Optimizing Protein Titer Production Using Animal Cells: Predictive Modeling and Recommendations for Enhanced Yield

Current biomanufacturing processes rely heavily on human expertise, struggling to adapt to the growing complexity of bioprocessing. Decision support tools based on machine learning models play a vital role in optimizing the production and timely detection of anomalies in the production line and can save thousands of dollars and reduce quality control costs. We have performed a thorough analysis on a pharmaceutical dataset to identify significant variables that can affect protein titer production. Based on this analysis, we develop machine learning-based models for production of the protein titer in the biopharmaceutical/biotechnology-based manufacturing industry. We envision the practical applications of our models in other biopharmaceutical and biotechnology industries, leading to increased productivity and cost-effectiveness in protein production within animal cells.