机器学习驱动的优化mRNA-脂质纳米粒子疫苗质量与XGBoost/贝叶斯方法和组合模型方法
Ravi Maharjan1, Ki Hyun Kim1,2, Kyeong Lee1
1BK21 FOUR Team and Integrated Research Institute for Drug Development, College of Pharmacy, Dongguk University, Gyeonggi, 10326, Republic of Korea.
Journal of pharmaceutical analysis
|January 6, 2025
概括
这项研究优化了使用机器学习的信使RNA-脂质纳米粒子 (mRNA-LNP) 生产. 自验证组合 (SVEM) 模型准确预测了最佳的脂质比率,以有效制造疫苗.
科学领域:
- 生物技术是生物技术.
- 制药科学 制药科学
- 材料科学 材料科学 材料科学
背景情况:
- 有效的疫苗制造依赖于精确控制纳米粒子配方.
- 传递 RNA-脂质纳米颗粒 (mRNA-LNP) 对于疫苗的输送至关重要,需要优化流程.
- 目前优化mRNA-LNP生产的方法可能耗时且资源密集.
研究的目的:
- 优化微流体条件和脂质混合比,以提高mRNA-LNP生产效率.
- 评估和比较机器学习模型,以预测最佳的mRNA-LNP配方.
- 确定影响mRNA-LNP特性的关键材料和过程属性.
主要方法:
- 使用I-最佳设计开发了24种不同的mRNA-LNP配方.
- 采用机器学习工具,包括XGBoost/贝叶斯优化和自验证组合 (SVEM),用于流程优化和脂质混合比率预测.
- 评估了关键反应,例如粒子大小 (PS),多分散度指数 (PDI),泽塔潜力,封装效率 (EE) 和回收率.
主要成果:
- 与XGBoost/贝叶斯优化 (>94%) 相比,SVEM模型显示出更高的预测准确度 (>97%).
- 实验验证证证实了SVEM的预测,实际粒子大小与预测值非常接近 (例如94-96nm与95-97nm).
- 像PDI和EE这样的关键参数也显示了SVEM预测和实验结果之间的密切一致.
结论:
- 机器学习,特别是SVEM模型,为优化mRNA-LNP生产提供了一个高度准确的方法.
- 优化的微流体条件和脂质比率可以显著提高疫苗制造效率.
- 这一数据驱动的战略有助于为疫苗应用开发高质量的mRNA-LNP配方.
更多相关视频
08:55Testing the In Vitro and In Vivo Efficiency of mRNA-Lipid Nanoparticles Formulated by Microfluidic Mixing
Published on: January 20, 2023
10.6K
13:54A Workflow for Lipid Nanoparticle LNP Formulation Optimization using Designed Mixture-Process Experiments and Self-Validated Ensemble Models SVEM
Published on: August 18, 2023
4.3K
相关概念视频
Improving Translational Accuracy
Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...
Improving Translational Accuracy
Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...
Vaccine Production
Vaccine production involves a sequence of upstream and downstream processes to generate a safe and effective immunological product. It begins with cultivating microorganisms, such as viruses or bacteria, to obtain antigenic material. For viral vaccines, mammalian host cells are grown in bioreactors and subsequently infected with the target virus. The virus replicates within the host cells, which are lysed to release viral particles. This lysate is then clarified through filtration or...
