Artificial neural network and regression coupled genetic algorithm to optimize parameters for enhanced xylitol production by Debaryomyces nepalensis in bioreactor

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.

A journey with psychedelic mushrooms: From historical relevance to biology, cultivation, medicinal uses, biotechnology, and beyond

Psychedelic mushrooms containing psilocybin and related tryptamines have long been used for ethnomycological purposes, but emerging evidence points to the potential therapeutic value of these mushrooms to address modern neurological, psychiatric health, and related disorders. As a result, psilocybin containing mushrooms represent a re-emerging frontier for mycological, biochemical, neuroscience, and pharmacology research. This work presents crucial information related to traditional use of psychedelic mushrooms, as well as research trends and knowledge gaps related to their diversity and distribution, technologies for quantification of tryptamines and other tryptophan-derived metabolites, as well as biosynthetic mechanisms for their production within mushrooms. In addition, we explore the current state of knowledge for how psilocybin and related tryptamines are metabolized in humans and their pharmacological effects, including beneficial and hazardous human health implications. Finally, we describe opportunities and challenges for investigating the production of psychedelic mushrooms and metabolic engineering approaches to alter secondary metabolite profiles using biotechnology integrated with machine learning. Ultimately, this critical review of all aspects related to psychedelic mushrooms represents a roadmap for future research efforts that will pave the way to new applications and refined protocols.

Machine learning in bioprocess development: from promise to practice

Fostered by novel analytical techniques, digitalization, and automation, modern bioprocess development provides large amounts of heterogeneous experimental data, containing valuable process information. In this context, data-driven methods like machine learning (ML) approaches have great potential to rationally explore large design spaces while exploiting experimental facilities most efficiently. Herein we demonstrate how ML methods have been applied so far in bioprocess development, especially in strain engineering and selection, bioprocess optimization, scale-up, monitoring, and control of bioprocesses. For each topic, we will highlight successful application cases, current challenges, and point out domains that can potentially benefit from technology transfer and further progress in the field of ML.

Artificial intelligence technologies in bioprocess: Opportunities and challenges

Bioprocess control and optimization are crucial for tapping the metabolic potential of microorganisms, and which have made great progress in the past decades. Combination of the current control and optimization technologies with the latest computer-based strategies will be a worth expecting way to improve bioprocess further. Recently, artificial intelligence (AI) emerged as a data-driven technique independent of the complex interactions used in mathematical models and has been gradually applied in bioprocess. In this review, firstly, AI-guided modeling approaches of bioprocess are discussed, which are widely applied to optimize critical process parameters (CPPs). Then, AI-assisted rapid detection and monitoring technologies employed in bioprocess are summarized. Next, control strategies according to the above two technologies in bioprocess are analyzed. Lastly, current research gaps and future perspectives on AI-guided optimization and control technologies are discussed. This review provides theoretical guidance for developing AI-guided bioprocess optimization and control technologies.

Using neural networks and a fuzzy inference system to evaluate the risk of wildfires and the pinpointing of firefighting stations in forests on the northern slopes of the Zagros Mountains, Iran (case study: Shimbar national wildlife preserve)

Predicting potential fire hazard zones in natural areas is one of the means of mitigating and managing fires. The current research focuses on the prioritizing of elements which contribute to the spread of fire and the special zoning of potentially dangerous areas in addition to the pinpointing of locations for the establishment of fire stations in forested areas in the Shimbar national reserve based on historical data spanning 2001 to 2018. The study utilizes elements (physiological, vegetation cover, meteorological, anthropological factors) contributing to wildfires as inputs into an artificial neural network and the development of a fuzzy inference system in order to produce fire zoning maps for the region under study. The map is divided into five sectors, i.e., minimum, low, moderate, high, and maximum risk of fire. The validation of the fire zoning map was evaluated at 0.83 and the RMSE error was 0.75. The results obtained show that 20% of the area under study is within the average risk category, 11% is within the high-risk category, and 10% is within the very high-risk category of a potential fire hazard. The most important variables were distance from a flowing source, i.e., river or stream, the land formation type, elevation, and the minimum temperature. The identification of suitable locations for firefighting stations was carried out by merging the fuzzy inference system model and Arc GIS, and the results obtained defined 16 possible locations. It was concluded that the application of hybrid models when dealing with the aforementioned variables is effective when seeking to determine locations for the establishment of firefighting stations and rural safety services; moreover, such hybrid models are highly efficacious for determining of fire hazard zones. It is proposed that hybrid models be applied on a large scale for the prevention, control, and management of fires throughout the country.

