Spectral analysis of the effect of inflow noise on a fed-batch fermentation for recombinant β-galactosidase

The fermentative production of β-galactosidase from Escherichia coli CSH50 containing the plasmid pOU140 is mechanistically complex and difficult to model in a nonideal bioreactor. A spectral analysis has been done for a fed-batch fermentation with Gaussian disturbances in the feed stream. Inflow noise converts a smooth operation into aperiodic motion, as observed in some chemical reactions also. The disturbances also cause significant differences in the frequency responses of intra-cellular (plasmid DNA and β-galactosidase) and extra-cellular (concentration and mass fraction of recombinant cells) variables, whose implications are discussed.     

Spectral analysis of a nonideal bioreactor for a recombinant fermentation with run-away plasmids

An earlier application of spectral analysis to an isothermal, perfectly mixed, fed-batch fermentation [11] showed that it provides insight into the performance and reveals significant differences between key extra-cellular and intra-cellular (recombinant) variables. To enable application on a practically useful scale, the methodology has been extended to a bioreactor with temperature variation, incomplete mixing of the broth and Gaussian disturbances. For the same system, #-galactosidase produced by a recombinant E. coli strain, there are again differences between the rDNA and its protein, on the one hand, and the concentration and mass fraction of recombinant cells, on the other. For certain combinations of the dilution rate and the degrees of mixing, the bioreactor is only marginally stable, implying that a sufficiently large disturbance can trigger unstable behavior.     

Hybrid neural network modeling of a full-scale industrial wastewater treatment process

In recent years, hybrid neural network approaches, which combine mechanistic and neural network models, have received considerable attention. These approaches are potentially very efficient for obtaining more accurate predictions of process dynamics by combining mechanistic and neural network models in such a way that the neural network model properly accounts for unknown and nonlinear parts of the mechanistic model. In this work, a full-scale coke-plant wastewater treatment process was chosen as a model system. Initially, a process data analysis was performed on the actual operational data by using principal component analysis. Next, a simplified mechanistic model and a neural network model were developed based on the specific process knowledge and the operational data of the coke-plant wastewater treatment process, respectively. Finally, the neural network was incorporated into the mechanistic model in both parallel and serial configurations. Simulation results showed that the parallel hybrid modeling approach achieved much more accurate predictions with good extrapolation properties as compared with the other modeling approaches even in the case of process upset caused by, for example, shock loading of toxic compounds. These results indicate that the parallel hybrid neural modeling approach is a useful tool for accurate and cost-effective modeling of biochemical processes, in the absence of other reasonably accurate process models.     

Prediction model of DnBP degradation based on BP neural network in AAO system

A laboratory-scale anaerobic-anoxic-oxic (AAO) system was established to investigate the fate of DnBP. A removal kinetic model including sorption and biodegradation was formulated, and kinetic parameters were evaluated with batch experiments under anaerobic, anoxic, oxic conditions. However, it is highly complex and is difficult to confirm the kinetic parameters using conventional mathematical modeling. To correlate the experimental data with available models or some modified empirical equations, an artificial neural network model based on multilayered partial recurrent back propagation (BP) algorithm was applied for the biodegradation of DnBP from the water quality characteristic parameters. Compared to the kinetic model, the performance of the network for modeling DnBP is found to be more impressive. The results showed that the biggest relative error of BP network prediction model was 9.95%, while the kinetic model was 14.52%, which illustrates BP model predicting effluent DnBP more accurately than kinetic model forecasting.     

Coupling of a neural filter and a neural controller for improvement of fermentation performance

The noise associated with fermentation processes is normally minimised by a filtering technique. However, sometimes the noise may be beneficial if it is properly regulated. For recombinant -galactosidase production in a fed-batch fermentation subject to Gaussian disturbances, it is shown that a neural network trained to act as a noise filter can allow disturbances of only a particular variance which maximizes -galactosidase synthesis. By coupling such a filter with a neural controller, the productivity may be enhanced beyond what is possible with static filtering and either proportional-integral-derivative or neural control.     

