The potential of random forest and neural networks for biomass and recombinant protein modeling in Escherichia coli fed‐batch fermentations

Product quality assurance strategies in production of biopharmaceuticals currently undergo a transformation from empirical “quality by testing” to rational, knowledge‐based “quality by design” approaches. The major challenges in this context are the fragmentary understanding of bioprocesses and the severely limited real‐time access to process variables related to product quality and quantity. Data driven modeling of process variables in combination with model predictive process control concepts represent a potential solution to these problems. The selection of statistical techniques best qualified for bioprocess data analysis and modeling is a key criterion. In this work a series of recombinant Escherichia coli fed‐batch production processes with varying cultivation conditions employing a comprehensive on‐ and offline process monitoring platform was conducted. The applicability of two machine learning methods, random forest and neural networks, for the prediction of cell dry mass and recombinant protein based on online available process parameters and two‐dimensional multi‐wavelength fluorescence spectroscopy is investigated. Models solely based on routinely measured process variables give a satisfying prediction accuracy of about ± 4% for the cell dry mass, while additional spectroscopic information allows for an estimation of the protein concentration within ± 12%. The results clearly argue for a combined approach: neural networks as modeling technique and random forest as variable selection tool.     

Control of cultivation processes for recombinant protein production: a review

The current state-of-the-art in control of cultivation processes for recombinant protein production is examined including the quantitative knowledge that can be activated for this purpose and the measurement techniques that can be employed for control at industrial manufacturing sites.     

Comparison of viable cell concentration estimation methods for a mammalian cell cultivation process

Various mechanistic and black-box models were applied for on-line estimations of viable cell concentrations in fed-batch cultivation processes for CHO cells. Data from six fed-batch cultivation experiments were used to identify the underlying models and further six independent data sets were used to determine the performance of the estimators. The performances were quantified by means of the root mean square error (RMSE) between the estimates and the corresponding off-line measured validation data sets. It is shown that even simple techniques based on empirical and linear model approaches provide a fairly good on-line estimation performance. Best results with respect to the validation data sets were obtained with hybrid models, multivariate linear regression technique and support vector regression. Hybrid models provide additional important information about the specific cellular growth rates during the cultivation.     

Open-loop control of the biomass concentration within the growth phase of recombinant protein production processes

Recombinant protein production processes are typically divided into two phases. In the first one, pure cell propagation takes place, while in the second one product formation is switched on within the cells by adding an inducer. In the initial biomass formation phase, the cell density is rather low and, hence, the measurement quantities that could be used to determine the process' state depict small values and are rather severely distorted by measurement noise. Because of these measurement problems, the fermentation cannot be reliably controlled by feedback control during this first production phase; instead, the process must be controlled in an open-loop fashion. The consequence, worked out in this paper, is to design substrate feed rate profiles for the growth phase in such a way that they are robust with respect to the main disturbances observed in practice. The robustness of the biomass formation is shown to be primarily dependent on the specific growth rate adjusted in the first hours. High batch-to-batch reproducibility can be obtained with exponential feeding profiles F(t) corresponding to specific growth rates micro(set) well below the maximal specific growth rate micro(max) of the organism. The reduction in the growth rate needed to obtain a robust process behavior depends on the inaccuracies in the initial biomass concentrations. Quantitative feed rate profiles were obtained by numerical simulation and these results were validated experimentally by means of a series of cultivation runs, where a recombinant pharmaceutical protein was produced. All experimental data confirmed the assumptions made in the robust process design study.     

Agroforestry systems for the production of woody biomass for energy transformation purposes

In the temperate zone, agroforestry systems come increasingly into focus as they offer an approach for the production of fuelwood, thus matching the increasing demand for a self-supply with bioenergy in rural decentralized areas. Because of the large area potential of marginal land, research activities aimed at a reliable estimation of the minimum productivity of fast–growing tree species under most unfavourable site conditions.Two agroforestry systems were established on reclaimed mine sites in NE-Germany (Lusatia) and Central Germany (Helmstedt). The yield potential and the sustainability of yields were studied for different clones of poplar (Populus spp.), willow (Salix viminalis L.), and black locust (Robinia pseudoacacia L.), considering different rotation periods (3-, 6-, and 9-year-rotation) and approaches of soil amelioration (mineral fertiliser, compost).On both sites the highest yields of woody biomass were found for R. pseudoacacia L. regardless of rotation period and amelioration measures. On loamy substrates in the Helmstedt mining district, all tree species and clones responded positively to soil amelioration measures.In the agroforestry system in Lusatia, special emphasis was given to the interaction between trees (R. pseudoacacia) and crops (Medicago sativa L.). Considering the land equivalent ratio (LER), R. pseudoacacia hedgerows have practically no negative influence on yields of M. sativa. Hence, with regard to an increasing demand for woody biomass, alley cropping with R. pseudoacacia and crops such as M. sativa may provide a promising alternative for future land use in the temperate zone.     