Recent advances in machine learning applications in metabolic engineering

Metabolic engineering encompasses several widely-used strategies, which currently hold a high seat in the field of biotechnology when its potential is manifesting through a plethora of research and commercial products with a strong societal impact. The genomic revolution that occurred almost three decades ago has initiated the generation of large omics-datasets which has helped in gaining a better understanding of cellular behavior. The itinerary of metabolic engineering that has occurred based on these large datasets has allowed researchers to gain detailed insights and a reasonable understanding of the intricacies of biosystems. However, the existing trail-and-error approaches for metabolic engineering are laborious and time-intensive when it comes to the production of target compounds with high yields through genetic manipulations in host organisms. Machine learning (ML) coupled with the available metabolic engineering test instances and omics data brings a comprehensive and multidisciplinary approach that enables scientists to evaluate various parameters for effective strain design. This vast amount of biological data should be standardized through knowledge engineering to train different ML models for providing accurate predictions in gene circuits designing, modification of proteins, optimization of bioprocess parameters for scaling up, and screening of hyper-producing robust cell factories. This review briefs on the premise of ML, followed by mentioning various ML methods and algorithms alongside the numerous omics datasets available to train ML models for predicting metabolic outcomes with high-accuracy. The combinative interplay between the ML algorithms and biological datasets through knowledge engineering have guided the recent advancements in applications such as CRISPR/Cas systems, gene circuits, protein engineering, metabolic pathway reconstruction, and bioprocess engineering. Finally, this review addresses the probable challenges of applying ML in metabolic engineering which will guide the researchers toward novel techniques to overcome the limitations.

Artificial neural network modeling and statistical optimization of medium components to enhance production of exopolysaccharide by Bacillus sp. EPS003

Microbial Exopolysaccharides (EPS) have a wide range of applications in food, cosmetics, agriculture, pharmaceutical industries, and environmental bioremediation. The present study aims at enhancing the production of EPS from a soil-isolate Bacillus sp. EPS003. Effects of carbon and nitrogen sources and process conditions were evaluated one factor at a time. Box-Behnken design has been used and a 2.5-fold increase in yield is reported after optimizing the most influential parameters sucrose, yeast extract, and peptone as identified by the Plackett-Burman method. An artificial neural network (ANN) with two different topologies (EPS-NN1 and EPS-NN2) was developed. On comparing prediction accuracy, EPS-NN2 formulated as one input layer with four input variables (sucrose, yeast extract, peptone, biomass), a single hidden layer with seven neurons and EPS yield in the output layer showed a high coefficient of determination (R²-0.98) and low error (NRMSE-0.024). This study concludes that the consideration of biomass value has increased the prediction accuracy of the model.

Thermoluminescence Characteristic of Calcite with Gaussian Process Regression Model of Machine Learning

Thermoluminescence (TL) is defined as a luminescent phenomenon that can be detected when an insulator or semiconductor is thermally stimulated. Defective crystals store radiation until they are stimulated. Thermoluminescence is a method of monitoring the absorbed dose of dosimeters. The irradiation crystal is heated to 500 °C to display the absorbed dose as a luminescent light. The thermoluminescence dosimetric properties of calcite obtained from nature were investigated in this study. Machine learning (ML) was also examined utilizing Gaussian process regression (GPR) for stimulated TL characteristics. According to the experimental output, the TL glow curve has two main peaks located at 90 ^o C and 240 ^o C with good dosimetric properties. In the four regression models of GPR, the data of heating rate of 3 ^o C/sec has the lowest residual.

