Optimization of Manufacturing Processes for Chinese Hamster Ovary (CHO) Cells Through the Application of Machine Learning

The objective of this Capstone project was to investigate the diverse applications of Machine Learning (ML) in the optimization of CHO cell bioprocesses. This investigation entailed a comprehensive review of multiple scholarly studies that employed an array of ML techniques, including Artificial Neural Networks (ANNs), Convolutional Neural Networks (CNNs), and Recurrent Neural Networks (RNNs). Each of these areas within the ML field possesses distinct characteristics that render them suitable for various applications within the bioprocessing domain. The experiments examined within this project evaluated the implementation of ML for protein glycosylation prediction, enhancement of monoclonal antibody titer, optimization of cell culture media, and evaluation of critical quality attributes pertinent to CHO cell bioprocesses. The findings demonstrate the capabilities of ML to aid in process optimization, serving as a foundational step towards the use of ML in CHO cell bioprocessing.