Applications of machine learning in antibody discovery, process development, manufacturing and formulation - Current trends, challenges, and opportunities

Aug 4, 2026

Upload by Sage

While machine learning (ML) has made significant contributions to the biopharmaceutical field, its support for quality-by-design based development and manufacturing of biologics remains in early stages, despite strong potential for automation. Adoption is accelerating as large-scale production datasets accumulate, driven by real-time monitoring of process variables and quality attributes via advanced process analytical technologies. ML enables accurate, flexible predictive analytics, monitoring, and control across upstream, downstream, and formulation for monoclonal antibodies. The review also details key challenges in processes, data, and model use, and highlights emerging digital biopharma directions.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages59

Reading Time2 h 29 min

AccessFree to download

Introduction

Scale-up and market drivers for biologics and monoclonal antibodies

Definitions and complexity of biologics

Upstream/downstream and manufacturing optimization needs

Computers and Chemical Engineering 182 (2024) 108585 | Review
Applications of machine learning in antibody discovery, process development, manufacturing and formulation: Current trends, challenges, and opportunities | | | | | --- | --- | --- | --- | | Thanh Tung Khuat a,∗, Robert Bassett b, Ellen Otteb, Alistair Grevis-James b, Bogdan Gabrys aa Complex Adaptive Systems Laboratory, The Data Science Institute, University of Technology Sydney, NSW 2007, Australia
b CSL Innovation, Melbourne, VIC 3000, Australia | | | | | A R T I C L E I N F O | | A B S T R A C T | | | Keywords: Biopharmaceuticals Machine learning Upstream Downstream Bioprocesses Digital twin
Soft sensors | | While machine learning (ML) has made significant contributions to the biopharmaceutical field, its applications are still in the early stages in terms of providing direct support for quality-by-design based development and manufacturing of biologics, hindering the enormous potential for bioprocesses automation from their development to manufacturing. However, the adoption of ML-based models instead of conventional multivariate data analysis methods is significantly increasing due to the accumulation of large-scale production data. This trend is primarily driven by the real-time monitoring of process variables and quality attributes of biopharmaceutical products through the implementation of advanced process analytical technologies. Given the complexity and multidimensionality of a bioproduct design, bioprocess development, and product manufacturing data, ML-based approaches are increasingly being employed to achieve accurate, flexible, and high-performing predictive models to address the problems of analytics, monitoring, and control within the biopharma field. This paper aims to provide a comprehensive review of the current applications of ML solutions in the design, monitoring, control, and optimisation of upstream, downstream, and product formulation processes of monoclonal antibodies. Finally, this paper thoroughly discusses the main challenges related to the bioprocesses themselves, process data, and the use of machine learning models in monoclonal antibody process development and manufacturing. Moreover, it offers further insights into the adoption of innovative machine learning methods and novel trends in the development of new digital biopharma solutions. | | 1. Introduction Over the past few years, biopharmaceutical products, also known as biologics, such as monoclonal antibodies (mAbs) and therapeutic proteins have become the best-selling drugs in the pharmaceutical market (Lu et al., 2020), and in 2021, seven of the top ten bestselling drugs worldwide were biologics (Urquhart, 2022) (see Fig. 1). According to the definition given by U.S. Food & Drug Administration (FDA) (2018), biologics include vaccines, monoclonal antibodies, blood and blood components, allergenics, somatic cells, tissues, gene therapy, and other therapeutic proteins. Unlike small molecule and chemically synthesised drugs, biologics are complicated, large mixtures of sugars, proteins or nucleic acids, and other substances which are not easy to identify and exactly characterise (Peters and Hennessey, 2020). Most of the biologics are produced by biotechnology in a living system such as microorganisms, plant, animal, or human cells, in which mammalian cells like the Chinese hamster ovary (CHO) cells, mouse myeloma (NS0), baby hamster kidney (BHK), human embryo kidney (HEK-293) and human retinal cells are typically used (Wurm, 2004). In recent years, the market for biologics has explosively grown with a percentage of new biological products approved by FDA every year since 2014 for treating various human diseases including cancers, autoimmune, metabolic and infectious diseases, always exceeding 20% of the total number of new approved drugs (De la Torre and Albericio, 2022) (see Fig. 2 for more details). Among biological products, mAbs emerge as the leading

FAQs

Why is machine learning increasingly adopted in biopharmaceutical development and manufacturing?

Machine learning adoption is growing because large-scale production data are accumulating and because real-time monitoring of process variables and quality attributes is improving through advanced process analytical technologies.

What parts of the monoclonal antibody lifecycle does the review focus on?

The review covers upstream, downstream, and product formulation processes, including design, monitoring, control, and optimization using ML-based solutions.

What challenges does the paper identify regarding ML in monoclonal antibody processes?

Key challenges relate to the bioprocesses themselves, the nature of process data, and the practical use of machine learning models during process development and manufacturing.

Aug 4, 2026

Upload by Sage