Modeling of malachite green adsorption onto novel polyurethane/SrFe12O19/clinoptilolite nanocomposite using response surface methodology and biogeography-based optimization-assisted multilayer neural network

This research studied the modeling of malachite green (MG) adsorption onto novel polyurethane/SrFe₁₂O₁₉/clinoptilolite (PU/SrM/CLP) nanocomposite from aqueous solutions by the application of biogeography-based optimization (BBO) algorithm-assisted multilayer neural networks (MNN-BBO) as a new evolutionary algorithm in environmental science. The PU/SrM/CLP nanocomposite was successfully fabricated and characterized by some spectroscopic analyses. Four variables influencing the removal efficiency were modeled by MNN-BBO and response surface methodology (RSM). The MNN-BBO model gave higher percentage removal (99.6%) about 7.6% compared to the RSM technique. Under optimal conditions obtained by MNN-BBO, the four independent variables including pH, shaking rate, initial concentration, and adsorbent dosage were 6.5, 255 rpm, 50 mg.L^-1, and 0.08 g, respectively. Under these conditions, the results were fitted well to the Langmuir isotherm with a monolayer maximum amount of sorbate uptake (q_max) of 68.49 mg.g^-1 and the pseudo-first-order kinetic pattern with the rate constant (K₁) of 0.01 min^-1 with the R² values of 0.9248 and 0.9980, respectively. The results of thermodynamics demonstrated that the MG uptake was not spontaneous due to the positive value of the adsorption ΔG. In addition, the positive values of ΔS (0.079 kJ/mol K) and ΔH (30.816 kJ/mol) indicated the feasible operation and endothermic approach, respectively. Besides, the wastewater investigations showed that the nanocomposite could be used as a new promising sorbent for efficient removal of MG (R% > 72) and magnetically separable from the real samples.

Enhanced production of alkane hydroxylase from Penicillium chrysogenum SNP5 (MTCC13144) through feed-forward neural network and genetic algorithm

Alkane hydroxylase (AlkB), a membrane-bound enzyme has high industrial demand; however, its economical production remains challenging due to its intrinsic nature and co-factor dependency. In the current study, various critical process parameters for optimum production of AlkB have been optimized through feed forward neural network (FFNN) and genetic algorithm (GA) models using Penicillium chrysogenum SNP5 (MTCC13144). AlkB specific activity under preliminary un-optimized conditions i.e., 1% hexadecane, 7.4 pH, 11 days incubation time, 28 °C incubation temperature and 1 ml of inoculum size was 100 U/mg. ‘One variable at a time’ (OVAT) strategy was used to identify optimum physicochemical parameters and then its output data was fed to develop a model of FFNN with ‘6-12-1’ topology. Outputs of FFNN were further optimized through GA to minimize errors and intensify search level. This has provided superior predictive performances with 0.053 U/mg overall mean absolute percentage error (MAPE), 6.801 U/mg root mean square errors (RMSE), and 0.987 overall correlation coefficient (R). The AlkB specific activity improved by 3.5-fold, i.e., from 100 U/mg under preliminary un-optimized conditions to 351.32 U/mg under optimum physicochemical conditions obtained through FFNN-GA hybrid method, i.e., hexadecane (carbon source): 1.56% v/v, FeSO₄: 0.63 mM, incubation temperature: 27.40 °C, pH: 7.38, incubation time: 12.35 days and inoculums size: 1.33 ml. The developed process would be a stepping stone to fulfill the high industrial demands of  Alkane hydroxylase.

Phenotype analysis of cultivation processes via unsupervised machine learning: Demonstration for Clostridium pasteurianum

A novel approach of phenotype analysis of fermentation-based bioprocesses based on unsupervised learning (clustering) is presented. As a prior identification of phenotypes and conditional interrelations is desired to control fermentation performance, an automated learning method to output reference phenotypes (defined as vector of biomass-specific rates) was developed and the necessary computing process and parameters were assessed. For its demonstration, time series data of 90 Clostridium pasteurianum cultivations were used which feature a broad spectrum of solventogenic and acidogenic phenotypes, while 14 clusters of phenotypic manifestations were identified. The analysis of reference phenotypes showed distinct differences, where potential conditionalities were exemplary isolated. Further, cluster-based balancing of carbon and ATP or the use of reference phenotypes as indicator for bioprocess monitoring were demonstrated to highlight the perks of this approach. Overall, such analysis depends strongly on the quality of the data and experimental validations will be required before conclusions. However, the automated, streamlined and abstracted approach diminishes the need of individual evaluation of all noisy dataset and showed promising results, which could be transferred to strains with comparably wide-ranging phenotypic manifestations or as indicators for repeated bioprocesses with clearly defined target.

