Multivariate statistical data analysis of cell‐free protein synthesis toward monitoring and control

Carlos A Duran‐Villalobos,Trevor Hallam,Beatrice Melinek,Barry Lennox,Daniel G Bracewell,Olotu Ogonah

doi:10.1002/aic.17257

Abstract

AbstractThe optimization and control of cell free protein synthesis (CFPS) presents an ongoing challenge due to the complex synergies and nonlinearities that cannot be fully explained in first principle models. This article explores the use of multivariate statistical tools for analyzing data sets collected from the CFPS of Cereulide monoclonal antibodies. During the collection of these data sets, several of the process parameters were modified to investigate their effect on the end‐point product (yield). Through the application of principal component analysis and partial least squares (PLS), important correlations in the process could be identified. For example, yield had a positive correlation with pH and NH3 and a negative correlation with CO2 and dissolved oxygen. It was also found that PLS was able to provide a long‐term prediction of product yield. The presented work illustrates that multivariate statistical techniques provide important insights that can help support the operation and control of CFPS processes.

Highlights

The integration of machine learning techniques into the operation and control of cell free protein synthesis (CFPS) systems[1] offers significant potential for improving productivity and the quality of materials manufactured using this relatively new processing technique
The specific case study for the analyses presented in this article is the scalable cell-free synthesis of monoclonal antibodies, using the cell-free lysate system developed by Sutro Biopharma.[24,25]
The colored dots in this figure represent the value in the score of the objects in the PC space: If the observation markers lie in the same quadrant of a blue line in the loadings plot, it suggests a strong association with that variable

Summary

Introduction

The integration of machine learning techniques into the operation and control of cell free protein synthesis (CFPS) systems[1] offers significant potential for improving productivity and the quality of materials manufactured using this relatively new processing technique. Models used to describe CFPS production processes and in particular its application to quality by design can only be useful in practice if sufficient process knowledge is available to explain the effect of critical process parameters on critical quality attributes. In this respect, mechanistic models offer great value in determining causality to support the optimization of CFPS processes[5,6]; the development of such models requires significant time and resources which typically make them impractical.[7] In addition, CFPS relies on a complex network of interacting reactions, reactants, and enzymatic catalysts, which are not yet fully understood.

Objectives

Methods

Results

Conclusion