Search

Spiders utilize their fine silk fibers for their aerial dispersal, known as ballooning. With this method, spiders can disperse hundreds of kilometers, reaching as high as 4.5 km. However, the passive dynamics of a ballooning model (a highly flexible filament and a spider body at the end of it) are not well understood. Here, we introduce a bead–spring model that takes into account the anisotropic drag of a fiber to investigate the passive dynamics by the various non-uniform flows: (i) a shear flow, (ii) a periodic vortex flow field, and (iii) a homogeneous turbulent flow. For the analysis of the wide range of parameters, we defined a dimensionless parameter, which is called "a ballooning number." The ballooning number is defined as the ratio of Stokes' fluid-dynamic force on a fiber by the non-uniform flow field to the gravitational force of a body. Our simulations show that the present model in a homogeneous turbulent flow exhibits the biased characteristic of slow settling with increasing turbulence. Upon investigating this phenomenon for a shear flows, it was found that the drag anisotropy of the filament structure is the main cause of the slow settling. Particularly, the cause of slow settling speed lies not only in the deformed geometrical shape but also in its generation of fluid-dynamic force in a non-uniform flow. Additionally, we found that the ballooning structure could become trapped in a vortex flow. These results help deepen our understanding of the passive dynamics of spiders ballooning in the atmospheric boundary layer. [ABSTRACT FROM AUTHOR]

Bioprocess development, optimization, and control in mini-bioreactor systems require information about essential process parameters, high data densities, and the ability to dynamically change process conditions. We present an integration approach combining a parallel mini-bioreactor system integrated into a liquid handling station (LHS) with a second LHS for offline analytics. Non-invasive sensors measure pH and DO online. Offline samples are collected every 20 min and acetate, glucose, and OD620 subsequently analyzed offline. All data are automatically collected, analyzed, formalized, and used for process control and optimization. Fed-batch conditions are realized via a slow enzymatic glucose release system. The integration approach was successfully used to apply an online experimental re-design method to eight Escherichia coli fed-batch cultivations. The method utilizes generated data to select the following experimental actions online in order to reach the optimization goal of estimating E. coli fed-batch model parameters with as high accuracy as possible. Optimal experimental designs were re-calculated online based on the experimental data and implemented by introducing pulses via the LHS to the running fermentations. The LHS control allows for various implementations of advanced control and optimization strategies in milliliter scale. [ABSTRACT FROM AUTHOR]

This paper deals with the development, analysis and reduction of a two-stage anaerobic digestion process model. A two-stage model based on a two-step mass-balance model is first developed. The model incorporates two stages as two similar CSTRs, with acidogenesis in the first stage and mainly methanogenesis in the second. Steady state optimization validates the benefits of two-stage operation compared to the traditional one-stage process. Using reaction invariants, the dynamics of the fast modes is neglected by projecting the overall system dynamics on the slow-motion invariant manifold, and the original 7-state model is reduced to a 5th order system. By neglecting the nearly unobservable modes, the model is further reduced to 3rd order. Simulation results show excellent accuracy of the reduced model compared to the detailed one. At the same time, they indicate the necessity of process control for stabilization of the system. [ABSTRACT FROM AUTHOR]

• Typically, hybrid models use conventional machine learning approaches together with mechanistic modeling. • The so obtained hybrid models are not completely interpretable by domain experts. • The Functional-Hybrid model instead use library of domain-specific knowledge blocks. • Functional-Hybrid models are interpretable. • Compared to conventional hybrid model, they require less data and are superior in extrapolation. Mathematical models used for the representation of (bio)-chemical processes can be grouped into two broad paradigms: white-box or mechanistic models, completely based on knowledgeor black-box data-driven models based on patterns observed in data. However, in the past two-decade, hybrid modeling that explores the synergy between the two paradigms has emerged as a pragmatic compromise. The data-driven part of these has been largely based on conventional machine learning algorithms (e.g., artificial neural network, support vector regression), which prevents interpretability of the finally learnt model by the domain experts. In this work, we present a novel hybrid modeling framework, the Functional-Hybrid model, that uses the ranked domain-specific functional beliefs together with symbolic regression to develop dynamic models. We demonstrate the successful implementation of the Functional-Hybrid model and its interpretability, focusing on applying chemical reaction kinetic principles to classical chemical reactions, biochemistry, ecology, physiology, and a bioreactor. Furthermore, we demonstrate that during interpolation, the Functional-Hybrid model performs similarly to a Hybrid-ANN hybrid model implementing a conventional ANN. However, it provides the advantage of being –to some extent– interpretable, unlike the conventional Hybrid-ANN model. Additionally, it is shown that the Functional-Hybrid model outperforms the Hybrid-ANN model for a very low number of experiments, making it more suitable when data is scarce. Finally, the Functional-Hybrid models show superior extrapolation capabilities compared to the Hybrid-ANN model. This improved performance can be attributed to the structure imposed by the functional transformations introduced in the Functional-Hybrid model. [ABSTRACT FROM AUTHOR]