Sequential modelling for carbohydrate and bioethanol production from Chlorella saccharophila CCALA 258: a complementary experimental and theoretical approach for microalgal bioethanol production

Bioethanol production from microalgal biomass is an attractive concept, and theoretical methods by which bioenergy can be produced indicate saving in both time and efficiency. The aim of the present study was to investigate the efficiencies of carbohydrate and bioethanol production by Chlorella saccharophila CCALA 258 using experimental, semiempirical, and theoretical methods, such as response surface methods (RSMs) and an artificial neural network (ANN) through sequential modeling. In addition, the interactive response surface modeling for determining the optimum conditions for the variables was assessed. The results indicated that the maximum bioethanol concentration was 11.20 g/L using the RSM model and 11.17 g/L using the ANN model under optimum conditions of 6% (v/v %) substrate and 4% (v/v %) inoculum at 96-h fermentation, pH 6, and 40 °C. In addition, the value of the experimental data for carbohydrate concentration was 0.2510 g/g biomass at ANN with the maximums of 50% (v/v) wastewater concentration, 4% (m/m) hydrogen peroxide concentration, and 6000 U/mL enzyme activity. Finally, although the RSM model was more effective than the ANN model for predicting bioethanol concentration, the ANN model yielded more precise values than the RSM model for carbohydrate concentration.

Improving the performance of machine learning models for biotechnology: The quest for deus ex machina

Machine learning is becoming an integral part of the Design-Build-Test-Learn cycle in biotechnology. Machine learning models learn from collected datasets such as omics data and predict a defined outcome, which has led to both production improvements and predictive tools in the field. Robust prediction of the behavior of microbial cell factories and production processes not only greatly increases our understanding of the function of such systems, but also provides significant savings of development time. However, many pitfalls when modeling biological data - bad fit, noisy data, model instability, low data quantity and imbalances in the data - cause models to suffer in their performance. Here we provide an accessible, in-depth analysis on the problems created by these pitfalls, as well as means of their detection and mediation, with a focus on supervised learning. Assessing the state of the art, we show that, currently, in-depth analyses of model performance are often absent and must be improved. This review provides a toolbox for the analysis of model robustness and performance, and simultaneously proposes a standard for the community to facilitate future work. It is further accompanied by an interactive online tutorial on the discussed issues.

Optimization of Position and Number of Hotspot Detectors Using Artificial Neural Network and Genetic Algorithm to Estimate Material Levels Inside a Silo

To realize efficient operation of a silo, level management of internal storage is crucial. In this study, to address the existing measurement limitations, a silo hotspot detector, which is typically utilized for internal silo temperature monitoring, was employed. The internal temperature data measured using the hotspot detectors were used to train an artificial neural network (ANN) algorithm to predict the level of the internal storage of the silo. The prediction accuracy was evaluated by comparing the predicted data with ground truth data. We combined the ANN model with the genetic algorithm (GA) to improve the prediction accuracy and establish efficient sensor installation positions and number to proceed with optimization. Simulation results demonstrated that the best predictive performance (up to 97% accuracy) was achieved when the ANN structure was 9-19-19-1. Furthermore, the numbers of efficient sensors and sensors positions determined using the proposed ANN-GA technique were reduced from seven to five or four, thereby ensuring economic feasibility.

Machine learning for metabolic engineering: A review

Machine learning provides researchers a unique opportunity to make metabolic engineering more predictable. In this review, we offer an introduction to this discipline in terms that are relatable to metabolic engineers, as well as providing in-depth illustrative examples leveraging omics data and improving production. We also include practical advice for the practitioner in terms of data management, algorithm libraries, computational resources, and important non-technical issues. A variety of applications ranging from pathway construction and optimization, to genetic editing optimization, cell factory testing, and production scale-up are discussed. Moreover, the promising relationship between machine learning and mechanistic models is thoroughly reviewed. Finally, the future perspectives and most promising directions for this combination of disciplines are examined.

Biovalorisation of crude glycerol and xylose into xylitol by oleaginous yeast Yarrowia lipolytica

Background

Xylitol is a commercially important chemical with multiple applications in the food and pharmaceutical industries. According to the US Department of Energy, xylitol is one of the top twelve platform chemicals that can be produced from biomass. The chemical method for xylitol synthesis is however, expensive and energy intensive. In contrast, the biological route using microbial cell factories offers a potential cost-effective alternative process. The bioprocess occurs under ambient conditions and makes use of biocatalysts and biomass which can be sourced from renewable carbon originating from a variety of cheap waste feedstocks.

