面向机器学习引导的CHO生物处理和媒体优化,以改善滴度和糖化
Ian Walsh1, Fumi Shozui2, Ayaka Sato2
1Bioprocessing Technology Institute, A*STAR, Singapore.
Biotechnology journal
|October 31, 2025
概括
机器学习模型准确地预测了中国仓鼠卵巢 (CHO) 细胞培养标位和糖化. 这种ML增强的实验设计 (DOE) 通过确定优化关键质量属性的关键因素来加速生物过程开发.
科学领域:
- 生物技术是生物技术.
- 生物加工是一种生物加工.
- 计算生物学 计算生物学
背景情况:
- 在生物处理中控制关键质量属性 (CQAs),如标位和糖化,至关重要,但由于复杂的介质和过程参数而具有挑战性.
- 传统的实验设计 (DOE) 方法与生物制造设计空间的高维和非线性性质作斗争.
研究的目的:
- 开发一种使用机器学习 (ML) 的计算工作流程,以预测在中国仓鼠卵巢 (CHO) 食批量培养中关键的CQA.
- 确定具有影响力的特征并生成组合媒体设计,以优化生物过程.
- 为了加速CHO工艺的开发,并有效地探索复杂的生物制造参数空间.
主要方法:
- 组装了一个全面的以糖甘为重点的CHO养批量数据集.
- 训练有素的ML模型来预测标位和主要的甘氨酸指标 (mannosylation,fucosylation,galactosylation).
- 采用混合的ML + 基于知识的策略来选择特征,并使用模拟回火的ML代理模型来进行主动学习.
主要成果:
- ML模型直接从初始介质和工艺参数获得了定位 (R2 ≈ 0.93) 和糖化 (R2 ≈ 0.79-0.95) 的高预测精度.
- 确定了15个可操作的特征,影响了基位和糖化,独立于核酸糖补充剂.
- 成功提出了一种介质组成和工艺参数组合,该组合可以将曼诺基化减少10%,同时增加标位.
结论:
- 通过ML增强的DOE显著加快了CHO流程开发,并优化了关键的质量属性.
- 开发的计算工作流程有效地探索复杂的生物制造设计空间.
- 这种方法可以精确控制生物过程的结果,提高效率和产品质量.
相关概念视频
Bioreactor Controls-I
Maintaining optimal conditions within fermenters is essential for maximizing microbial productivity and ensuring process efficiency. This lesson focuses on key parameters—temperature, foam, pH, carbon dioxide, oxygen, and pressure—and their precise measurement and control strategies in fermentation systems.Temperature ControlTemperature regulation is critical due to the exothermic nature of many fermentation processes. In small laboratory fermenters, temperature is commonly monitored using...
Bioreactor Controls-II
In aerobic fermentations, oxygen is vital for microbial growth and metabolite production. Since air comprises only about 20% oxygen and the gas is poorly soluble in water—just 9 ppm at 20°C—supplying sufficient oxygen becomes a critical challenge, especially in high-demand processes like yeast growth or citric acid production. Even a fully saturated broth may offer only a few seconds of oxygen availability.To address this, sterile or scrubbed air is introduced into the fermentor via a sparger...
Bioreactor Controls-III
Strain improvement is a foundational strategy in industrial microbiology aimed at maximizing microbial productivity, particularly because natural isolates typically yield commercially valuable products in very low concentrations. Although optimizing the culture medium and environmental conditions can improve yields, these adjustments are inherently limited by the organism’s genetic potential. As a result, the focus shifts toward genetic modifications to enhance biosynthetic capacity. The...
Methods of Medium Optimization
Optimizing growth media enhances microbial proliferation and maximizes product yield. Statistical experimental design methodologies provide structured and reproducible approaches, offering progressively higher levels of robustness and efficiency.The One-Factor-at-a-Time (OFAT) MethodThe One-Factor-at-a-Time (OFAT) method involves adjusting a single variable while keeping all others constant. However, it cannot detect interactions between variables, often leading to suboptimal outcomes when...
Upstream Processing
Upstream processing represents a critical phase in biomanufacturing, wherein biological systems such as microorganisms, mammalian cells, or insect cells are cultivated to produce therapeutic proteins, vaccines, enzymes, or other biologically derived products. This phase encompasses all steps from the selection and genetic manipulation of the production organism to the cultivation of cells in bioreactors under tightly controlled environmental conditions.Host Selection and Genetic OptimizationThe...
Production of Alcohol
Continuous fermentation is a key strategy in industrial ethanol production, particularly when efficiency, scalability, and high yields are essential. This approach allows for uninterrupted operation and optimized resource utilization. The primary feedstock, corn starch, undergoes enzymatic hydrolysis facilitated by α-amylase and glucoamylase. These enzymes break down the starch into fermentable sugars such as glucose, which are readily assimilated by fermentative microorganisms.Fermentation...


