Machine Learning To Accelerate Antibody Manufacturing

Monoclonal antibodies are lab-made proteins used to treat various diseases but face a manufacturing bottleneck.
Researchers at the University of Oklahoma have developed a machine learning model to accelerate the manufacturing timeline of monoclonal antibodies.
The model aims to identify high-yield cell lines more quickly, which could reduce production costs and time.
The machine learning model correctly identified high-performing clones 76.2% of the time and accurately forecasted production trajectories.
The research includes a partnership with Wheeler Bio, a contract development and manufacturing organization focused on antibody therapies.
The project is part of a $35 million initiative by the U.S. Economic Development Administration to enhance Oklahoma’s biotechnology industry.
The study emphasizes the combination of academic research and industrial application to address real-world challenges in biopharmaceutical manufacturing.