Result

In this study, biotransformation of xylose to xylitol was investigated using Yarrowia lipolytica, an oleaginous yeast which was firstly grown on a glycerol/glucose for screening of co-substrate, followed by media optimisation in shake flask, scale up in bioreactor and downstream processing of xylitol. A two-step medium optimization was employed using central composite design and artificial neural network coupled with genetic algorithm. The yeast amassed a concentration of 53.2 g/L xylitol using pure glycerol (PG) and xylose with a bioconversion yield of 0.97 g/g. Similar results were obtained when PG was substituted with crude glycerol (CG) from the biodiesel industry (titer: 50.5 g/L; yield: 0.92 g/g). Even when xylose from sugarcane bagasse hydrolysate was used as opposed to pure xylose, a xylitol yield of 0.54 g/g was achieved. Xylitol was successfully crystallized from PG/xylose and CG/xylose fermentation broths with a recovery of 39.5 and 35.3%, respectively.

Conclusion

To the best of the author’s knowledge, this study demonstrates for the first time the potential of using Y. lipolytica as a microbial cell factory for xylitol synthesis from inexpensive feedstocks. The results obtained are competitive with other xylitol producing organisms.

Optimizing the Rail Profile for High-Speed Railways Based on Artificial Neural Network and Genetic Algorithm Coupled Method

Though the high-speed railways are seen as a sustainable form of transportation, the fact that the rail wear in high-speed railways negatively affects the running safety and riding comfort, as well as the maintenance of railways, has drawn a wide range of concerns among researchers and scholars. In order to reduce the rail wear and achieve the goal of sustainable transportation, this paper proposes an ingenious optimization program of rail profiles based on the artificial neural network (ANN) and genetic algorithm (GA) coupled method. The candidate solutions of the nonlinear GA programming model are regarded as the inputs of the trained ANN model. Meanwhile, the outputs of the trained ANN model serve as the objective functions of the GA model. The computational results show that the optimized rail profile not only has superior performances in terms of the wheel/rail wear and contact conditions, but also maintains good dynamic performances. Therefore, this study can provide the theoretical and practical basis for the design and the preventive grinding of rails in the high-speed railways. Also, the ANN-GA coupled model can be extended and further employed on the optimization of other rail profiles.

Rapid extraction of copper ions in water, tea, milk and apple juice by solvent-terminated dispersive liquid-liquid microextraction using p-sulfonatocalix (4) arene: optimization by artificial neural networks coupled bat inspired algorithm and response surface methodology

A bat inspired algorithm with the aid of artificial neural networks (ANN-BA) has been used for the first time in chemistry and food sciences to optimize solvent-terminated dispersive liquid-liquid microextraction (ST-DLLME) as a green, fast and low cost technique for determination of Cu²⁺ ions in water and food samples using p-sulfonatocalix (4) arene as a complexing reagent. For this purpose, the influence of four important factors four factors which was influenced on the extraction efficiency such as salt addition, solution pH and disperser and extraction solvent volumes were investigated. Central composite design (CCD) as a comparative technique was employed for optimization of ST-DLLME efficiency. The ANN-BA optimization technique was regarded as a superior model due to its higher value of extraction efficiency (about 7.21%) compared to CCD method. Under ANN-BA optimal conditions, the limit of quantitation (S/N = 10), limit of detection (S/N = 3) and linear range were 0.35, 0.12 and 0.35-1000 µg L^-1, respectively. In these circumstances, the percentage recoveries for drinking tea, apple juice, milk, bottled drinking water, river and well water spiked with 0.05, 0.1 and 0.2 mg L^-1 of Cu²⁺ ions were in the acceptable range (91.4-107.1%). In comparison to other methods, the developed ST-DLLME method showed the lowest solvent and sample consumption, shortest value of extraction time, most suitable determination and detection limits and linear range with simple and low cost apparatus. Additionally, the use of bat inspired algorithm as a powerful metaheuristic algorithm with the aid of artificial networks is another advantage of the present work.