A robust soft sensor to monitor 1,3-propanediol fermentation process by Clostridium butyricum based on artificial neural network

A comprehensive review on the application of neural network model in microbial fermentation

The development of high-performance strains and the continuous breakthrough of strain screening technology also pose challenges to downstream fermentation optimization and scale-up. Therefore, neural network models are utilized to optimize the fermentation process to meet the goals of boosting yield or lowering cost, with the use of artificial intelligence technology in conjunction with the peculiarities of the fermentation process. High-performance strains' yield rise and fermentation process amplification will be sped up with the aid of neural network models. This paper offers a helpful review for anyone interested in state-of-the-art microbial fermentation processes, as it thoroughly reviews the application of neural network models in predicting fermentation yield, optimizing the fermentation process, and monitoring the fermentation process.

Bringing a scalable adaptive hybrid modeling framework closer to industrial use: Application on a multiscale fungal fermentation

Digitalization has paved the way for new paradigms such as digital shadows and digital twins for fermentation processes, opening the door for real-time process monitoring, control, and optimization. With a digital shadow, real-time model adaptation to accommodate complex metabolic phenomena such as metabolic shifts of a process can be monitored. Despite the many benefits of digitalization, the potential has not been fully reached in the industry. This study investigates the development of a digital shadow for a very complex fungal fermentation process in terms of microbial physiology and fermentation operation on pilot-scale at Novonesis and the challenges thereof. The process has historically been difficult to optimize and control due to a lack of offline measurements and an absence of biomass measurements. Pilot-scale and lab-scale fermentations were conducted for model development and validation. With all available pilot-scale data, a data-driven soft sensor was developed to estimate the main substrate concentration (glucose) with a normalized root mean squared error (N-RMSE) of 2%. This robust data-driven soft sensor was able to estimate accurately in lab-scale (volume < 20× pilot) with a N-RMSE of 7.8%. A hybrid soft sensor was developed by combining the data-driven soft sensor with a mass balance to estimate the glycerol and biomass concentrations on pilot-scale data with N-RMSEs of 11% and 21%, respectively. A digital shadow modeling framework was developed by coupling a mechanistic model (MM) with the hybrid soft sensor. The digital shadow modeling framework significantly improved the predictability compared with the MM. The contribution of this study brings the application of digital shadows closer to industrial implementation. It demonstrates the high potential of using this type of modeling framework for scale-up and leads the way to a new generation of in silico-based process development.

Modeling and optimization of sporulation by Bacillus licheniformis BF-002 based on dynamics and recurrent neural networks

Bacillus licheniformis is widely utilized in disease prevention and environmental remediation. Spore quantity is a critical factor in determining the quality of microbiological agents containing vegetative cells. To improve the understanding of Bacillus licheniformis BF-002 strain culture, a hybrid model integrating traditional dynamic modeling and recurrent neural network was developed. This model enabled the optimization of carbon/nitrogen source feeding rates, pH, temperature and agitation speed using genetic algorithms. Carbon and nitrogen source consumption in the optimal duplicate batches showed no significant difference compared to the control batch. However, the spore quantity in the broth increased by 16.2% and 35.2% in the respective duplicate batches. Overall, the hybrid model outperformed the traditional dynamic model in accurately tracking the cultivation dynamics of Bacillus licheniformis, leading to increased spore production when used for optimizing cultivation conditions.

Enhancing and monitoring spore production in Clostridium butyricum using pH-based regulation strategy and a robust soft sensor based on back-propagation neural networks

Clostridium butyricum is a probiotic that forms anaerobic spores and plays a crucial role in regulating gut microbiota. However, the total viable cell count and spore yield of C. butyricum in industrial production are comparatively low. To this end, we investigated the metabolic characteristics of the strain and proposed three distinct pH regulation strategies for enhancing spore production. In addition, precise measurement of fermentation parameters such as substrate concentration, total viable cell count, and spore concentration is crucial for successful industrial probiotics production. Nevertheless, online measurement of these intricate parameters in the fermentation of C. butyricum poses a considerable challenge owing to the complex, nonlinear, multivariate, and strongly coupled characteristics of the production process. Therefore, we analyzed the capacitance and conductivity acquired from a viable cell sensor as the core parameters for the fermentation process. Subsequently, a robust soft sensor was developed using a seven-input back-propagation neural network model with input variables of fermentation time, capacitance, conductivity, pH, initial total sugar concentration, ammonium ion concentration, and calcium ion concentration. The model enables the online monitoring of total viable biomass count, substrate concentrations, and spore yield, and can be extended to similar fermentation processes with pH changes as a characteristic feature.

