Moving towards an era of hybrid modelling: advantages and challenges of coupling mechanistic and data-driven models for upstream pharmaceutical bioprocesses

In silico simulation of glycosylation and related pathways

Glycans participate in a vast number of recognition systems in diverse organisms in health and in disease. However, glycans cannot be sequenced because there is no sequencer technology that can fully characterize them. There is no "template" for replicating glycans as there are for amino acids and nucleic acids. Instead, glycans are synthesized by a complicated orchestration of multitudes of glycosyltransferases and glycosidases. Thus glycans can vary greatly in structure, but they are not genetically reproducible and are usually isolated in minute amounts. To characterize (sequence) the glycome (defined as the glycans in a particular organism, tissue, cell, or protein), glycosylation pathway prediction using in silico methods based on glycogene expression data, and glycosylation simulations have been attempted. Since many of the mammalian glycogenes have been identified and cloned, it has become possible to predict the glycan biosynthesis pathway in these systems. By then incorporating systems biology and bioprocessing technologies to these pathway models, given the right enzymatic parameters including enzyme and substrate concentrations and kinetic reaction parameters, it is possible to predict the potentially synthesized glycans in the pathway. This review presents information on the data resources that are currently available to enable in silico simulations of glycosylation and related pathways. Then some of the software tools that have been developed in the past to simulate and analyze glycosylation pathways will be described, followed by a summary and vision for the future developments and research directions in this area.

Building blocks needed for mechanistic modeling of bioprocesses: A critical review based on protein production by CHO cells

This paper reviews the key building blocks needed to develop a mechanistic model for use as an operational production tool. The Chinese Hamster Ovary (CHO) cell, one of the most widely used hosts for antibody production in the pharmaceutical industry, is considered as a case study. CHO cell metabolism is characterized by two main phases, exponential growth followed by a stationary phase with strong protein production. This process presents an appropriate degree of complexity to outline the modeling strategy. The paper is organized into four main steps: (1) CHO systems and data collection; (2) metabolic analysis; (3) formulation of the mathematical model; and finally, (4) numerical solution, calibration, and validation. The overall approach can build a predictive model of target variables. According to the literature, one of the main current modeling challenges lies in understanding and predicting the spontaneous metabolic shift. Possible candidates for the trigger of the metabolic shift include the concentration of lactate and carbon dioxide. In our opinion, ammonium, which is also an inhibiting product, should be further investigated. Finally, the expected progress in the emerging field of hybrid modeling, which combines the best of mechanistic modeling and machine learning, is presented as a fascinating breakthrough. Note that the modeling strategy discussed here is a general framework that can be applied to any bioprocess.

Integration Approaches to Model Bioreactor Hydrodynamics and Cellular Kinetics for Advancing Bioprocess Optimisation

Large-scale bioprocesses are increasing globally to cater to the larger market demands for biological products. As fermenter volumes increase, the efficiency of mixing decreases, and environmental gradients become more pronounced compared to smaller scales. Consequently, the cells experience gradients in process parameters, which in turn affects the efficiency and profitability of the process. Computational fluid dynamics (CFD) simulations are being widely embraced for their ability to simulate bioprocess performance, facilitate bioprocess upscaling, downsizing, and process optimisation. Recently, CFD approaches have been integrated with dynamic Cell reaction kinetic (CRK) modelling to generate valuable information about the cellular response to fluctuating hydrodynamic parameters inside large production processes. Such coupled approaches have the potential to facilitate informed decision-making in intelligent biomanufacturing, aligning with the principles of "Industry 4.0" concerning digitalisation and automation. In this review, we discuss the benefits of utilising integrated CFD-CRK models and the different approaches to integrating CFD-based bioreactor hydrodynamic models with cellular kinetic models. We also highlight the suitability of different coupling approaches for bioprocess modelling in the purview of associated computational loads.