Parallel hybrid modeling methods for a full-scale cokes wastewater treatment plant

Parallel hybrid modeling methods are applied to a full-scale cokes wastewater treatment plant. Within the hybrid model structure, a mechanistic model specifies the basic dynamics of the relevant process and a non-parametric model compensates for the inaccuracy of the mechanistic model. First, a simplified mechanistic model is developed based on Activated Sludge Model No. 1 and the specific process knowledge of the cokes wastewater treatment process. Then, the mechanistic model is combined with five different non-parametric models--feedforward back-propagation neural network, radial basis function network, linear partial least squares (PLS), quadratic PLS and neural network PLS (NNPLS)--in parallel configuration. These models are identified with the same data obtained from the plant operation to predict dynamic behavior of the process. The performance of each parallel hybrid model is compared based on their ease of model building, prediction accuracy and interpretability. For this application, the parallel hybrid model with NNPLS as non-parametric model gives better performance than other parallel hybrid models. In addition, the NNPLS model is used to analyze the behavior of the operation data in the reduced space and allows for fault detection and isolation.     

Neural and Hybrid Optimizations of the Fed-Batch Synthesis of Poly-β-Hydroxybutyrate by Ralstonia eutropha in a Nonideal Bioreactor

Poly-β-hydroxybutyrate (PHB) is synthesized by some microorganisms under stressful conditions. Despite its properties being comparable to those of synthetic polymers, and its biocompatibility and biodegradability, low productivities have dampened commercial interest in microbial PHB production. To increase production efficiency, a fed-batch fermentation with Ralstonia eutropha was optimized recently through a neural-cum-dispersion model (D-model) incorporating incomplete dispersion and noise in the feed streams. The approach described in the work has been improved in two ways: first by a model comprising neural networks only (N-model) and then by a hybrid neural model (H-model) with a mathematical component. At optimum dispersion, PHB production through the N-model optimization was 35% more than by the D-model, and this was enhanced by a further 58% using hybrid optimization. Recognizing that the D-model itself more than doubled the PHB production compared to a noise-free fully dispersed bioreactor, the present results establish hybrid neural optimization as a viable method for PHB production improvement under realistic conditions.     

A hybrid representation approach for modelling complex dynamic bioprocesses

This paper considers the use of hybrid models to represent the dynamic behaviour of biotechnological processes. Each hybrid model consists of a set of non linear differential equations and a neural model. The set of differential equations attempts to describe as much as possible the phenomenology of the process whereas neural networks model predict some key parameters that are an essential part of the phenomenological model. The neural model is obtained indirectly, that is, using the prediction errors of one or more state variables to adjust its weights instead of successive presentations of input-output data of the neural network. This approach allows to use actual measurements to derive a suitable neural model that not only represents the variation of some key parameters but it is also able to partly include dynamic behaviour unaccounted for by the phenomenological model. The approach is described in detail using three test cases: (1) the fermentation of glucose to gluconic acid by the micro-organism Pseudomonas ovalis, (2) the growth of filamentous fungi in a solid state fermenter, and (3) the propagation of filamentous fungi growing on a 2-D solid substrate. Results for the three applications clearly demon- strate that using a hybrid model is a viable alternative for modelling complex biotechnological bioprocesses.     

A Four-Stage Hybrid Model for Hydrological Time Series Forecasting

Hydrological time series forecasting remains a difficult task due to its complicated nonlinear, non-stationary and multi-scale characteristics. To solve this difficulty and improve the prediction accuracy, a novel four-stage hybrid model is proposed for hydrological time series forecasting based on the principle of ‘denoising, decomposition and ensemble’. The proposed model has four stages, i.e., denoising, decomposition, components prediction and ensemble. In the denoising stage, the empirical mode decomposition (EMD) method is utilized to reduce the noises in the hydrological time series. Then, an improved method of EMD, the ensemble empirical mode decomposition (EEMD), is applied to decompose the denoised series into a number of intrinsic mode function (IMF) components and one residual component. Next, the radial basis function neural network (RBFNN) is adopted to predict the trend of all of the components obtained in the decomposition stage. In the final ensemble prediction stage, the forecasting results of all of the IMF and residual components obtained in the third stage are combined to generate the final prediction results, using a linear neural network (LNN) model. For illustration and verification, six hydrological cases with different characteristics are used to test the effectiveness of the proposed model. The proposed hybrid model performs better than conventional single models, the hybrid models without denoising or decomposition and the hybrid models based on other methods, such as the wavelet analysis (WA)-based hybrid models. In addition, the denoising and decomposition strategies decrease the complexity of the series and reduce the difficulties of the forecasting. With its effective denoising and accurate decomposition ability, high prediction precision and wide applicability, the new model is very promising for complex time series forecasting. This new forecast model is an extension of nonlinear prediction models.     

