Harnessing the potential of machine learning for advancing “Quality by Design” in biomanufacturing

Abstract

Ensuring consistent high yields and product quality are key challenges in biomanufacturing. Even minor deviations in critical process parameters (CPPs) such as media and feed compositions can significantly affect product critical quality attributes (CQAs). To identify CPPs and their interdependencies with product yield and CQAs, design of experiments, and multivariate statistical approaches are typically used in industry. Although these models can predict the effect of CPPs on product yield, there is room to improve CQA prediction performance by capturing the complex relationships in high-dimensional data. In this regard, machine learning (ML) approaches offer immense potential in handling non-linear datasets and thus are able to identify new CPPs that could effectively predict the CQAs. ML techniques can also be synergized with mechanistic models as a ‘hybrid ML’ or ‘white box ML’ to identify how CPPs affect the product yield and quality mechanistically, thus enabling rational design and control of the bioprocess. In this review, we describe the role of statistical modeling in Quality by Design (QbD) for biomanufacturing, and provide a generic outline on how relevant ML can be used to meaningfully analyze bioprocessing datasets. We then offer our perspectives on how relevant use of ML can accelerate the implementation of systematic QbD within the biopharma 4.0 paradigm.

Exploring the potential of machine learning for more efficient development and production of biopharmaceuticals

“…The typical goal of using ML is to develop predictive models based on statistical associations among the features in a given dataset. These models can further predict a range of outputs, in terms of binary responses, continuous values or categorical labels 2 …”

“…Such integration will enable real‐time association of external factors, like market demand, supplier inventories, patient experience, and public emergencies, with the internal information, namely laboratory data, modeling and simulation outcomes, energy and resource management, and so forth. This connection between the internal and external data sources will enable unprecedented real‐time monitoring, control, prediction and responsiveness in biopharmaceutical manufacturing and development 1,2 …”

“…Machine learning is still in its infancy for directly supporting QbD based development and manufacturing of biopharmaceuticals 2 . However, the adoption of ML‐based models in place of conventional multi‐variate‐data‐analysis (MVDA) is increasing with the accumulation of large‐scale data due to real‐time monitoring of process variables and quality attributes of products, through the implementation of process analytical technology in biopharmaceuticals manufacturing 2,9 .…”

A methodology for co-simulation-based optimization of biofabrication protocols

“…Several computational methodologies for biofabrication process design exist in the state-of-the-art. Design-of-Experiments (DoE) [23, 26] supports strategic and effective research design by enabling efficient, systematic exploration and exploitation of complex design spaces [7, 14]. A variety of DoE approaches exist [11], and they prove adequate to tackle multi-factorial problems in the optimization of directed cell differentiation [3, 18], and tissue engineering scaffolds [25].…”

“…A variety of DoE approaches exist [11], and they prove adequate to tackle multi-factorial problems in the optimization of directed cell differentiation [3, 18], and tissue engineering scaffolds [25]. DoE can be combined with Machine Learning (ML) and Artificial Neural Networks (ANN) to improve the accuracy of the bioprocess model [23].…”