Li Hao, Zhao Yayi, Xu Chenjie
Department of Biomedical Engineering, College of Biomedicine, City University of Hong Kong, Tat Chee Ave, Kowloon, Hong Kong SAR, China.
Institute of Digital Medicine, City University of Hong Kong, Tat Chee Ave, Kowloon, Hong Kong SAR, China.
Nano Converg. 2025 Jul 15;12(1):35. doi: 10.1186/s40580-025-00502-4.
A significant amount of effort has been poured into optimizing the delivery system that is demanded by novel therapeutic modalities. Lipid nanoparticle presents as a solution to transfect cells safely and efficiently with nucleic acid-based therapeutics. Among the components that make up the lipid nanoparticle, ionizable lipids are crucial for the transfection efficiency. Traditionally, the design of ionizable lipids relies on literature search and personal experience. With advancements in computer science, we argue that the use of machine learning can accelerate the design of ionizable lipids systematically. Assuming researchers in lipid nanoparticle synthesis may come from various backgrounds, an entry-level guide is needed to outline and summarize the general workflow of incorporating machine learning for those unfamiliar with it. We hope this can jumpstart the use of machine learning in their projects.
大量的精力投入到优化新型治疗方式所需的递送系统中。脂质纳米颗粒是一种能够安全、高效地用基于核酸的疗法转染细胞的解决方案。在构成脂质纳米颗粒的成分中，可电离脂质对转染效率至关重要。传统上，可电离脂质的设计依赖于文献检索和个人经验。随着计算机科学的进步，我们认为使用机器学习可以系统地加速可电离脂质的设计。鉴于脂质纳米颗粒合成领域的研究人员可能来自不同背景，需要一份入门指南来概述和总结将机器学习应用于不熟悉该技术的人员的一般工作流程。我们希望这能推动他们在项目中使用机器学习。