A comparison of methods for predicting vegetation type

Predictive modeling of vegetation patterns has wide application in vegetation science. In this paper I discuss three methods of predictive modeling using data from the alpine treeline ecotone as a case study. The study area is a portion of Glacier National Park, Montana. Parametric general linear models (GLM), artificial neural networks (ANN) and classification tree (CT) methods of predicting vegetation type are compared to determine the relative strength of each predictive approach and how they may be used in concert to increase understanding of important vegetation – environment relations. For each predictive method, vegetation type within the alpine treeline ecotone is predicted using a suite of environmental indicator variables including elevation, moisture potential, solar radiation potential, snow potential index, and disturbance history. Results from each of the predictive methods are compared against the real vegetation types to determine the relative accuracy of the methods.When the entire data field is examined (i.e., not evaluated by smaller spatial aggregates of data) the ANN procedure produces the most accurate predictions (=0.571); the CT predictions are the least accurate (=0.351). The predicted patterns of vegetation on the landscape are considerably different using the three methods. The GLM and CT methods produce large contiguous swaths of vegetation types throughout the study area, whereas the ANN method produces patterns with much more heterogeneity and smaller patches.When predictions are compared to reality at catchment scale, it becomes evident that the accuracy of each method varies depending upon the specific situation. The ANN procedure remains the most accurate method in the majority of the catchments, but both the GLM and PCT produce the most accurate classifications in at least one basin each.The variability in predictive ability of the three methods tested here indicates that there may not be a single best predictive method. Rather it may be important to use a suite of predictive models to help understand the environment – vegetation relationships. The ability to use multiple predictive methods to determine which spatial subunits of a landscape are outliers is important when identifying locations useful for climate change monitoring studies.     

On-line biomass estimation in biosurfactant production process by <Emphasis Type="Italic">Candida lipolytica</Emphasis> UCP 988

Biomass is an important variable in biosurfactant production process. However, such bioprocess variable, usually, is collected by sampling and determined by off-line analysis, with significant time delay. Therefore, simple and reliable on-line biomass estimation procedures are highly desirable. An artificial neural network model (ANN) is presented for the on-line estimation of biomass concentration, in biosurfactant production by Candida lipolytica UCP 988, as a nonlinear function of pH and dissolved oxygen. Several configurations were evaluated while developing the optimal ANN model. The optimal ANN model consists of one hidden layer with four neurons. The performance of the ANN was checked using experimental data. The results obtained indicate a very good predictive capacity for the ANN-based software sensor with values of R2 of 0.969 and RMSE of 0.021 for biomass concentration. Estimated biomass using the ANN was proved to be a simple, robust and accurate method.     

Monitoring of recombinant protein production using bioluminescence in a semiautomated fermentation process

On-line optimization of fermentation processes can be greatly aided by the availability of information on the physiological state of the cell. The goal of our "BioLux" research project was to design a recombinant cell capable of intracellular monitoring of product synthesis and to use it as part of an automated fermentation system. A recombinant plasmid was constructed containing an inducible promoter that controls the gene coding for a model protein and the genes necessary for bioluminescence. The cells were cultured in microfermenters equipped with an on-line turbidity sensor and a specially designed on-line light sensor capable of continuous measurement of bioluminescence. Initial studies were done under simple culture conditions, and a linear correlation between luminescence and protein production was obtained. Such specially designed recombinant bioluminescent cells can potentially be applied for model-based inference of intracellular product formation, as well as for optimization and control of recombinant fermentation processes.     

Modeling of biomass production by Spirulinaplatensis as function of phosphate concentrations and pH regimes

Batch cultivations were performed to evaluate the influence of phosphate concentrations (0.25, 0.5, 0.75, and 1.0 g L⁻¹) for pH regimes (9.5, 10.0, and 10.5) on the biomass production by Spirulina platensis. The best condition for cell growth (3.099 g L⁻¹) was found at 0.5 g L⁻¹ phosphate and pH value of 10.0. Cultivation time, phosphate, and pH caused to increase significantly (p < 0.01) in biomass production by S. platensis. Lag time was observed up to 4 h. After then, biomass production increased sharply (p < 0.01) from 0.020 g L⁻¹ to 2.063, 2.213, 1.532, and 0.797 g L⁻¹ at 0.25, 0.5, 0.75, and 1.0 g L⁻¹ phosphate values, respectively. Modified Gompertz model could be regarded as sufficient to describe the biomass production by S. platensis with high determination coefficients and low sum of square value indicated that. Biological parameters for biomass production were successfully predicted by modified Gompertz model.     

