Integrating Ensemble NSGA-II for Multi-Objective Process Optimization: Refolding of Proinsulin as a Case Study

Area of Science:

• Biotechnology
• Protein Engineering
• Bioprocess Optimization

Background:

• In vitro refolding of biotherapeutic inclusion bodies is a critical bottleneck in protein production.
• Current refolding optimization methods, like Design of Experiments (DoE), are often inefficient, costly, and labor-intensive.
• Offline analysis and large sampling requirements hinder rapid development.

Purpose of the Study:

• To develop and validate a novel knowledge-based refolding optimization framework for proinsulin.
• To contrast the efficiency and effectiveness of this new method against traditional DoE-based approaches.
• To enhance biotherapeutic production yield and reduce experimental workload.

Main Methods:

• Real-time monitoring of refolding using Fourier Transform Infrared (FTIR), Oxidation Reduction Potential (ORP), and Reverse Phase-High Performance Liquid Chromatography (RP-HPLC).
• Segmentation of the reaction into two distinct phases (0-2h and 2-6h) based on monitored data.
• Application of a multi-objective optimization (MOO) method combining XGBoost and a Non-dominated Sorting Genetic Algorithm II (NSGA-II) optimizer.

Main Results:

• The knowledge-based optimization achieved a yield of 65% ± 1.78%, a 20% relative increase compared to the traditional DoE method (54% ± 2.62%).
• Identified linear correlations between parameters within the segmented reaction phases.
• Demonstrated a combined screening and optimization capability, reducing experimental efforts by approximately 50%.

Conclusions:

• The proposed knowledge-based optimization framework offers a more efficient and effective approach to biotherapeutic refolding.
• This method significantly enhances protein yield and reduces the time and resources required for optimization.
• The integrated approach has broad applicability for optimizing complex bioprocesses.