Application of artificial intelligence-based methods in bioelectrochemical systems: Recent progress and future perspectives

Bioelectrochemical Systems (BESs) leverage microbial metabolic processes to either produce electricity by degrading organic matter or consume electricity to assist metabolism, and can be used for various applications such as energy production, wastewater treatment, and bioremediation. Given the intricate mechanisms of BESs, the application of artificial intelligence (AI)-based methods have been proposed to enhance the performance of BESs due to their capability to identify patterns and gain insights through data analysis. This review focuses on the analysis and comparison of AI algorithms commonly used in BESs, including artificial neural network (ANN), genetic programming (GP), fuzzy logic (FL), support vector regression (SVR), and adaptive neural fuzzy inference system (ANFIS). These algorithms have different features, such as ANN's simple network structure, GP's use in the training process, FL's human-like thought process, SVR's high prediction accuracy and robustness, and ANFIS's combination of ANN and FL features. The AI-based methods have been applied in BESs to predict microbial communities, products or substrates, and reactor performance, which can provide valuable information and improve system efficiency. Limitations of AI-based methods for predicting and optimizing BESs and recommendations for future development are also discussed. This review demonstrates the potential of AI-based methods in optimizing BESs and provides valuable information for the future development of this field.

From Spatial-Temporal Multiscale Modeling to Application: Bridging the Valley of Death in Industrial Biotechnology

The Valley of Death confronts industrial biotechnology with a significant challenge to the commercialization of products. Fortunately, with the integration of computation, automation and artificial intelligence (AI) technology, the industrial biotechnology accelerates to cross the Valley of Death. The Fourth Industrial Revolution (Industry 4.0) has spurred advanced development of intelligent biomanufacturing, which has evolved the industrial structures in line with the worldwide trend. To achieve this, intelligent biomanufacturing can be structured into three main parts that comprise digitalization, modeling and intellectualization, with modeling forming a crucial link between the other two components. This paper provides an overview of mechanistic models, data-driven models and their applications in bioprocess development. We provide a detailed elaboration of the hybrid model and its applications in bioprocess engineering, including strain design, process control and optimization, as well as bioreactor scale-up. Finally, the challenges and opportunities of biomanufacturing towards Industry 4.0 are also discussed.

A robust soft sensor based on artificial neural network for monitoring microbial lipid fermentation processes using Yarrowia lipolytica

Microbial oils produced by Yarrowia lipolytica offer an environmentally friendly and sustainable alternative to petroleum as well as traditional lipids from animals and plants. The accurate measurement of fermentation parameters, including the substrate concentration, dry cell weight, and lipid accumulation, is the foundation of process control, which is indispensable for industrial lipid production. However, it remains a great challenge to measure the complex parameters online during the lipid fermentation process, which is nonlinear, multivariate, and characterized by strong coupling. As a type of AI technology, the artificial neural network model is a powerful tool for handling extremely complex problems, and it can be employed to develop a soft sensor to monitor the microbial lipid fermentation process of Y. lipolytica. In this study, we first analyzed and emphasized the volume of sodium hydroxide and dissolved oxygen concentration as central parameters of the fermentation process. Then, a soft sensor based on a four-input artificial neural network model was developed, in which the input variables were fermentation time, dissolved oxygen concentration, initial glucose concentration, and additional volume of sodium hydroxide. This provides the possibility of online monitoring of dry cell weight, glucose concentration, and lipid production with high accuracy, which can be extended to similar fermentation processes characterized by the addition of bases or acids, as well as changes of the dissolved oxygen concentration.

Prediction of surfactin fermentation with Bacillus subtilis DSM10 by response surface methodology optimized artificial neural network

Biosurfactants produced by Bacillus species are an emerging group of surface-active molecules. They have excellent surface tension reducer and high emulsifier properties. Generally, the biosurfactant fermentation leads to a low product concentration. Therefore, our goal was to investigate Bacillus subtilis DSM10 production and improve the biosurfactant content in the broth by media optimization via response surface methodology. The optimal combinations of the investigated factors were determined as the following: pH = 9, glucose = 20 g/L, and NH₄ NO₃  = 2 g/L. Under the optimized conditions, the formed surfactin strain reduced surface tension in the broth by 48% (from 72 to 37 mN/m) and the isolated product by 63% (from 72 to 27 mN/m). An artificial neural network was built based on the results of response surface methodology to predict the product quality and the harvesting time of broth. Thus, finally, the model can predict the final cell and product amount, and even their time course, with around 90% reliability.

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.