Factors affecting the bioconversion of Philippine tung seed by black soldier fly larvae for the production of protein and oil-rich biomass

A systematic study on the use of Philippine tung (Reutealis trisperma) seed as a substrate for the cultivation of black soldier fly larvae (Hermetia illucens) was performed. The characteristics of Reutealis trisperma seed from two different locations: West Java and Papua, were determined. The seed has a relatively high oil (37.6–39.2%, dry weight) and protein content (14.9–28.2%, dry weight). The effect of cake content in the substrate (0–20%, wet weight), moisture content in the substrate (50–70%, wet weight), feeding rate (50–100?mg/larva/d), lighting condition (dark-light) and substrate depth in a rearing container (4–10?cm) was performed. An optimum prepupal biomass productivity of 123.4?g/m²/d was obtained (20%, wet weight of cake content in the substrate, 60%, wet weight of moisture content in the substrate, 100?mg/larva/d, dark, 6?cm substrate depth). The protein and oil content of the biomass were also determined to evaluate the effect of Reutealis trisperma seed as a substrate for the cultivation of black soldier fly larvae to produce protein and oil-rich biomass. The oil content in the biomass was also extracted and the fatty acid composition was identified. The prepupal biomass has a relatively high amount of protein (45%, dry weight) and oil content (26.6%, dry weight) and is suitable for cattle feed application.     

Spruce sugars and poultry hydrolysate as growth medium in repeated fed-batch fermentation processes for production of yeast biomass

Bioprocess and Biosystems Engineering - The production of microbial protein in the form of yeast grown on lignocellulosic sugars and nitrogen-rich industrial residues is an attractive approach for...     

Simple control of fed-batch processes for recombinant protein production with E. coli

A very simple but effective process control technique is proposed that leads to a high batch-to-batch reproducibility with respect to biomass concentration as well as the specific biomass growth rate profiles in E. coli fermentations performed during recombinant protein production. It makes use of the well-established temperature controllers in currently used fermenters, but takes its information from the difference between the controlled culture temperature T (cult) and the temperature T (coolin) of the coolant fed to the fermenter's cooling jacket as adjusted by the fermenter temperature controller. For process control purposes this measured difference is corrected regarding stirrer influences and cumulated before it is used as a new process control variable. As a spin-off of this control, it becomes possible to estimate online the oxygen mass transfer rates and the corresponding k(L)a values during the real cultivation process.     

生物强化技术在生物质沼气制备过程中的应用及研究进展

生物强化技术通过为特定的生物过程"设计"微生物,进而作为一种提升反应系统活力和性能的手段被应用于生物质沼气制备过程,以便加快发酵系统启动时间、增加原料利用率、缩短酸败系统的恢复时间、降低高有机负荷的抑制作用等。本文针对以木质纤维素为原料的沼气制备中的生物强化技术,从生物强化菌剂的构建及标准、生物强化作用的影响因素、生物强化作用机制的探究等几个方面来阐述目前国内外生物强化技术在生物质沼气制备过程中的应用与研究进展,以及存在的问题和解决方案。     

Anaerobic biotechnological approaches for production of liquid energy carriers from biomass

In recent years, increasing attention has been paid to the use of renewable biomass for energy production. Anaerobic biotechnological approaches for production of liquid energy carriers (ethanol and a mixture of acetone, butanol and ethanol) from biomass can be employed to decrease environmental pollution and reduce dependency on fossil fuels. There are two major biological processes that can convert biomass to liquid energy carriers via anaerobic biological breakdown of organic matter: ethanol fermentation and mixed acetone, butanol, ethanol (ABE) fermentation. The specific product formation is determined by substrates and microbial communities available as well as the operating conditions applied. In this review, we evaluate the recent biotechnological approaches employed in ethanol and ABE fermentation. Practical applicability of different technologies is discussed taking into account the microbiology and biochemistry of the processes.     