Designing hybrid life cycle assessment models based on uncertainty and complexity

Purpose

Despite the wide use of LCA for environmental profiling, the approach for determining the system boundary within LCA models continues to be subjective and lacking in mathematical rigor. As a result, life cycle models are often developed in an ad hoc manner, and are difficult to compare. Significant environmental impacts may be inadvertently left out. Overcoming this shortcoming can help elicit greater confidence in life cycle models and their use for decision making.

Methods

This paper describes a framework for hybrid life cycle model generation by selecting activities based on their importance, parametric uncertainty, and contribution to network complexity. The importance of activities is determined by structural path analysis—which then guides the construction of life cycle models based on uncertainty and complexity indicators. Information about uncertainty is from the available life cycle inventory; complexity is quantified by cost or granularity. The life cycle model is developed in a hierarchical manner by adding the most important activities until error requirements are satisfied or network complexity exceeds user-specified constraints.

Results and Discussion

The framework is applied to an illustrative example for building a hybrid LCA model. Since this is a constructed example, the results can be compared with the actual impact, to validate the approach. This application demonstrates how the algorithm sequentially develops a life cycle model of acceptable uncertainty and network complexity. Challenges in applying this framework to practical problems are discussed.

Conclusion

The presented algorithm designs system boundaries between scales of hybrid LCA models, includes or omits activities from the system based on path analysis of environmental impact contribution at upstream network nodes, and provides model quality indicators that permit comparison between different LCA models.

     

Using historical data for bioprocess optimization: modeling wine characteristics using artificial neural networks and archived process information

Optimization of fermentation processes is a difficult task that relies on an understanding of the complex effects of processing inputs on productivity and quality outputs. Because of the complexity of these biological systems, traditional optimization methods utilizing mathematical models and statistically designed experiments are less effective, especially on a production scale. At the same time, information is being collected on a regular basis during the course of normal manufacturing and process development that is rarely fully utilized. We are developing an optimization method in which historical process data is used to train an artificial neural network for correlation of processing inputs and outputs. Subsequently, an optimization routine is used in conjunction with the trained neural network to find optimal processing conditions given the desired product characteristics and any constraints on inputs. Wine processing is being used as a case study for this work. Using data from wine produced in our pilot winery over the past 3 years, we have demonstrated that trained neural networks can be used successfully to predict the yeast-fermentation kinetics, as well as chemical and sensory properties of the finished wine, based solely on the properties of the grapes and the intended processing. To accomplish this, a hybrid neural network training method, Stop Training with Validation (STV), has been developed to find the most desirable neural network architecture and training level. As industrial historical data will not be evenly spaced over the entire possible search space, we have also investigated the ability of the trained neural networks to interpolate and extrapolate with data not used during training. Because a company will utilize its own existing process data for this method, the result of this work will be a general fermentation optimization method that can be applied to fermentation processes to improve quality and productivity.     

Optimization of auto-induction medium for G-CSF production by Escherichia coli using artificial neural networks coupled with genetic algorithm

Granulocyte colony-stimulating factor (G-CSF) is a cytokine widely used in cancer patients receiving high doses of chemotherapeutic drugs to prevent the chemotherapy-induced suppression of white blood cells. The production of recombinant G-CSF should be increased to meet the increasing market demand. This study aims to model and optimize the carbon source of auto-induction medium to enhance G-CSF production using artificial neural networks coupled with genetic algorithm. In this approach, artificial neural networks served as bioprocess modeling tools, and genetic algorithm (GA) was applied to optimize the established artificial neural network models. Two artificial neural network models were constructed: the back-propagation (BP) network and the radial basis function (RBF) network. The root mean square error, coefficient of determination, and standard error of prediction of the BP model were 0.0375, 0.959, and 8.49 %, respectively, whereas those of the RBF model were 0.0257, 0.980, and 5.82 %, respectively. These values indicated that the RBF model possessed higher fitness and prediction accuracy than the BP model. Under the optimized auto-induction medium, the predicted maximum G-CSF yield by the BP-GA approach was 71.66 %, whereas that by the RBF-GA approach was 75.17 %. These predicted values are in agreement with the experimental results, with 72.4 and 76.014 % for the BP-GA and RBF-GA models, respectively. These results suggest that RBF-GA is superior to BP-GA. The developed approach in this study may be helpful in modeling and optimizing other multivariable, non-linear, and time-variant bioprocesses.     

