Prediction of lipid nanoparticles for mRNA vaccines by the machine learning algorithm

Sodium alginate coating simultaneously increases the biosafety and immunotherapeutic activity of the cationic mRNA nanovaccine

“…Overall, SA alleviated the tissue toxicity and inflammation induced by DOTAP-OVA in mice, which suggests that modifying SA establishes a safer formulation of liposome/mRNA complexes for gene therapy. Although the effect of delivery based on the ionizable lipids has been clinically recognized, there are still some patients with adverse reactions 44 . Therefore, it is necessary to continuously explore new delivery systems.…”

The applications of machine learning to predict the forming of chemically stable amorphous solid dispersions prepared by hot-melt extrusion

“…Previous studies have demonstrated that the important substructures associated with the corresponding models were extracted by implementing IG feature importance analysis. For example, ML has successfully been applied to predict lipid nanoparticle (LNP)-based mRNA vaccine, and the important ionized lipid's substructures were extracted based on IG analysis ( Wang et al, 2022 ). Furthermore, the substructures of compounds and solvents were studied from the ML solubility prediction model ( Ye and Ouyang, 2021 ).…”

Converting Nanotoxicity Data to Information Using Artificial Intelligence and Simulation

“…In 2010, SMILES-based nanodescriptors for organic ligands decorating the surfaces of 109 fluorescent magnetic NPs with similar cores were used to train a ML model . In other QNTR studies, SMILES strings were also employed to represent the surface , and core − composition of NPs (Figure A) and simple nanostructures such as fullerenes (Figure B). , SMILES-based nanodescriptors can be calculated using commercial and open-source software packages. , SMILES-based nanodescriptors have also been used to model the toxicity of ligand-grafted MWCNTs, , GNPs, , and MONPs. , For example, Dragon and MOE descriptors encoding surface ligand features were used to model protein binding, acute cytotoxicity, and immune response induced by 83 MWCNTs . However, the SMILES-based nanodescriptors cannot distinguish NPs with the same surfaces but different cores, or different sizes and shapes.…”