CHO Cell Bioreactor Model: Fed-batch to Continuous Transfer

bioRxiv (preprint)

Overview

Researchers have developed a metabolism-informed model for monoclonal antibody (mAb) production that enables the transfer of knowledge from fed-batch to continuous biomanufacturing. By embedding mass balances into a learning architecture, the model serves as a soft sensor to infer intracellular fluxes and forecast extracellular concentrations. This approach bridges the gap between mechanistic understanding and real-time predictive control in continuous cultures.

In Depth

Key Findings

The metabolism-informed model demonstrates that transferring knowledge from fed-batch to continuous operation via fine-tuning outperforms other transfer learning variants. Analysis reveals that the growth rate in fed-batch, rather than the operating mode itself, is the dominant axis of metabolic variation. Additionally, the model identifies potential regime-shift signals that can guide experimental design for the characterization of continuous cultures.

Technical Details

The model embeds reactor- and intracellular-level mass balances directly within a learning architecture. To maintain flexibility, the design deliberately avoids explicit steady-state assumptions and predefined kinetic rate laws. This structure relies on the fact that conservation relationships remain invariant across different operating modes, providing a structural prior for cross-regime generalization.

The researchers benchmarked the system using three metabolic optimization strategies and two different metabolic network sizes. The results indicated that:

• Concentration predictions remained indistinguishable regardless of the network size.
• Exchange flux fidelity was agnostic to the choice of solver.
• Network size only impacted the level of intracellular interpretability.

Background & Context

Chinese hamster ovary (CHO) cells are the primary platform for mAb manufacturing. While the industry is shifting from fed-batch toward continuous operation to achieve sustained high cell densities and volumetric productivities, continuous processes require reliable real-time monitoring and control. Current efforts are hindered by sparse extracellular measurements and limited intracellular observability.

Significance & Outlook

By functioning as a soft sensor, this model allows for the inference of unmeasured intracellular fluxes and one-step ahead forecasting of extracellular concentrations. This methodology provides a route toward digital models that are grounded in first-principles knowledge while remaining adaptive enough for deployment under regime shifts in biomanufacturing.

Get our weekly technology intelligence — free

Receive an infographic that lets you judge at a glance whether each field’s analysis report is worth reading.

Subscribe Free — Weekly Tech Intelligence

By subscribing, you’ll receive Troy-Technical’s weekly technology intelligence newsletter.

• Your email and selected fields are used only to deliver the newsletter.
• We never share your information with third parties.
• You can unsubscribe anytime via the link in each email.

See our Privacy Policy for details.

Agree & Continue ▶

Takes about a minute · Unsubscribe anytime