Model-based control system design to manage process parameters in mammalian cell culture for biopharmaceutical manufacturing

To enhance the robustness and flexibility of biopharmaceutical manufacturing, a paradigm shift toward methods of continuous processing, such as perfusion, and fundamental technologies for high-throughput process development are being actively investigated. The continuous upstream process must establish an advanced control strategy to ensure a "State of Control" before operation. Specifically, feedforward and feedback control must address the complex fluctuations that occur during the culture process and maintain critical process parameters in appropriate states. However, control system design for industry-standard mammalian cell culture processes is still often performed in a laborious trial-and-error manner. This paper provides a novel control approach in which controller specifications to obtain desired control characteristics can be determined systematically by combining a culture model with control theory. In the proposed scheme, control conditions, such as PID parameters, can be specified mechanistically based on process understanding and control requirements without qualitative decision making or specific preliminary experiments. The effectiveness of the model-based control algorithm was verified by control simulations assuming perfusion Chinese hamster ovary culture. As a tool to assist in the development of control strategies, this study will reduce the high operational workload that is a serious problem in continuous culture and facilitate the digitalization of bioprocesses.

How Not to Make the Joint Extended Kalman Filter Fail with Unstructured Mechanistic Models

The unstructured mechanistic model (UMM) allows for modeling the macro-scale of a phenomenon without known mechanisms. This is extremely useful in biomanufacturing because using the UMM for the joint estimation of states and parameters with an extended Kalman filter (JEKF) can enable the real-time monitoring of bioprocesses with unknown mechanisms. However, the UMM commonly used in biomanufacturing contains ordinary differential equations (ODEs) with unshared parameters, weak variables, and weak terms. When such a UMM is coupled with an initial state error covariance matrix P(t=0) and a process error covariance matrix Q with uncorrelated elements, along with just one measured state variable, the joint extended Kalman filter (JEKF) fails to estimate the unshared parameters and state simultaneously. This is because the Kalman gain corresponding to the unshared parameter remains constant and equal to zero. In this work, we formally describe this failure case, present the proof of JEKF failure, and propose an approach called SANTO to side-step this failure case. The SANTO approach consists of adding a quantity to the state error covariance between the measured state variable and unshared parameter in the initial P(t = 0) of the matrix Ricatti differential equation to compute the predicted error covariance matrix of the state and prevent the Kalman gain from being zero. Our empirical evaluations using synthetic and real datasets reveal significant improvements: SANTO achieved a reduction in root-mean-square percentage error (RMSPE) of up to approximately 17% compared to the classical JEKF, indicating a substantial enhancement in estimation accuracy.

Current challenges and recent advances on the path towards continuous biomanufacturing

Continuous biopharmaceutical manufacturing is currently a field of intense research due to its potential to make the entire production process more optimal for the modern, ever-evolving biopharmaceutical market. Compared to traditional batch manufacturing, continuous bioprocessing is more efficient, adjustable, and sustainable and has reduced capital costs. However, despite its clear advantages, continuous bioprocessing is yet to be widely adopted in commercial manufacturing. This article provides an overview of the technological roadblocks for extensive adoptions and points out the recent advances that could help overcome them. In total, three key areas for improvement are identified: Quality by Design (QbD) implementation, integration of upstream and downstream technologies, and data and knowledge management. First, the challenges to QbD implementation are explored. Specifically, process control, process analytical technology (PAT), critical process parameter (CPP) identification, and mathematical models for bioprocess control and design are recognized as crucial for successful QbD realizations. Next, the difficulties of end-to-end process integration are examined, with a particular emphasis on downstream processing. Finally, the problem of data and knowledge management and its potential solutions are outlined where ontologies and data standards are pointed out as key drivers of progress.

Simulation and optimization of nutrient uptake and biomass formation using a multi-parameter Monod-type model of tobacco BY-2 cell suspension cultures in a stirred-tank bioreactor

Introduction

Tobacco (Nicotiana tabacum) cv Bright Yellow-2 (BY-2) cell suspension cultures enable the rapid production of complex protein-based biopharmaceuticals but currently achieve low volumetric productivity due to slow biomass formation. The biomass yield can be improved with tailored media, which can be designed either by laborious trial-and-error experiments or systematic, rational design using mechanistic models, linking nutrient consumption and biomass formation.

