Wei Wang

Lipid nanoparticles (LNPs) have proven effective in mRNA delivery, as evidenced by COVID-19 vaccines. Its key ingredient, ionizable lipids, is traditionally optimized by inefficient and costly experimental screening. This study leverages artificial intelligence (AI) and virtual screening to facilitate the rational design of ionizable lipids by predicting two key properties of LNPs, apparent pKa and mRNA delivery efficiency. Nearly 20 million ionizable lipids were evaluated through two iterations of AI-driven generation and screening, yielding three and six new molecules, respectively. In mouse test validation, one lipid from the initial iteration, featuring a benzene ring, demonstrated performance comparable to the control DLin-MC3-DMA (MC3). Notably, all six lipids from the second iteration equaled or outperformed MC3, with one exhibiting efficacy akin to a superior control lipid SM-102. Furthermore, the AI model is interpretable in structure-activity relationships. Ionizable lipid optimization is essential for mRNA therapy via lipid nanoparticle, but experimental screening is investment-intensive. Here, authors developed AI models achieving the rational design of lipid molecules and fast high-throughput screening.

Lipid nanoparticle (LNP) is commonly used to deliver mRNA vaccines. Currently, LNP optimization primarily relies on screening ionizable lipids by traditional experiments which consumes intensive cost and time. Current study attempts to apply computational methods to accelerate the LNP development for mRNA vaccines. Firstly, 325 data samples of mRNA vaccine LNP formulations with IgG titer were collected. The machine learning algorithm, lightGBM, was used to build a prediction model with good performance (R2 > 0.87). More importantly, the critical substructures of ionizable lipids in LNPs were identified by the algorithm, which well agreed with published results. The animal experimental results showed that LNP using DLin-MC3-DMA (MC3) as ionizable lipid with an N/P ratio at 6:1 induced higher efficiency in mice than LNP with SM-102, which was consistent with the model prediction. Molecular dynamic modeling further investigated the molecular mechanism of LNPs used in the experiment. The result showed that the lipid molecules aggregated to form LNPs, and mRNA molecules twined around the LNPs. In summary, the machine learning predictive model for LNP-based mRNA vaccines was first developed, validated by experiments, and further integrated with molecular modeling. The prediction model can be used for virtual screening of LNP formulations in the future.

With the increasing utilisation of information technology and artificial intelligence in practical control systems, particularly in large-scale distributed networked systems, growing concerns have arisen regarding the potential disclosure of individuals' sensitive information to adversaries. Consequently, safeguarding privacy security against the rapidly increasing risk of privacy leakages has become a top priority in modern control systems. This survey aims to offer a comprehensive review of privacy-preserving control and filtering problems in networked control system. First, we review some basic introductions to the privacy-preserving mechanisms from the perspective of control community. Then, we present recent advancements in the design of privacy-preserving strategies for various control and filtering problems. Moreover, several possible future research topics are outlined on the privacy-preserving issue of control systems.