Between the Poles of Data‐Driven and Mechanistic Modeling for Process Operation

Solle, Dörte; Hitzmann, Bernd; Herwig, Christoph; Remelhe, Manuel; Ulonska, Sophia; Wuerth, Lynn; Prata, Adrian; Steckenreiter, Thomas

doi:10.1002/cite.201600175

Cited by 86 publications

(61 citation statements)

References 67 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…However, with the option for high‐throughput experimentation more input factors can be taken into account, increasing the predictive power of DoE approaches. Noteworthy, optimization by DoE provides information about improved conditions, but lacks mechanistic understanding unless mechanistic instead of black‐box models are used . Perhaps the most valuable feature of DoE applications is the systematic investigation of non‐intuitive input factors, that are not considered to be important when relying on biased expert knowledge or educated guesses alone.…”

Section: Microbioreactor (Mbr) Systemsmentioning

confidence: 99%

Microbioreactor Systems for Accelerated Bioprocess Development

Hemmerich

Noack

Wiechert

et al. 2018

Biotechnology Journal

124

129

View full text Add to dashboard Cite

In recent years, microbioreactor (MBR) systems have evolved towards versatile bioprocess engineering tools. They provide a unique solution to combine higher experimental throughput with extensive bioprocess monitoring and control, which is indispensable to develop economically and ecologically competitive bioproduction processes. MBR systems are based either on down-scaled stirred tank reactors or on advanced shaken microtiter plate cultivation devices. Importantly, MBR systems make use of optical measurements for non-invasive, online monitoring of important process variables like biomass concentration, dissolved oxygen, pH, and fluorescence. The application range of MBR systems can be further increased by integration into liquid handling robots, enabling automatization and, thus standardization, of various handling and operation procedures. Finally, the tight integration of quantitative strain phenotyping with bioprocess development under industrially relevant conditions greatly increases the probability of finding the right combination of producer strain and bioprocess control strategy. This review will discuss the current state of the art in the field of MBR systems and we can readily conclude that their importance for industrial biotechnology will further increase in the near future.

show abstract

Section: Microbioreactor (Mbr) Systemsmentioning

confidence: 99%

Microbioreactor Systems for Accelerated Bioprocess Development

Hemmerich

Noack

Wiechert

et al. 2018

Biotechnology Journal

124

129

View full text Add to dashboard Cite

show abstract

“…Alternatively, atline analytical devices, such as HPLC systems, can give important process insight along the sequence of unit operations in the manufacturing platform (Karst et al, 2017). In cases where CQAs cannot be directly assessed, soft‐sensor‐based technologies can provide relevant substitutes (Kroll, Hofer, Ulonska, Kager, & Herwig, 2017; Mandenius & Gustavsson, 2015; Narayanan et al, 2019; Sokolov, Feidl, Morbidelli, & Butte, 2018; Solle et al, 2017; Sommeregger et al, 2017).…”

Section: Introductionmentioning

confidence: 99%

Process‐wide control and automation of an integrated continuous manufacturing platform for antibodies

Feidl

Vogg

Wolf

et al. 2020

Biotech & Bioengineering

View full text Add to dashboard Cite

Integrated continuous manufacturing is entering the biopharmaceutical industry. The main drivers range from improved economics, manufacturing flexibility, and more consistent product quality. However, studies on fully integrated production platforms have been limited due to the higher degree of system complexity, limited process information, disturbance, and drift sensitivity, as well as difficulties in digital process integration. In this study, we present an automated end‐to‐end integrated process consisting of a perfusion bioreactor, CaptureSMB, virus inactivation (VI), and two polishing steps to produce an antibody from an instable cell line. A supervisory control and data acquisition (SCADA) system was developed, which digitally integrates unit operations and analyzers, collects and centrally stores all process data, and allows process‐wide monitoring and control. The integrated system consisting of bioreactor and capture step was operated initially for 4 days, after which the full end‐to‐end integrated run with no interruption lasted for 10 days. In response to decreasing cell‐specific productivity, the supervisory control adjusted the loading duration of the capture step to obtain high capacity utilization without yield loss and constant antibody quantity for subsequent operations. Moreover, the SCADA system coordinated VI neutralization and discharge to enable constant loading conditions on the polishing unit. Lastly, the polishing was sufficiently robust to cope with significantly increased aggregate levels induced on purpose during virus inactivation. It is demonstrated that despite significant process disturbances and drifts, a robust process design and the supervisory control enabled constant (optimum) process performance and consistent product quality.

show abstract

“…The selection of the type of model depends on the goal of the modeling activity but is often highly biased by the expertise of the responsible team, available (commercial) software solutions and general resources such as time and labor. However, the essential qualities of models are the ability to describe the relevant reality with sufficient accuracy, the capability for model transfer to different situations (model robustness) and the simplicity of its development and interpretation (Solle et al, ).…”

Section: Introductionmentioning

confidence: 99%

“…The data‐driven part simplifies the management of system complexity and the estimation of model parameters and sensitivity. A detailed description of application, definition, advantages, and disadvantages of all the three modeling techniques is reported in a recent general review by (Solle et al, ) and from the perspective of QbD and PAT by (Simon et al, ; Teixeira, Oliveira, Alves, & Carrondo, ; Von Stosch et al, ).…”

Section: Introductionmentioning

confidence: 99%

“…There can be different possible architectures for developing hybrid models as described in detail in (Solle et al, ; Thompson & Kramer, ). These are referred to as serial when using the Black‐box models to estimate unknown terms in the mechanistic equations, parallel when using the Black‐box model to reduce the errors made by the mechanistic model or a hybrid approach when the mechanistic model is used to generate data for training Black‐box models.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A new generation of predictive models: The added value of hybrid models for manufacturing processes of therapeutic proteins

Narayanan

Sokolov

Morbidelli

et al. 2019

Biotech & Bioengineering

View full text Add to dashboard Cite

Due to the lack of complete understanding of metabolic networks and reaction pathways, establishing a universal mechanistic model for mammalian cell culture processes remains a challenge. Contrarily, data-driven approaches for modeling these processes lack extrapolation capabilities. Hybrid modeling is a technique that exploits the synergy between the two modeling methods. Although mammalian cell cultures are among the most relevant processes in biotechnology and indeed looks ideal for hybrid modeling, their application has only been proposed but never developed in the literature. This study provides a quantitative assessment of the improvement brought by hybrid models with respect to the state-of-the-art statistical predictive models in the context of therapeutic protein production. This is illustrated using a dataset obtained from a 3.5 L fed-batch experiment. With the goal to robustly define the process design space, hybrid models reveal a superior capability to predict the time evolution of different process variables using only the initial and process conditions in comparison to the statistical models. Hybrid models not only feature more accurate prediction results but also demonstrate better robustness and extrapolation capabilities. For the future application, this study highlights the added value of hybrid modeling for model-based process optimization and design of experiments. K E Y W O R D Sbiopharmaceuticals, hybrid models, mammalian culture, predictive modeling, process optimization

show abstract

Between the Poles of Data‐Driven and Mechanistic Modeling for Process Operation

Cited by 86 publications

References 67 publications

Microbioreactor Systems for Accelerated Bioprocess Development

Microbioreactor Systems for Accelerated Bioprocess Development

Process‐wide control and automation of an integrated continuous manufacturing platform for antibodies

A new generation of predictive models: The added value of hybrid models for manufacturing processes of therapeutic proteins

Contact Info

Product

Resources

About