Automated high-throughput liquid handling operations in biolabs necessitate miniaturised and automatised equipment for effective space utilisation and system integration. This paper presents a thermal segment microwell plate control unit designed for enhanced microwell-based experimentation in liquid handling setups. The development of this device stems from the need to move towards geometry standardization and system integration of automated lab equipment. It incorporates features based on Smart Sensor and Sensor 4.0 concepts. An enzymatic activity assay is implemented with the developed device on a liquid handling station, allowing fast characterisation via a high-throughput approach. The device outperforms other comparable devices in certain metrics based on automated liquid handling requirements and addresses the needs of future biolabs in automation, especially in highthroughput screening. [ABSTRACT FROM AUTHOR]

The bacterium Escherichia coli is a widely used organism in biotechnology. For high space-time yields, glucose-limited fed-batch technology is the industry standard; this is because an overflow metabolism of acetate occurs at high glucose concentrations. As an interesting alternative, various strains with limited glucose uptake have been developed. However, these have not yet been characterized under process conditions. To demonstrate the efficiency of our previously developed high-throughput robotic platform, in the present work, we characterized three different exemplary E. coli knockout (KO) strains with limited glucose uptake capacities at three different scales (microtiter plates, 10 mL bioreactor system and 100 mL bioreactor system) under excess glucose conditions with different initial glucose concentrations. The extensive measurements of growth behavior, substrate consumption, respiration, and overflow metabolism were then used to determine the appropriate growth parameters using a mechanistic mathematical model, which allowed for a comprehensive comparative analysis of the strains. The analysis was performed coherently with these different reactor configurations and the results could be successfully transferred from one platform to another. Single and double KO mutants showed reduced specific rates for substrate uptake qSmax and acetate production qApmax; meanwhile, higher glucose concentrations had adverse effects on the biomass yield coefficient YXSem. Additional parameters compared to previous studies for the oxygen uptake rate and carbon dioxide production rate indicated differences in the specific oxygen uptake rate qOmax. This study is an example of how automated robotic equipment, together with mathematical model-based approaches, can be successfully used to characterize strains and obtain comprehensive information more quickly, with a trade-off between throughput and analytical capacity. [ABSTRACT FROM AUTHOR]

Nucleoside analogues are important compounds for the treatment of viral infections or cancers. While (chemo-)enzymatic synthesis is a valuable alternative to traditional chemical methods, the feasibility of such processes is lowered by the high production cost of the biocatalyst. As continuous enzyme membrane reactors (EMR) allow the use of biocatalysts until their full inactivation, they offer a valuable alternative to batch enzymatic reactions with freely dissolved enzymes. In EMRs, the enzymes are retained in the reactor by a suitable membrane. Immobilization on carrier materials, and the associated losses in enzyme activity, can thus be avoided. Therefore, we validated the applicability of EMRs for the synthesis of natural and dihalogenated nucleosides, using one-pot transglycosylation reactions. Over a period of 55 days, 2′-deoxyadenosine was produced continuously, with a product yield >90%. The dihalogenated nucleoside analogues 2,6-dichloropurine-2′-deoxyribonucleoside and 6-chloro-2-fluoro-2′-deoxyribonucleoside were also produced, with high conversion, but for shorter operation times, of 14 and 5.5 days, respectively. The EMR performed with specific productivities comparable to batch reactions. However, in the EMR, 220, 40, and 9 times more product per enzymatic unit was produced, for 2′-deoxyadenosine, 2,6-dichloropurine-2′-deoxyribonucleoside, and 6-chloro-2-fluoro-2′-deoxyribonucleoside, respectively. The application of the EMR using freely dissolved enzymes, facilitates a continuous process with integrated biocatalyst separation, which reduces the overall cost of the biocatalyst and enhances the downstream processing of nucleoside production. [ABSTRACT FROM AUTHOR]

