Advanced Bioprocessing Sensor and Analytical Technologies for Induced Pluripotent Stem Cell Culture Online Monitoring and Automation

This project aims to develop a robust online sensor to monitor the redox state; and build a knowledge graph hybrid model-based machine learning (KG-ML) algorithm to advance the scientific understanding of metabolism.

Project status

100% Completed

Industry Need

Large-scale production of induced pluripotent stem cells (iPSCs) is essential for cell therapies and regenerative medicines, but iPSCs’ productivity and pluripotency are highly sensitive to culture conditions.
Subtle changes in culture conditions can lead to stress which can result in cell populations with heterogeneous differentiation potential.

Solution

Develop a robust online sensor to monitor the redox state
Build a knowledge graph hybrid model-based machine learning (KG-ML) algorithm to advance the scientific understanding of metabolism by correlating online sensing and the intracellular metabolic state of the culture process
This model-based monitoring and control will safeguard that the culture follows the expected trajectory for successful growth and expansion.

To achieve these goals and facilitate large-scale production of iPSCs, the team will conduct the following tasks:

Refine the two-photon excitation (TPE) fluorescence optical sensor to iPSC cultures for real-time measurement of the intracellular redox state
Gain knowledge of iPSC metabolism in static and bioreactor cultures with aggregates and on microcarriers
Extend the KG-ML framework to improve the prediction and control of iPSCs outcomes

Outputs/Deliverables

Coming Soon

Impacts

Improved capability to assess iPSC intracellular metabolic state in real time

Advanced understanding of iPSC metabolism in static and bioreactor cultures with aggregates and on microcarriers

Reliable TPE sensor developed with predictive power for iPSC pluripotency, growth rate, and productivity (CQAs)

Improved prediction and control of CQAs and outcomes for iPSC culture process using online sensing and KG-ML framework

Publications

Wang, K., Xie, W., & Harcum, S. W. (2024). Metabolic regulatory network kinetic modeling with multiple isotopic tracers for iPSCs. Biotechnology and Bioengineering, 121(4), 1336–1354. https://doi.org/10.1002/bit.28609

Zheng, H., Harcum, S. W., Pei, J., & Xie, W. (2024). Stochastic biological system-of-systems modelling for iPSC culture. Communications Biology, 7(1), 39–w. https://doi.org/10.1038/s42003-023-05653-w

Zheng, H., Xie, W., Ryzhov, I. O., & Xie, D. (2022). Policy Optimization in Dynamic Bayesian Network Hybrid Models of Biomanufacturing Processes. INFORMS Journal on Computing, 35(1), 66–82. https://doi.org/10.1287/ijoc.2022.1232

Wang, B., Xie, W., Martagan, T., Akcay, A., & Ravenstein, B. van. (2024). Biomanufacturing Harvest Optimization With Small Data. Production and Operations Management, 33(12), 2381-2400. https://doi.org/10.1177/10591478241270130 (Original work published 2024)