Building a dendritic actin filament network branch by branch: models of filament orientation pattern and force generation in lamellipodia

We review mathematical and computational models of the structure, dynamics, and force generation properties of dendritic actin networks. These models have been motivated by the dendritic nucleation model, which provided a mechanistic picture of how the actin cytoskeleton system powers cell motility. We describe how they aimed to explain the self-organization of the branched network into a bimodal distribution of filament orientations peaked at 35° and ??35° with respect to the direction of membrane protrusion, as well as other patterns. Concave and convex force–velocity relationships were derived, depending on network organization, filament, and membrane elasticity and accounting for actin polymerization at the barbed end as a Brownian ratchet. This review also describes models that considered the kinetics and transport of actin and diffuse regulators and mechanical coupling to a substrate, together with explicit modeling of dendritic networks.     

Enzymatic synthesis of amoxicillin: avoiding limitations of the mechanistic approach for reaction kinetics

A recurrent doubt that occurs to the enzyme‐kinetics modeler is, When should I stop adding parameters to my mechanistic model in order to fit a non‐conventional behavior? This problem becomes more and more involving when the complexity of the reaction network increases. This work intends to show how the use of artificial neural networks may circumvent the need of including an overwhelming number of parameters in the rate equations obtained through the classical, mechanistic approach. We focus on the synthesis of amoxicillin by the reaction of p‐OH‐phenylglycine methyl ester and 6‐aminopenicillanic acid, catalyzed by penicillin G acylase immobilized on glyoxyl‐agarose, at 25°C and pH 6.5. The reaction was carried on a batch reactor. Three kinetic models of this system were compared: a mechanistic, a semi‐empiric, and a hybrid–neural network (NN). A semi‐empiric, simplified model with a reasonable number of parameters was initially built‐up. It was able to portray many typical process conditions. However, it either underestimated or overestimated the rate of synthesis of amoxicillin when substrates' concentrations were low. A more complex, full‐scale mechanistic model that could span all operational conditions was intractable for all practical purposes. Finally, a hybrid model, that coupled artificial neural networks (NN) to mass‐balance equations was established, that succeeded in representing all situations of interest. Particularly, the NN could predict with accuracy reaction rates for conditions where the semi‐empiric model failed, namely, at low substrate concentrations, a situation that would occur, for instance, at the end of a fed‐batch industrial process. © 2002 Wiley Periodicals, Inc. Biotechnol Bioeng 80: 622–631, 2002.     

Overview of mathematical computer models of striated sphincter muscles

The present paper compares current mathematical striated sphincter models. Current models are subdivided in four categories: (1) simple models, (2) implementations of the urethral resistance relation, (3) models with realistic muscle dynamics and (4) finite element models. In our research group a neural network model, representing Onuf's nucleus, the spinal motor nucleus that innervates the external urethral and anal sphincters, was developed. A realistic sphincter model is needed to test the neural network. To decide whether or not a model is applicable in our research two requirements should be fulfilled: (1) the presence of realistic muscle dynamics preferably by implementation of a Huxley type muscle model and (2) the model should consist of more than one muscle unit to form a more dimensional model. Reviewing the literature, if a myogenic sphincter is modelled, mainly the Hill-equation is applied. Moreover, single muscle unit models are published. In general a multi-unit muscle model of the sphincter is lacking, prohibiting the study of the inherent properties of sphincter muscles, which could give information on the realistic behaviour of elements in circular muscles. It is concluded that the functionality of current sphincter models is limited for our purpose.     

