Cell culture plays a crucial role in the production of biologics. When introducing process changes as part of a design of experiment (DOE), accurately modeling the behavior of the cell culture process is challenging as the process involves multiple interdependent growth and production phases, only some of which may be impacted by process changes. Traditional parametric non-linear models struggle to effectively capture this complexity, while non-parametric models alone can be disjointed and difficult to correlate with DOE parameters.

To address this issue, functional DOE simplifies the complexity into principal components and correlates the changes with DOE parameters. This approach enables the creation of a prediction profiler, which can optimize cell culture parameters from small scale data and use them to predict behavior during larger-scale production. The entire process can be performed within the Functional Data Explorer Platform in JMP Pro and can provide a more efficient approach for optimizing cell culture processes.