News Express: UM develops new AI model for virtual screening to enhance mRNA lipid nanoparticle delivery systems
新聞快訊：澳大新mRNA篩選AI模型提高藥物研發效率
人工智能驅動用於mRNA遞送的可電離脂質的合理設計流程圖
An overview of AI-driven rational design of ionisable lipids for mRNA lipid nanoparticles
澳大新mRNA篩選AI模型提高藥物研發效率
澳門大學中華醫藥研究院、中藥質量研究國家重點實驗室、健康科學學院副教授歐陽德方的研究團隊開發了用於mRNA脂質納米粒（LNP）遞送體系虛擬篩選的人工智能模型，可提高藥物研發效率。相關研究成果以“人工智能驅動用於mRNA遞送的可電離脂質的合理設計”為題發表於國際學術期刊《自然通訊》，並已申請了國際專利。
COVID-19 mRNA疫苗的成功證明脂質納米粒在mRNA遞送方面是極有效的。然而，LNP其中的關鍵成分——可電離脂質，傳統上需要通過低效且成本高昂的實驗篩選進行優化。因此，研究團隊開發了人工智能模型，預測LNP的關鍵屬性——表觀pKa和mRNA遞送效率，通過虛擬篩選促進可電離脂質的合理設計。在兩輪人工智能模型驅動的生成和篩選中，共評估了近2,000萬個可電離脂質，分別產生了三個和六個新分子。在小鼠測試驗證中，第一輪篩選獲得的一個含有苯環的脂質表現出與對照分子DLin-MC3-DMA（MC3）相當的性能，而第二輪篩選獲得的所有六個脂質都達到了或超過了MC3的性能，其中一個脂質的效果類似於一個更優的對照脂質SM-102。此外，研究團隊開發的人工智能模型在結構—活性關係上具有可解釋性。這項研究證明了人工智能技術在mRNA藥物產品開發中的巨大潛力，可提高藥物研發效率。
該研究由澳門大學與復旦大學團隊合作完成。歐陽德方為該研究的共同第一通訊作者，澳大博士後王維與博士生吳依洋為共同第一作者。該研究獲得澳門大學項目（檔案編號：MYRG-CRG2022-00008-ICMS、ICMS/RTO/EP160/2023）， 澳門特別行政區科學技術發展基金（檔案編號：0071/2024/RIA1、005/2023/SKL），深圳市科技創新委員會（檔案編號：SGDX20210823103802016），珠海市科技創新局（檔案編號：ZH22017002210010PWC），國家自然科學基金（檔案編號：32301174）的共同資助。研究亦獲上海張江mRNA國際創新中心、復旦大學mRNA藥物轉化中心、上海藍鵲生物醫藥有限公司的支持。完整文章可查閱：https://www.nature.com/articles/s41467-024-55072-6。
欲瀏覽官網版可登入以下連結:
https://www.um.edu.mo/zh-hant/news-and-press-releases/press-release/detail/60209/
UM develops new AI model for virtual screening to enhance mRNA lipid nanoparticle delivery systems
The research team led by Ouyang Defang, associate professor in the Institute of Chinese Medical Sciences, the State Key Laboratory of Quality Research in Chinese Medicine, and the Faculty of Health Sciences at the University of Macau (UM), has developed an artificial intelligence (AI) model for virtual screening of mRNA lipid nanoparticle (LNP) delivery systems. This breakthrough is set to increase the efficiency of AI-driven mRNA drug development. The research, titled ‘Artificial intelligence-driven rational design of ionisable lipids for mRNA delivery’, has been published in the international journal Nature Communications, and an international patent has been filed.
LNPs have proven effective for mRNA delivery, as demonstrated by the COVID-19 vaccines. However, their key component, ionisable lipids, has traditionally been optimised through inefficient and costly experimental screening. Therefore, the research team developed an AI model and used virtual screening to facilitate the rational design of ionisable lipids by predicting two key properties of LNPs, apparent pKa and mRNA delivery efficiency. Nearly 20 million ionisable 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, showed comparable performance to the control DLin-MC3-DMA (MC3). Notably, all six lipids from the second iteration equalled or outperformed MC3, with one exhibiting efficacy similar to a superior control lipid SM-102. Furthermore, the AI model is interpretable in terms of structure-activity relationships. This study demonstrates the immense potential of AI technology in the development of mRNA drug products, which can improve the efficiency of drug development.
The study was conducted in collaboration between UM and Fudan University. Ouyang Defang is the co-first corresponding author of the study, with UM postdoctoral fellow Wang Wei and PhD student Wu Yiyang as co-first authors. The research was supported by UM (File No: MYRGCRG2022-00008-ICMS, ICMS/RTO/EP160/2023), Science and Technology Development Fund of the Macao SAR (File No: 0071/2024/RIA1, 005/2023/SKL), Shenzhen Science and Technology Innovation Commission (File No: SGDX20210823103802016), Zhuhai Science and Technology Innovation Bureau (File No: ZH22017002210010PWC), National Natural Science Foundation of China (File No: 32301174). The research was also supported by the Shanghai Zhangjiang mRNA Innovation and Translation Center, Fudan Center for mRNA Translational Research, and Shanghai RNACure. The full text of the study is available at https://www.nature.com/articles/s41467-024-55072-6.
To read the news on UM’s official website, please visit the following link:
https://www.um.edu.mo/news-and-press-releases/press-release/detail/60209/