EUVIMEDEUVIMED entdecken
Crossref

Innovating cell culture process development with deep learning‐powered robotic experimentation using the first I ndustrial S mart L ab F ramework

Shuting Xu, Yanting Huang, Xin Shen iD, Rongjia Mao, Yiming Song, Wanying Ye, Lijun Wang, Xiaoxiao Tong, Yun Cao, Ruiqiang Sun iD, Hang Zhou, Weichang Zhou

Biotechnology Progress · 2025

DOI: 10.1002/btpr.70051 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biotechnology Progress, 2025. Verfasst von Shuting Xu und 11 weiteren Autor:innen. Beteiligte Einrichtung: Cell Culture Process Development (CCPD), WuXi Biologics, Shanghai, China. Hauptthemen laut MEDLINE: Deep Learning, Robotics, Cell Culture Techniques. Die Arbeit nennt 39 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Traditional biologics process development, including antibody and recombinant protein production, typically relies on labor‐intensive, iterative cell culture optimization to determine optimal process parameters. To address this inefficiency, we introduced the Industrial Smart Lab Framework for Cell Culture (ISLFCC), an autonomous laboratory that combines deep learning and robotic experimentation to enhance cell culture processes. In this system, robotic arms sample various bioreactors for analysis, and the IoT system transmits these analysis results to decoder‐only transformer deep learning models. Based on these analysis results, these models predict future cell states and recommend optimal actions, which are then executed by automation devices through the IoT system, such as adjusting nutrient feeds and temperature shifts. In a comparative case study, our AI‐driven process development for three different cell clones resulted in an average titer increase of 26.8% and maintained lactate levels below 1 g/L without rebound in the late phase within just a single batch, surpassing traditional three‐stage empirical process development methods. Moreover, our approach has greatly automated cell culture to ensure enhanced reproducibility, data accuracy, adaptability to various cell lines, and seamless scalability across production scales, marking the first implementation of high‐throughput automated cell culture in 3 and 15 L bioreactors. By merging AI with robotic execution, ISLFCC provides a transformative framework that accelerates biologics development, representing a paradigm shift towards autonomous, data‐driven biomanufacturing.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningRoboticsCell Culture TechniquesBioreactorsAnimalsCHO CellsCricetulus

Stichwörter der Autor:innen

IoTcell culturedeep learningdigital twinprocess developmentrobotic experimentationsmart lab

Autor:innen und Institutionen

  • Cell Culture Process Development (CCPD), WuXi Biologics, Shanghai, China.

ORCID-Profile: Xin Shen, Ruiqiang Sun

Zitieren

APA

Xu, S., Huang, Y., Shen, X., Mao, R., Song, Y., Ye, W., Wang, L., Tong, X., Cao, Y., Sun, R., Zhou, H., & Zhou, W. (2025). Innovating cell culture process development with deep learning‐powered robotic experimentation using the first I ndustrial S mart L ab F ramework. Biotechnology Progress. https://doi.org/10.1002/btpr.70051

Vancouver

Xu S, Huang Y, Shen X, Mao R, Song Y, Ye W, et al. Innovating cell culture process development with deep learning‐powered robotic experimentation using the first I ndustrial S mart L ab F ramework. Biotechnology Progress. 2025. doi:10.1002/btpr.70051

BibTeX
@article{Xu2025,
  author = {Xu, Shuting and Huang, Yanting and Shen, Xin and Mao, Rongjia and Song, Yiming and Ye, Wanying and Wang, Lijun and Tong, Xiaoxiao and Cao, Yun and Sun, Ruiqiang and Zhou, Hang and Zhou, Weichang},
  title = {Innovating cell culture process development with deep learning‐powered robotic experimentation using the first I ndustrial S mart L ab F ramework},
  journal = {Biotechnology Progress},
  year = {2025},
  doi = {10.1002/btpr.70051}
}
Kostenlos starten →