Saccharomyces cerevisiae is a popular expression system for recombinant proteins. In most cases, production processes are performed as carbon-limited fed-batch cultures to avoid aerobic ethanol formation. Especially for constitutive expression systems, the specific product formation rate depends on the specific growth rate. The development of optimal feeding strategies strongly depends on laboratory-scale cultivations, which are time and resource consuming, especially when continuous experiments are carried out. It is therefore beneficial for accelerated process development to look at alternatives. In this study, S. cerevisiae AH22 secreting a heterologous endo-polygalacturonase (EPG) was characterized in microwell plates with an enzyme-based fed-batch medium. Through variation of the glucose release rate, different growth profiles were established and the impact on EPG secretion was analyzed. Product formation rates of 200-400 U (gx h)−1 were determined. As a reference, bioreactor experiments using the change-stat cultivation technique were performed. The growth-dependent product formation was analyzed over dilution rates of D = 0.01-0.35 with smooth change of D at a rate of 0.003 h−2. EPG production was found to be comparable with a qp of 400 U (gx h)−1 at D = 0.27 h−1. The presented results indicate that parallel miniaturized fed-batch cultures can be applied to determine product formation profiles of putative production strains. With further automation and parallelization of the concept, strain characterization can be performed in shorter time. [ABSTRACT FROM AUTHOR]

• A method for building models based on an S-system formalism is proposed. • ODE integration is avoided for model training by a two-stage incremental approach. • Models tailorable in complexity by the maximum number of parameters. • Trained closed-form expressions easily interpretable. Mathematical modeling helps guide experiments more effectively, support process monitoring and control tasks, stabilize product quality, increase consumer safety, or ease specific decision-making tasks for subject matter experts. However, constructing accurate process models can be challenging, especially with bioprocesses, due to complex metabolic mechanisms and data scarcity. This work proposes a method for building models combining a mass balance backbone with a canonical kinetic representation, i.e., the S-system formalism. The model structure and parameters that best describe the studied system are automatically identified by solving a mixed-integer nonlinear programming (MINLP) problem. Following an incremental approach, the integration of ordinary differential equations is avoided. Numerical examples show that our method performs similarly to models based on artificial neural networks, outperforming them in some cases while providing an analytical, closed-form model. Such expressions can be more easily interpreted and optimized in existing algebraic modeling systems. [ABSTRACT FROM AUTHOR]

Immobilization of enzymes has been considered as an efficient approach to facilitate enzyme recovery and to improve biocatalyst stability. However, multimeric nucleoside phosphorylases, useful for the synthesis of modified nucleosides, encounter several challenges to their immobilization, including requirement for high enzymatic load, and poor retention of enzyme activity. In this study, multimeric enzymes pyrimidine nucleoside phosphorylase (PyNP) and purine nucleoside phosphorylase (PNP), from Thermus thermophilus and Geobacillus thermoglucosidasius , respectively, were successfully immobilized on the magnetic beads with cross-linked polyethyleneimine and epoxide functional groups, resulting in high enzyme loading (up to 0.4 and 1.3 g per g dry beads), high enzyme activity maintenance (41% and 83% under the highest enzyme loading), and improved enzyme stability. The screening of the immobilization conditions showed that binding buffer pH, enzyme loading amount, binding temperature and binding time are important factors for the immobilization yield. The application of the immobilized enzymes in the synthesis of 2,6-dihalogenated purine nucleosides achieved with high substrate conversion (78.5–85.5%) as well as high productivity (1.5–2.0 g L −1 h −1 ). [ABSTRACT FROM AUTHOR]

We discuss the application of a nonlinear model predictive control (MPC) and moving horizon estimation (MHE) framework to achieve optimal operation of E. coli fed-batch cultivations with intermittent bolus feeding. 24 parallel experiments were considered in a high-throughput mini-bioreactor platform at a 10 mL scale. The robotic facility can run up to 48 fed-batch processes in parallel with automated liquid handling, online and at-line analytics. Three main challenges emerge in implementing the model-based monitoring and control framework: First, the inputs are given in an instantaneous pulsed form by bolus injections; second, online and at-line measurement frequencies are severely imbalanced; and third, optimization for the distinctive multiple reactors can be either parallelized or integrated. We address these challenges by incorporating the concept of impulsive control systems, formulating multi-rate MHE with identifiability analysis, and suggesting criteria for deciding the reactor configuration. In this study, we present the key elements and background theory of the implementation with in silico simulations for bacterial fed-batch cultivations. • An MPC and MHE framework is developed for E. coli cultivations in 24 mini-bioreactors. • Bolus feeding for mini-scale bioreactors is modeled as an impulsive control system. • A complex macro-kinetic growth model is fitted with multi-rate measurements. • Three criteria are proposed to classify 24 parallel mini-bioreactors into optimization groups. [ABSTRACT FROM AUTHOR]

