EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Toward Machine Learning‐Guided CHO Bioprocess and Media Optimization for Improved Titer and Glycosylation

Ian Walsh, Fumi Shozui, Ayaka Sato, Uiseon Park, Hyeran Cho, Charmaine Chia iD

Biotechnology Journal · 2025-11 · 20 (11)

DOI: 10.1002/biot.70149 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biotechnology Journal, online erschienen am 31.10.2025. Verfasst von Ian Walsh und 5 weiteren Autor:innen. Beteiligte Einrichtung: Bioprocessing Technology Institute, A*STAR, Singapore. Hauptthemen laut MEDLINE: Machine Learning, Culture Media. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Precise control of critical quality attributes, including titer and glycosylation, is essential in bioprocessing, yet conventional design‑of‑experiments methods are challenged by the high‐dimensional, nonlinear design space for media and process parameters. We assemble a comprehensive glycan‑focused Chinese hamster ovary (CHO) fed‑batch dataset and develop a computational workflow (i) to train machine learning (ML) models to predict key CQAs, (ii) apply a hybrid ML + knowledge‐based strategy to select potentially impactful features, and (iii) generate combinatorial media designs. The resulting models predict final titer ( R 2 ≈ 0.93) and major glycan metrics—mannosylation, fucosylation, galactosylation ( R 2 ≈ 0.79–0.95)—directly from initial media composition and process parameters without requiring spent media analysis. Feature selection shortlisted 20 features out of 76 for a second‐tier validation, from which 15 were confirmed as actionable levers impacting titer and glycosylation, uncovering glycan effects independent of nucleotide sugar supplementation. Finally, we incorporated our workflow, utilizing a ML surrogate model coupled with simulated annealing, in a proof‑of‑concept active learning step, successfully proposing a media composition and process parameter combination that reduced mannosylation by 10% while increasing titer. Together, these results underscore how ML‑enhanced DOE can accelerate CHO process development and explore complex biomanufacturing spaces with greater efficiency.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Machine LearningCulture MediaCHO CellsAnimalsCricetulusGlycosylationPolysaccharidesCricetinaeBatch Cell Culture Techniques

Stichwörter der Autor:innen

CHO cell culturebioprocess optimizationcritical quality attributesdesign of experimentsfeature selectionglycosylationmachine learningsimulated annealingtiter

Substanzen

Culture MediaPolysaccharides

Autor:innen und Institutionen

  • Bioprocessing Technology Institute, A*STAR, Singapore.
  • Ajinomoto Co., Inc., Chuo, Japan.
  • Ajinomoto CELLiST Korea Co., Inc., Korea.
  • Bioinformatics Institute, A*STAR, Singapore.

ORCID-Profile: Charmaine Chia

Zitieren

APA

Walsh, I., Shozui, F., Sato, A., Park, U., Cho, H., & Chia, C. (2025). Toward Machine Learning‐Guided CHO Bioprocess and Media Optimization for Improved Titer and Glycosylation. Biotechnology Journal, 20(11). https://doi.org/10.1002/biot.70149

Vancouver

Walsh I, Shozui F, Sato A, Park U, Cho H, Chia C. Toward Machine Learning‐Guided CHO Bioprocess and Media Optimization for Improved Titer and Glycosylation. Biotechnology Journal. 2025;20(11). doi:10.1002/biot.70149

BibTeX
@article{Walsh2025,
  author = {Walsh, Ian and Shozui, Fumi and Sato, Ayaka and Park, Uiseon and Cho, Hyeran and Chia, Charmaine},
  title = {Toward Machine Learning‐Guided CHO Bioprocess and Media Optimization for Improved Titer and Glycosylation},
  journal = {Biotechnology Journal},
  year = {2025},
  volume = {20},
  number = {11},
  doi = {10.1002/biot.70149}
}
Kostenlos starten →