The Select of Different Formulations for Lipid Nanoparticles in Gene Delivery Based on Decision Tree

The launch of two mRNA vaccines based on lipid nanoparticle delivery systems is considered to be a milestone in the field of nucleic acid delivery. Lipid nanoparticles (LNP) are a class of lipid vesicles composed of a variety of lipids, whose uniform lipid core is surrounded by the outer lipid membrane. The lipid components of such carriers mainly include ionizable lipids, phospholipids, cholesterol, and polyethylene glycol lipids, which are self-assembled into nanoparticles through molecular interactions during preparation. It effectively prevents the enzymatic degradation of RNA during in vivo delivery and delivers RNA to cells in a safe and effective manner. The formulation of a LNP is closely related to its efficacy, so this paper first compares the transfection rate and physicochemical properties of EGFP-mRNA and siRNA lipid nanoparticles prepared in the same way. Then, the amount of phospholipids most suitable for LNP formulation was selected by comparing the properties of LNPs containing different DSPC ratios, and the difference of the results was analyzed by paired sample T test. Finally, the decision tree algorithm was used to analyze the effect of single component on the encapsulation efficiency of EGFP-mRNA and siRNA lipid nanoparticles, and the intelligent optimization algorithm such as particle swarm optimization algorithm can be used to optimize the ratio formula in the future.