Methods

Here we developed an iterative experiment-modeling-optimization workflow to gradually refine such a model and its predictions, based on collected data concerning BY-2 cell macronutrient consumption (sucrose, ammonium, nitrate and phosphate) and biomass formation.

Results and discussion

The biomass formation was well predicted by an unstructured segregated mechanistic Monod-type model as long as the nutrient concentrations did not approach zero (we omitted phosphate, which was completely depleted). Multi-criteria optimization for sucrose and biomass formation indicated the best tradeoff (in a Paretian sense) between maximum biomass yield and minimum process time by reducing the initial sucrose concentration, whereas the inoculation biomass could be increased to maximize the biomass yield or minimize the process time, which we confirmed in calibration experiments. The model became inaccurate at biomass densities > 8 g L-1 dry mass when sucrose was almost depleted. We compensated for this limitation by including glucose and fructose as sucrose hydrolysis products in the model. The remaining offset between the simulation and experimental data might be resolved by including intracellular pools of sucrose, ammonium, nitrate and phosphate. Overall, we demonstrated that iterative models can be used to systematically optimize conditions for bioreactor-based processes.

Cell-culture process optimization via model-based predictions of metabolism and protein glycosylation

In order to meet the rising demand for biologics and become competitive on the developing biosimilar market, there is a need for process intensification of biomanufacturing processes. Process development of biologics has historically relied on extensive experimentation to develop and optimize biopharmaceutical manufacturing. Experimentation to optimize media formulations, feeding schedules, bioreactor operations and bioreactor scale up is expensive, labor intensive and time consuming. Mathematical modeling frameworks have the potential to enable process intensification while reducing the experimental burden. This review focuses on mathematical modeling of cellular metabolism and N-linked glycosylation as applied to upstream manufacturing of biologics. We review developments in the field of modeling cellular metabolism of mammalian cells using kinetic and stoichiometric modeling frameworks along with their applications to simulate, optimize and improve mechanistic understanding of the process. Interest in modeling N-linked glycosylation has led to the creation of various types of parametric and non-parametric models. Most published studies on mammalian cell metabolism have performed experiments in shake flasks where the pH and dissolved oxygen cannot be controlled. Efforts to understand and model the effect of bioreactor-specific parameters such as pH, dissolved oxygen, temperature, and bioreactor heterogeneity are critically reviewed. Most modeling efforts have focused on the Chinese Hamster Ovary (CHO) cells, which are most commonly used to produce monoclonal antibodies (mAbs). However, these modeling approaches can be generalized and applied to any mammalian cell-based manufacturing platform. Current and potential future applications of these models for Vero cell-based vaccine manufacturing, CAR-T cell therapies, and viral vector manufacturing are also discussed. We offer specific recommendations for improving the applicability of these models to industrially relevant processes.

Steps toward a digital twin for functional food production with increased health benefits

Lactobacillus rhamnosus (L. rhamnosus) is a commensal bacterium with health-promoting properties and with a wide range of applications within the food industry. To improve and optimize the control of L. rhamnosus biomass production in batch and fed-batch bioprocesses, this study proposes the application of artificial neural network (ANN) modelling to improve process control and monitoring, with potential future implementation as a basis for a digital twin. Three ANNs were developed using historical data from ten bioprocesses. These ANNs were designed to predict the biomass in batch bioprocesses with different media compositions, predict biomass in fed-batch bioprocesses, and predict the growth rate in fed-batch bioprocesses. The immunomodulatory effect of the L. rhamnosus samples was examined and found to elicit an anti-inflammatory response as evidenced by the inhibition of IL-6 and TNF-α secretion. Overall, the findings of this study reinforce the potential of ANN modelling for bioprocess optimization aimed at improved control for maximising the volumetric productivity of L. rhamnosus as an immunomodulatory agent with applications in the functional food industry.

