Related Experiment Video
Updated: Jul 26, 2025
13:26
Automated Gel Size Selection to Improve the Quality of Next-generation Sequencing Libraries Prepared from Environmental Water Samples
Published on: April 17, 2015
10.6K
Next-generation cell line selection methodology leveraging data lakes, natural language generation and advanced data
Stephen Goldrick1, Haneen Alosert1, Clare Lovelady2
1Department of Biochemical Engineering, University College London, London, United Kingdom.
Frontiers in Bioengineering and Biotechnology
|June 21, 2023
Summary
A new cell line development (CLD) methodology uses digitalization, a manufacturability index (MI), and machine learning for autonomous, data-driven clone selection. This approach improves biopharmaceutical development by identifying risks and optimizing processes for better product quality.
Area of Science:
- Biopharmaceutical Development
- Biotechnology
- Process Engineering
Background:
- Cell line development is critical in biopharmaceutical manufacturing.
- Incomplete clone characterization causes project delays and impacts commercial success.
- Current methods may miss crucial process-related quality issues.
Purpose of the Study:
- To introduce a novel, autonomous, data-driven cell line development (CLD) methodology.
- To enhance lead clone selection through digitalization, advanced analytics, and automated reporting.
- To address limitations in conventional cell line development processes.
Main Methods:
- Digitalization of process data into a structured data lake.
- Calculation of a cell line manufacturability index (MI) for clone performance.
- Application of machine learning (ML) to identify process risks and critical quality attributes (CQAs).
- Automated report generation using natural language generation (NLG).
Main Results:
- The CLD methodology successfully selected a lead clone for a recombinant Chinese hamster ovary (CHO) cell line.
- It identified sub-optimal conditions affecting trisulfide bond (TSB) concentration, a known product quality issue.
- The approach revealed quality issues missed by conventional methods.
Conclusions:
- The CLD methodology, aligned with Industry 4.0 principles, enhances decision-making in biopharmaceutical development.
- Digitalization, data lakes, predictive analytics, and autonomous reporting are key benefits.
- This novel approach leads to more informed and efficient cell line selection and process optimization.
