Monoclonal antibodies are complex biological products, and glycosylation and other critical quality attributes (CQAs) can significantly influence their efficacy, stability, and pharmacokinetics. These attributes are sensitive to the metabolic state of the production cell and bioprocess conditions, making their prediction and control an important challenge in biomanufacturing. This project will develop mechanistic and data-driven models to predict mAb glycosylation and CQAs from cellular metabolism and process variables. Genome-scale metabolic models, metabolomics, glycosylation pathway models, and machine learning will be integrated and validated using CHO cell culture, metabolomics, and glycomics data. The goal is to enable predictive and rational optimization of bioprocesses for consistent mAb quality.