Implementing Scale Dependencies into Computational Bioreactor Models to Facilitate Predictions of Protein Glycosylation in Different Environments

This project will deliver an interoperable simulation platform integrating spatially resolved CFD flow fields, cell kinetics, and mechanistic models of intracellular glycosylation.

Industry Need

Although significant progress has been made in modeling glycosylation at small scales, scaling up to production bioreactors introduces complex gradients in nutrients, oxygen, pH, and metabolite concentrations that affect cellular performance, ultimately impacting product properties including glycosylation. This project addresses a critical need in upstream bioprocess modeling: predicting the impact of hydrodynamic heterogeneity on glycosylation outcomes during CHO based protein production for different scales and bioreactor configurations.

Approach

This proposal brings together researchers in mammalian hydrodynamics and systems biology of glycosylation to deliver an interoperable simulation platform integrating spatially resolved CFD flow fields, cell kinetics, and mechanistic models of intracellular glycosylation. The platform will help NIIMBL partners build digital twins, accelerate process development, validate glycosylation robustness, and inform scale-up decisions. This platform can be extended in the future to include automated CFD routines and machine learning-assisted parameter optimization. The tool will support both research and industrial applications, advancing NIIMBL’s mission to accelerate biomanufacturing innovation.

Impacts

The platform will help NIIMBL partners build digital twins, accelerate process development, validate glycosylation robustness, and inform scale-up decisions.

This platform can be extended in the future to include automated CFD routines and machine learning-assisted parameter optimization. The tool will support both research and industrial applications, advancing NIIMBL’s mission to accelerate biomanufacturing innovation.

Value Statement/Outcomes

This project seeks to demonstrate a computational bioreactor modeling platform that improves scale-up predictions for drug companies, resulting in quality benefits across large-scale manufacturing.

Outputs/Deliverables

Deliverables include Python software packages, step-by-step example workflows, training materials, and graphical output of key metrics (e.g., glycan distribution over time, pH and oxygen maps).