Development of meaningful and reliable analytical assays in the (bio)pharmaceutical industry can often be challenging, involving tedious trial and error experimentation. In this work, an automated analytical workflow using an AI-based algorithm for streamlined method development and optimization is presented. Chromatographic methods are developed and optimized from start to finish by a feedback-controlled modeling approach using readily available LC instrumentation and software technologies, bypassing manual user intervention. With the use of such tools, the time requirement of the analyst is drastically minimized in the development of a method. Herein key insights on chromatography system control, automatic optimization of mobile phase conditions, and final separation landscape for challenging multicomponent mixtures are presented ( small molecules drug, peptides, proteins, and vaccine products) showcased by a detailed comparison of a chiral method development process. The work presented here illustrates the power of modern chromatography instrumentation and AI-based software to accelerate the development and deployment of new separation assays across (bio)pharmaceutical modalities while yielding substantial cost-savings, method robustness, and fast analytical turnaround.

中文翻译：

---

使用反馈控制液相色谱优化改进药物模式的测定开发

在（生物）制药行业中开发有意义且可靠的分析方法通常具有挑战性，涉及繁琐的试错实验。在这项工作中，提出了一种使用基于人工智能的算法来简化方法开发和优化的自动化分析工作流程。色谱方法的开发和优化自始至终采用反馈控制建模方法，使用现成的 LC 仪器和软件技术，绕过了用户手动干预。通过使用此类工具，分析人员在方法开发过程中所需的时间大大减少。本文通过对手性方法开发过程的详细比较，提出了关于色谱系统控制、流动相条件自动优化以及具有挑战性的多组分混合物（小分子药物、肽、蛋白质和疫苗产品）的最终分离情况的关键见解。这里介绍的工作说明了现代色谱仪器和基于人工智能的软件的力量，可以加速跨（生物）制药模式的新分离测定的开发和部署，同时实现大量成本节约、方法稳健性和快速分析周转。