Optimization of a fermentation medium using neural networks and genetic algorithms

Artificial neural networks and genetic algorithms are used to model and optimize a fermentation medium for the production of the enzyme hydantoinase by Agrobacterium radiobacter. Experimental data reported in the literature were used to build two neural network models. The concentrations of four medium components served as inputs to the neural network models, and hydantoinase or cell concentration served as a single output of each model. Genetic algorithms were used to optimize the input space of the neural network models to find the optimum settings for maximum enzyme and cell production. Using this procedure, two artificial intelligence techniques have been effectively integrated to create a powerful tool for process modeling and optimization.     

Preliminary screening of neural network configurations for bioreactor applications

Summary A methodology and a computer code have been devised to perform a preliminary analysis of six types of neural networks commonly employed for bioreactor problems. Both static and time-varying data can be analysed, and the values of the parameters and/or sampling times can be chosen according to the system behavior. The results help to select a suitable network configuration for detailed training and application. This is illustrated for a fed-batch fermentation to produce recombinant -galactosidase.     

A hybrid genetic--neural algorithm for modeling the biodegradation process of DnBP in AAO system

A hybrid artificial neural network - genetic algorithm numerical technique was successfully developed to model, and to simulate the biodegradation process of di-n-butyl phthalate in an anaerobic/anoxic/oxic (AAO) system. The fate of DnBP was investigated, and a removal kinetic model including sorption and biodegradation was formulated. To correlate the experimental data with available models or some modified empirical equations, the steady state model equations describing the biodegradation process have been solved using genetic algorithm (GA) and artificial neural network (ANN) from the water quality characteristic parameters. Compared with the kinetic model, the performance of the GA-ANN for modeling the DnBP was found to be more impressive. The results show that the predicted values well fit measured concentrations, which was also supported by the relatively low RMSE (0.2724), MAPE (3.6137) and MSE (0.0742)and very high R (0.9859) values, and which illustrates the GA-ANN model predicting effluent DnBP more accurately than the mechanism model forecasting.     

Mathematical modeling and optimization of cellulase protein production using Trichoderma reesei RL-P37

The enzyme cellulase, a multienzyme complex made up of several proteins, catalyzes the conversion of cellulose to glucose in an enzymatic hydrolysis-based biomass-to-ethanol process. Production of cellulase enzyme proteins in large quantities using the fungus Trichoderma reesei requires understanding the dynamics of growth and enzyme production. The method of neural network parameter function modeling, which combines the approximation capabilities of neural networks with fundamental process knowledge, is utilized to develop a mathematical model of this dynamic system. In addition, kinetic models are also developed. Laboratory data from bench-scale fermentations involving growth and protein production by T. reesei on lactose and xylose are used to estimate the parameters in these models. The relative performances of the various models and the results of optimizing these models on two different performance measures are presented. An approximately 33% lower root-mean-squared error (RMSE) in protein predictions and about 40% lower total RMSE is obtained with the neural network-based model as opposed to kinetic models. Using the neural network-based model, the RMSE in predicting optimal conditions for two performance indices, is about 67% and 40% lower, respectively, when compared with the kinetic models. Thus, both model predictions and optimization results from the neural network-based model are found to be closer to the experimental data than the kinetic models developed in this work. It is shown that the neural network parameter function modeling method can be useful as a "macromodeling" technique to rapidly develop dynamic models of a process.     

A Robust In Vivo-Like Persistent Firing Supported by a Hybrid of Intracellular and Synaptic Mechanisms

Persistent firing is believed to support short-term information retention in the brain. Established hypotheses make use of the recurrent synaptic connectivity to support persistent firing. However, this mechanism is known to suffer from a lack of robustness. On the other hand, persistent firing can be supported by an intrinsic cellular mechanism in multiple brain areas. However, the consequences of having both the intrinsic and the synaptic mechanisms (a hybrid model) on persistent firing remain largely unknown. The goal of this study is to investigate whether a hybrid neural network model with these two mechanisms has advantages over a conventional recurrent network based model. Our computer simulations were based on in vitro recordings obtained from hippocampal CA3 pyramidal cells under cholinergic receptor activation. Calcium activated non-specific cationic (CAN) current supported persistent firing in the Hodgkin-Huxley style cellular models. Our results suggest that the hybrid model supports persistent firing within a physiological frequency range over a wide range of different parameters, eliminating parameter sensitivity issues generally recognized in network based persistent firing. In addition, persistent firing in the hybrid model is substantially more robust against distracting inputs, can coexist with theta frequency oscillations, and supports pattern completion.