From observational to actionable: rethinking omics in biologics production

As the era of omics continues to expand with increasing ubiquity and success in both academia and industry, omics-based experiments are becoming commonplace in industrial biotechnology, including efforts to develop novel solutions in bioprocess optimization and cell line development. Omic technologies provide particularly valuable 'observational' insights for discovery science, especially in academic research and industrial R&D; however, biomanufacturing requires a different paradigm to unlock 'actionable' insights from omics. Here, we argue the value of omic experiments in biotechnology can be maximized with deliberate selection of omic approaches and forethought about analysis techniques. We describe important considerations when designing and implementing omic-based experiments and discuss how systems biology analysis strategies can enhance efforts to obtain actionable insights in mammalian-based biologics production.

Hybrid modeling in bioprocess dynamics: Structural variabilities, implementation strategies, and practical challenges

Hybrid modeling, with an appropriate blend of the mechanistic and data-driven framework, is increasingly being adopted in bioprocess modeling, model-based experimental design (digital-twin), identification of critical process parameters, and optimization. However, the development of a hybrid model from experimental data is an inherently complex workflow, involving designed experiments, selection of the data-driven process, identification of model parameters, assessment fitness, and generalization capability. Depending on the complexity of the process system and purpose, each piece of these modules can flexibly be incorporated into the puzzle. However, this extra flexibility can be a cause of concern to trace an "optimal" model structure. In this paper, the development of hybrid models in a common bioprocess system, selection of data-driven components and their mapping to states, choice of parameter identification techniques, and model quality assurance are revisited. The challenges associated with hybrid-model development, and corrective actions have also been reviewed. The review also suggests the lack of data, and code sharing in communal repositories can be a hurdle in the exploration, and expansion of those tools in a bioprocess system.

Hybrid model development for parameter estimation and process optimization of hydrophobic interaction chromatography

Hydrophobic Interaction Chromatography (HIC) is often employed as a polishing step to remove aggregates for the purification of therapeutic proteins in the biopharmaceutical industry. To accelerate the process development and save the costs of performing time- and resource-intensive experiments, advanced model-based process design and optimization are necessary. Due to the unclear adsorption mechanism of the salt-dependent interaction between the protein and resin, the development of an accurate mechanistic model to describe the complex HIC behavior is challenging. In this work, an isotherm derived from Wang et al. is modified by adding three extra parameters together with an equilibrium dispersive model to represent the HIC process. To reduce the development effort of isotherm equations and extract missing information from the available data, a hybrid model is constructed by combining a simple and well-known multi-component Langmuir isotherm (MCL) with a neural network (NN). It is observed that the structure of the hybrid model is of critical importance to the accuracy of the developed model. During parameter estimation, a regularization strategy is incorporated to prevent overfitting. Furthermore, the impact of NN structures and regularization rates are comprehensively investigated. One of the interesting findings was that a simple NN with one hidden layer with two nodes and sigmoid as the activation function, significantly outperforms the mechanistic model, with a 62% improvement in accuracy in calibration and 31.4% in validation. To ensure the generalizability of the developed hybrid model, an in-silico dataset is generated using the mechanistic model to test the extrapolation capability of the hybrid model. Process optimization is also carried out to find the optimal operating conditions under product quality constraints using the developed hybrid model.

Big data and machine learning driven bioprocessing - Recent trends and critical analysis

Given the potential of machine learning algorithms in revolutionizing the bioengineering field, this paper examined and summarized the literature related to artificial intelligence (AI) in the bioprocessing field. Natural language processing (NLP) was employed to explore the direction of the research domain. All the papers from 2013 to 2022 with specific keywords of bioprocessing using AI were extracted from Scopus and grouped into two five-year periods of 2013-to-2017 and 2018-to-2022, where the past and recent research directions were compared. Based on this procedure, selected sample papers from recent five years were subjected to further review and analysis. The result shows that 50% of the publications in the past five-year focused on topics related to hybrid models, ANN, biopharmaceutical manufacturing, and biorefinery. The summarization and analysis of the outcome indicated that implementing AI could improve the design and process engineering strategies in bioprocessing fields.

