Machine learning-based prediction of optimal fermentation conditions for soluble protein production in periplasm of Escherichia coli

Abstract

Maximizing the expression of recombinant proteins in E. coli has tremendous research interest. To date, studies have attempted optimizing factors relating to either gene expression (expression-level) or fermentation process (process-level) conditions to achieve high yields of RPP. However, understanding the combinatorial influence of expression and process-level factors is crucial for achieving the desired protein yields. This thesis demonstrates that a machine learning model based on expression and process levels can effectively predict the optimized fermentation conditions. The developed tool will enable researchers to predict optimal conditions for maximal recombinant protein production, reducing the time-consuming and expensive trial and error experiments.