In situ carbon dioxide capture to co-produce 1,3-propanediol, biohydrogen and micro-nano calcium carbonate from crude glycerol by Clostridium butyricum

Background

Climate change caused by greenhouse gas emission has become a global hot topic. Although biotechnology is considered as an environmentally friendly method to produce chemicals, almost all biochemicals face carbon dioxide emission from inevitable respiration and energy metabolism of most microorganisms. To cater for the broad prospect of biochemicals, bioprocess optimization of diverse valuable products is becoming increasingly important for environmental sustainability and cleaner production. Based on Ca(OH)₂ as a CO₂ capture agent and pH regulator, a bioprocess was proposed for co-production of 1,3-propanediol (1,3-PDO), biohydrogen and micro-nano CaCO₃ by Clostridium butyricum DL07.

Results

In fed-batch fermentation, the maximum concentration of 1,3-PDO reached up to 88.6 g/L with an overall productivity of 5.54 g/L/h. This productivity is 31.9% higher than the highest value previously reports (4.20 g/L/h). In addition, the ratio of H₂ to CO₂ in exhaust gas showed a remarkable 152-fold increase in the 5 M Ca(OH)₂ group compared to 5 M NaOH as the CO₂ capture agent. Green hydrogen in exhaust gas ranged between 17.2% and 20.2%, with the remainder being N₂ with negligible CO₂ emissions. During CO₂ capture in situ, micro-nano calcite particles of CaCO₃ with sizes in the range of 300 nm to 20 µm were formed simultaneously. Moreover, when compared with 5M NaOH group, the concentrations of soluble salts and proteins in the fermentation broth of 5 M Ca(OH)₂ group were notably reduced by 53.6% and 44.1%, respectively. The remarkable reduction of soluble salts and proteins would contribute to the separation of 1,3-PDO.

Conclusions

Ca(OH)₂ was used as a CO₂ capture agent and pH regulator in this study to promote the production of 1,3-PDO. Meanwhile, micro-nano CaCO₃ and green H₂ were co-produced. In addition, the soluble salts and proteins in the fermentation broth were significantly reduced.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s13068-022-02190-2.

Mutation of rpoS is Beneficial for Suppressing Organic Acid Secretion During 1,3-Propandiol Biosynthesis in Klebsiella pneumoniae

In this study, to reduce the formation of organic acid during 1,3-propanediol biosynthesis in Klebsiella pneumoniae, a method combining UV mutagenesis and high-throughput screening with pH color plates was employed to obtain K. pneumoniae mutants. When compared with the parent strain, the total organic acid formation by the mutant decreased, whereas 1,3-propanediol biosynthesis increased after 24 h anaerobic shake flask culture. Subsequently, genetic changes in the mutant were analyzed by whole-genome sequencing and verified by signal gene deletion. Mutation of the rpoS gene was confirmed to contribute to the regulation of organic acid synthesis in K. pneumoniae. Besides, rpoS deletion eliminated the formation of 2,3-butanediol, the main byproduct produced during 1,3-propanediol fermentation, indicating the role of rpoS in metabolic regulation in K. pneumoniae. Thus, a K. pneumoniae mutant was developed, which could produce lower organic acid during 1,3-propanediol fermentation due to an rpoS mutation in this study.

Challenges in the Development of Soft Sensors for Bioprocesses: A Critical Review

Among the greatest challenges in soft sensor development for bioprocesses are variable process lengths, multiple process phases, and erroneous model inputs due to sensor faults. This review article describes these three challenges and critically discusses the corresponding solution approaches from a data scientist’s perspective. This main part of the article is preceded by an overview of the status quo in the development and application of soft sensors. The scope of this article is mainly the upstream part of bioprocesses, although the solution approaches are in most cases also applicable to the downstream part. Variable process lengths are accounted for by data synchronization techniques such as indicator variables, curve registration, and dynamic time warping. Multiple process phases are partitioned by trajectory or correlation-based phase detection, enabling phase-adaptive modeling. Sensor faults are detected by symptom signals, pattern recognition, or by changing contributions of the corresponding sensor to a process model. According to the current state of the literature, tolerance to sensor faults remains the greatest challenge in soft sensor development, especially in the presence of variable process lengths and multiple process phases.

Bioprocess Control: Current Progress and Future Perspectives

Typical bioprocess comprises of different unit operations wherein a near optimal environment is required for cells to grow, divide, and synthesize the desired product. However, bioprocess control caters to unique challenges that arise due to non-linearity, variability, and complexity of biotech processes. This article presents a review of modern control strategies employed in bioprocessing. Conventional control strategies (open loop, closed loop) along with modern control schemes such as fuzzy logic, model predictive control, adaptive control and neural network-based control are illustrated, and their effectiveness is highlighted. Furthermore, it is elucidated that bioprocess control is more than just automation, and includes aspects such as system architecture, software applications, hardware, and interfaces, all of which are optimized and compiled as per demand. This needs to be accomplished while keeping process requirement, production cost, market value of product, regulatory constraints, and data acquisition requirements in our purview. This article aims to offer an overview of the current best practices in bioprocess control, monitoring, and automation.