Pyrimidine and purine nucleoside phosphorylases catalyze the reversible phosphorolytic cleavage and formation of the glycosidic bond of purine and pyrimidine nucleosides, respectively, and are thus, key catalysts for the synthesis of new compounds. The selection of the best combination of enzyme and reaction conditions is not trivial, as each of the two enzymes can perform the reaction in two directions and thus, also competes with the other one for the reaction intermediates. A generic approach to the solution of this problem based on the formulation of a mixed integer dynamic optimization program using MOSAICmodeling is presented. [ABSTRACT FROM AUTHOR]

Machine learning (ML) is becoming increasingly crucial in many fields of engineering but has not yet played out its full potential in bioprocess engineering. While experimentation has been accelerated by increasing levels of lab automation, experimental planning and data modeling are still largerly depend on human intervention. ML can be seen as a set of tools that contribute to the automation of the whole experimental cycle, including model building and practical planning, thus allowing human experts to focus on the more demanding and overarching cognitive tasks. First, probabilistic programming is used for the autonomous building of predictive models. Second, machine learning automatically assesses alternative decisions by planning experiments to test hypotheses and conducting investigations to gather informative data that focus on model selection based on the uncertainty of model predictions. This review provides a comprehensive overview of ML-based automation in bioprocess development. On the one hand, the biotech and bioengineering community should be aware of the potential and, most importantly, the limitation of existing ML solutions for their application in biotechnology and biopharma. On the other hand, it is essential to identify the missing links to enable the easy implementation of ML and Artificial Intelligence (AI) tools in valuable solutions for the bio-community. There is no one-fits-all procedure; however, this review should help identify the potential for automating model building by combining first-principles biotechnology knowledge and ML methods to address the reproducibility crisis in bioprocess development. • A Survey from the Perspective of Automated Bioprocess Development. • Comprehensive overview of Machine Learning -based automation in bioprocess development. • Recent ML implementation across several important subfields of bioprocess systems. • Two crucial challenges remaining the bottleneck of bioprocess automation. • Active Learning, Automation, Reinforcement Learning, Transfer Learning. [ABSTRACT FROM AUTHOR]

• In aerobic processes, the dynamics of the dissolved oxygen is usually very fast. • Both continuous and fed-batch models can be reduced by approximating fast dynamics. • The invariant manifold method leads to an accurate reduced model. • The reduced model is useful for simplifying the observer design problem. Owing to the increasing demand for large scale and high efficiency in manufacturing processes, computer aided tools for process operation and control are rapidly gaining popularity. An important state variable in aerobic processes is the dissolved oxygen, which can be easily measured online and is an important indicator of the metabolic activity. However, due to the fast kinetics of the oxygen transfer, dynamical models describing aerobic bioprocesses tend to be highly stiff. This can lead to significant numerical problems hampering its use for fixed step discretization methods and computationally costly applications such as computer fluid dynamics. In this work we use the slow-motion invariant manifold and the quasi steady state assumption methods to eliminate the differential equation describing the dissolved oxygen (the fast mode). By doing this, the tractability of the model is significantly increased with a neglectable loss in description power. The reduced model is also useful for simplifying the observer design problems, which is demonstrated by a state and parameter estimation example at the end of the work. [ABSTRACT FROM AUTHOR]

• A framework for uncertainty analysis of biokinetic models is presented. • Parameter identifiability is deduced from the dynamic sensitivity matrix. • A mechanistic model of Escherichia coli was analyzed for parameter identifiability and uncertainty of model outputs. • Model prediction accuracy is improved up to 15 times, after regularization. Mechanistic models are simplifications of bio-physical systems, for which the true values of the model parameters are sometimes unknown. Therefore, before using model-based predictions to study or improve a process, it is essential to ensure that the outputs of the model are reliable. This paper covers the development and application of a framework for practical identifiability and uncertainty analyses of dynamic growth models for bioprocesses. By exploring the numerical properties of the sensitivity matrix, a simple algorithm to determine the presence of non-identifiable parameters in models with high output uncertainty is presented. The framework detects the existence of non-identifiable parameters within the model and proposes a regularisation technique, in conjunction with Monte Carlo Analysis. As an example, the framework was used to analyse a macro-kinetic growth model of Escherichia coli describing a fed-batch process. The results show a reduction in the uncertainty of model outputs from a maximum coefficient of variation of 748% to 5% after regularization, and a 15-fold improvement in the accuracy of model predictions for two independent validation datasets. The presented framework aims to improve the reliability of model predictions and promote a more thorough handling of dynamical models to extend their use in biotechnology. [ABSTRACT FROM AUTHOR]