rAAV Manufacturing: The Challenges of Soft Sensing during Upstream Processing

Recombinant adeno-associated virus (rAAV) is the most effective viral vector technology for directly translating the genomic revolution into medicinal therapies. However, the manufacturing of rAAV viral vectors remains challenging in the upstream processing with low rAAV yield in large-scale production and high cost, limiting the generalization of rAAV-based treatments. This situation can be improved by real-time monitoring of critical process parameters (CPP) that affect critical quality attributes (CQA). To achieve this aim, soft sensing combined with predictive modeling is an important strategy that can be used for optimizing the upstream process of rAAV production by monitoring critical process variables in real time. However, the development of soft sensors for rAAV production as a fast and low-cost monitoring approach is not an easy task. This review article describes four challenges and critically discusses the possible solutions that can enable the application of soft sensors for rAAV production monitoring. The challenges from a data scientist's perspective are (i) a predictor variable (soft-sensor inputs) set without AAV viral titer, (ii) multi-step forecasting, (iii) multiple process phases, and (iv) soft-sensor development composed of the mechanistic model.

NyctiDB: A non-relational bioprocesses modeling database supported by an ontology

Strategies to exploit and enable the digitalization of industrial processes are on course to become game-changers in optimizing (bio)chemical facilities. To achieve this, these industries face an increasing need for process models and, as importantly, an efficient way to store the models and data/information. Therefore, this work proposes developing an online information storage system that can facilitate the reuse and expansion of process models and make them available to the digitalization cycle. This system is named NyctiDB, and it is a novel non-relational database coupled with a bioprocess ontology. The ontology supports the selection and classification of bioprocess models focused information, while the database is in charge of the online storage of said information. Through a series of online collections, NyctiDB contains essential knowledge for the design, monitoring, control, and optimization of a bioprocess based on its mathematical model. Once NyctiDB has been implemented, its applicability and usefulness are demonstrated through two applications. Application A shows how NyctiDB is integrated inside the software architecture of an online educational bioprocess simulator. This implies that NyctiDB provides the information for the visualization of different bioprocess behaviours and the modifications of the models in the software. Moreover, the information related to the parameters and conditions of each model is used to support the users’ understanding of the process. Additionally, application B illustrates that NyctiDB can be used as AI enabler to further the research in this field through open-source and reliable data. This can, in fact, be used as the information source for the AI frameworks when developing, for example, hybrid models or smart expert systems for bioprocesses. Henceforth, this work aims to provide a blueprint on how to collect bioprocess modeling information and connect it to facilitate and empower the Internet-of-Things paradigm and the digitalization of the biomanufacturing industries.

Progress toward rapid, at-line N-glycosylation detection and control for recombinant protein expression

Proteins continue to represent a large fraction of the therapeutics market, reaching over a hundred billion dollars in market size globally. One key feature of protein modification that can affect both structure and function is the addition of glycosylation following protein folding, leading to regulatory requirements for the accurate assessment of protein attributes, including glycan structures. The non-template-driven, innately heterogeneous N-glycosylation process thus requires accurate detection to robustly generate protein therapies. A challenge exists in the timely detection of protein glycosylation without labor-intensive manipulation. In this article, we discuss progress toward N-glycoprotein control, focusing on novel control strategies and the advancement of rapid, high-throughput analysis methods.

Toward predictive engineering of gene circuits

Many synthetic biology applications rely on programming living cells using gene circuits - the assembly and wiring of genetic elements to control cellular behaviors. Extensive progress has been made in constructing gene circuits with diverse functions and applications. For many circuit functions, however, it remains challenging to ensure that the circuits operate in a predictable manner. Although the notion of predictability may appear intuitive, close inspection suggests that it is not always clear what constitutes predictability. We dissect this concept and how it can be confounded by the complexity of a circuit, the complexity of the context, and the interplay between the two. We discuss circuit engineering strategies, in both computation and experiment, that have been used to improve the predictability of gene circuits.

