Predicting Algorithm of DNA Recombination Site Based on Machine Learning

DNA recombination is one of the basic tools used in genetic engineering, where the site-specific recombination of cells and viruses can achieve various biological functions. However, the traditional recombination system is limited to the sequence containing similar consensus sequences or the sequences are highly similar to each other. In order to solve the problem of sequence restriction and study the correlation between the structure and function of the recombination site. This article focuses on the attC site of the bacterial integration subsystem, analyzes and quantifies the structural characteristics of the attC site, and combines XGBoost regression algorithm to analyze the synthesized attC site. The recombination frequency is predicted, compared with the existing technologies, the effect of this algorithm is significantly improved. By analyzing the relationship between structure and function, the design algorithm of attC sites can be effectively improved.