A novel feeding strategy for enhanced protein production by fed-batch fermentation in recombinant Pichia pastoris

A novel feeding strategy for enhanced protein production of hepatitis B virus surface antigen (HBsAg) in fed-batch fermentation, recombinant Pichia pastoris, has been developed. A minimal salt medium was used to grow cells in the initial batch fermentation, followed by a glycerol+salts fed-batch phase. At the end of the fed-batch phase a dry cell weight of 130 g l⁻¹ was achieved. In the absence of basal salts, the same amount of glycerol feed resulted in only 90 g l⁻¹ cell dry weight. When a limited amount of casamino acids were also included every 24 h during methanol induction, there was a two-fold increase in expression levels of HBsAg. After 192 h of induction, the expression levels of HBsAg (soluble and insoluble) reached >1 g l⁻¹ using the Mut⁻ strain. Thus, the use of basal salts in the glycerol feed, along with the addition of limited amounts of casamino acids with the methanol feed, resulted in an increased expression of total HBsAg.     

On-line monitoring and control of substrate concentrations in biological processes by flow injection analysis systems

Concentrations of substrates, glucose, and ammionia in biological processes have been on-line monitored by using glucose-flow injection (FIA) and ammonia-FIA systems. Based on the on-line monitored data the concentrations of substrates have been controlled by an on-off controller, a PID controller, and a neural network (NN) based controller. A simulation program has been developed to test the control quality of each controller and to estimate the control parameters. The on-off controller often produced high oscillations at the set point due to its low robustness. The control quality of a PID controller could have been improved by a high analysis frequency and by a short residence time of sample in a FIA system. A NN-based controller with 3 layers has been developed, and a 3(input)-2(hidden)-1(output) network structure has been found to be optimal for the NN-based controller. The performance of the three controllers has been tested in a simulated process as well as in a cultivation process ofSaccharomyces cerevisiae, and the performance has also been compared to simulation results. The NN-based controller with the 3-2-1 network structure was robust and stable against some disturbances, such as a sudden injection of distilled water into a biological process.     

Use of oil bodies and oleosins in recombinant protein production and other biotechnological applications

Oil bodies obtained from oilseeds have been exploited for a variety of applications in biotechnology in the recent past. These applications are based on their non-coalescing nature, ease of extraction and presence of unique membrane proteins—oleosins. In suspension, oil bodies exist as separate entities and, hence, they can serve as emulsifying agent for a wide variety of products, ranging from vaccines, food, cosmetics and personal care products. Oil bodies have found significant uses in the production and purification of recombinant proteins with specific applications. The desired protein can be targeted to oil bodies in oilseeds by affinity tag or by fusing it directly to the N or C terminal of oleosins. Upon targeting, the hydrophobic domain of oleosin embeds into the TAG matrix of oil body, whereas the protein fused with N and/or C termini is exposed on the oil body surface, where it acquires correct confirmation spontaneously. Oil bodies with the attached foreign protein can be separated easily from other cellular components. They can be used directly or the protein can be cleaved from the fusion. The desired protein can be a pharmaceutically important polypeptide (e.g. hirudin, insulin and epidermal growth factor), a neutraceutical polypeptide (somatotropin), a commercially important enzyme (e.g. xylanase), a protein important for improvement of crops (e.g. chitinase) or a multimeric protein. These applications can further be widened as oil bodies can also be made artificially and oleosin gene can be expressed in bacterial systems. Thus, a protein fused to oleosin can be expressed in Escherichia coli and after cell lysis it can be incorporated into artificial oil bodies, thereby facilitating the extraction and purification of the desired protein. Artificial oil bodies can also be used for encapsulation of probiotics. The manipulation of oleosin gene for the expression of polyoleosins has further expanded the arena of the applications of oil bodies in biotechnology.     

Evaluation of pelleting as a pre-processing step for effective biomass deconstruction and fermentation

Densification of bulky forages by pelleting reduces their transportation, handling, and storage costs. Because of high shearing force and frictional heating during the pelleting process, it is hypothesized that pelleting of lignocellulosic biomass could also partially deconstruct its complex structure and facilitate bioethanol production. In this study, pelleted wheat straw, corn stover, big bluestem, and sorghum stalk were evaluated for sugars and ethanol production, and compared with those of unpelleted biomasses. Mass recovery after alkali pretreatment increased by 14%, 11%, 2%, and 5%, respectively, in unpelleted biomasses. Lignin content reduced significantly more in pelleted samples for all types of biomass, except sorghum stalk. Volumetric productivity of enzymatic hydrolysis was 23%, 21%, 20% and 12% higher, respectively, in pelleted samples; ethanol yield on the basis of released sugars did not differ significantly between pelleted and unpelleted samples. These results indicate that the pelleting process led to better enzymatic hydrolysis of pretreated biomasses without affecting the quality of sugars for fermentation. However, overall yield of ethanol from the raw biomass was not significantly higher in pelleted biomasses because of higher mass loss during pretreatment process. In our study, we propose a schematic for complete utilization of various byproducts for enhanced economic viability.