Model-based intensification of CHO cell cultures: One-step strategy from fed-batch to perfusion

There is a growing interest in continuous processing of the biopharmaceutical industry. However, the technology transfer from traditional batch-based processes is considered a challenge as protocol and tools still remain to be established for their usage at the manufacturing scale. Here, we present a model-based approach to design optimized perfusion cultures of Chinese Hamster Ovary cells using only the knowledge captured during small-scale fed-batch experiments. The novelty of the proposed model lies in the simplicity of its structure. Thanks to the introduction of a new catch-all variable representing a bulk of by-products secreted by the cells during their cultivation, the model was able to successfully predict cellular behavior under different operating modes without changes in its formalism. To our knowledge, this is the first experimentally validated model capable, with a single set of parameters, to capture culture dynamic under different operating modes and at different scales.

Nonlinear Reactor Design Optimization With Embedded Microkinetic Model Information

Despite the success of multiscale modeling in science and engineering, embedding molecular-level information into nonlinear reactor design and control optimization problems remains challenging. In this work, we propose a computationally tractable scale-bridging approach that incorporates information from multi-product microkinetic (MK) models with thousands of rates and chemical species into nonlinear reactor design optimization problems. We demonstrate reduced-order kinetic (ROK) modeling approaches for catalytic oligomerization in shale gas processing. We assemble a library of six candidate ROK models based on literature and MK model structure. We find that three metrics—quality of fit (e.g., mean squared logarithmic error), thermodynamic consistency (e.g., low conversion of exothermic reactions at high temperatures), and model identifiability—are all necessary to train and select ROK models. The ROK models that closely mimic the structure of the MK model offer the best compromise to emulate the product distribution. Using the four best ROK models, we optimize the temperature profiles in staged reactors to maximize conversions to heavier oligomerization products. The optimal temperature starts at 630–900K and monotonically decreases to approximately 560 K in the final stage, depending on the choice of ROK model. For all models, staging increases heavier olefin production by 2.5% and there is minimal benefit to more than four stages. The choice of ROK model, i.e., model-form uncertainty, results in a 22% difference in the objective function, which is twice the impact of parametric uncertainty; we demonstrate sequential eigendecomposition of the Fisher information matrix to identify and fix sloppy model parameters, which allows for more reliable estimation of the covariance of the identifiable calibrated model parameters. First-order uncertainty propagation determines this parametric uncertainty induces less than a 10% variability in the reactor optimization objective function. This result highlights the importance of quantifying model-form uncertainty, in addition to parametric uncertainty, in multi-scale reactor and process design and optimization. Moreover, the fast dynamic optimization solution times suggest the ROK strategy is suitable for incorporating molecular information in sequential modular or equation-oriented process simulation and optimization frameworks.

Modern Sensor Tools and Techniques for Monitoring, Controlling, and Improving Cell Culture Processes

The growing biopharmaceutical industry has reached a level of maturity that allows for the monitoring of numerous key variables for both process characterization and outcome predictions. Sensors were historically used in order to maintain an optimal environment within the reactor to optimize process performance. However, technological innovation has pushed towards on-line in situ continuous monitoring of quality attributes that could previously only be estimated off-line. These new sensing technologies when coupled with software models have shown promise for unique fingerprinting, smart process control, outcome improvement, and prediction. All this can be done without requiring invasive sampling or intervention on the system. In this paper, the state-of-the-art sensing technologies and their applications in the context of cell culture monitoring are reviewed with emphasis on the coming push towards industry 4.0 and smart manufacturing within the biopharmaceutical sector. Additionally, perspectives as to how this can be leveraged to improve both understanding and outcomes of cell culture processes are discussed.

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

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.

Process Analytical Technologies - Advances in bioprocess integration and future perspectives

Process Analytical Technology (PAT) instruments include analyzers capable of measuring physical and chemical process parameters and key attributes with the goal of optimizing process controls. PAT in the form of a probe or sensor is designed to integrate within the pharmaceutical manufacturing line and is coupled with computing equipment to perform chemometric modeling for result interpretation and multilayer statistical control of processes. PAT solutions are intended for understanding bioprocesses with a goal to control quality at all stages of product manufacturing and achieve quality by design (QbD). The goal of PAT implementation is to promote real-time release of products to decrease the cycle time and cost of production. This review focuses on the applications of PAT solutions at different stages of the manufacturing process for vaccine production, the advantages, challenges at present state, and the vision of the future development of biopharmaceutical industries.