Dual-carbon-source mixed fermentation metabolic regulation system for single-cell protein production

Abstract

The application discloses a double-carbon-source mixed fermentation metabolic regulation system for single-cell protein production and belongs to the technical field of fermentation engineering.The application solves the problem that the separate utilization efficiency of methanol and methane is low in the prior art, and the metabolic pathways of microorganisms for the two carbon sources are different, so that efficient single-cell protein production is difficult to realize, through comprehensive analysis of various data related to historical parameters, key metabolic nodes that have a significant influence on the yield and quality of single-cell protein can be accurately identified, reasonable weights are allocated to the key metabolic nodes, so that the production quality of single-cell protein in fermentation engineering is optimized, and the realization of the production target is ensured; meanwhile, according to the production target of the actual fermentation engineering, the optimization cost of the key metabolic nodes is evaluated, excessive optimization of high-cost nodes is avoided, the economic benefit of single-cell protein production is improved, the cost budget of fermentation engineering is reduced, and the flexibility of fermentation engineering and the scientificity of metabolic regulation decision-making are enhanced.

Description

Technical Field

[0001] This invention relates to the field of fermentation engineering technology, specifically to a dual-carbon-source mixed fermentation metabolic regulation system for single-cell protein production. Background Technology

[0002] Single-cell protein (SCP) is an important microbial protein resource, characterized by high protein content and easy digestibility and absorption. Traditional SCP production primarily relies on fermentation of a single carbon source, such as glucose or starch, but this approach suffers from high costs and limited resources. With increasing emphasis on renewable energy and waste utilization, methanol and methane have gained attention as potential carbon sources. Methanol is a low-carbon alcohol obtainable through biomass gasification and natural gas reforming, while methane is a major component of natural gas.

[0003] However, in existing technologies, the utilization efficiency of methanol and methane alone is low, and the metabolic pathways of microorganisms for these two carbon sources are different, making it difficult to achieve efficient single-cell protein production.

[0004] Therefore, it does not meet the existing needs, so we propose a dual-carbon-source mixed fermentation metabolic regulation system for single-cell protein production. Summary of the Invention

[0005] The purpose of this invention is to provide a dual-carbon-source mixed fermentation metabolic regulation system for single-cell protein production. By comprehensively analyzing various data related to historical parameters, it can accurately identify key metabolic nodes that significantly affect the yield and quality of single-cell proteins, assign them reasonable weights, thereby optimizing the production quality of single-cell proteins in fermentation engineering, improving the economic efficiency of single-cell protein production, reducing the cost budget of fermentation engineering, and enhancing the flexibility of fermentation engineering and the scientific nature of metabolic regulation decisions, thus solving the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A dual-carbon-source mixed fermentation metabolic regulation system for single-cell protein production, the system comprising: a parameter acquisition unit and an analysis and regulation unit, the analysis and regulation unit comprising: an intelligent analysis module, a weight allocation module and a metabolic regulation module;

[0008] The parameter acquisition unit is configured to integrate online monitoring technology within the fermenter, and to acquire temperature, pH, dissolved oxygen, methanol and methane concentration parameters during the fermentation process through temperature sensors, pH sensors, dissolved oxygen sensors, and methanol and methane concentration sensors.

[0009] The intelligent analysis module is configured to train a convolutional neural network model based on historical parameters, predict flux changes of different metabolic pathways in a fermentation process based on the convolutional neural network model, identify key metabolic nodes, and generate different metabolic regulation strategies.

[0010] The weight allocation module is configured to allocate different weights to each key metabolic node according to the importance of the single-cell protein production target.

[0011] The metabolic regulation module is configured to optimize the generated metabolic regulation strategies based on the production target and the weights, to obtain a metabolic regulation strategy suitable for the current production target.

[0012] Further, the weight allocation module allocates different weights to each key metabolic node according to the importance of the single-cell protein production target, specifically:

[0013] Obtain the same production target in the historical parameters, and the fermentation parameters, single-cell protein yield data, consumption rate and conversion rate of methanol and methane, metabolite concentration, expression level of related genes in the key metabolic pathway, activity and protein expression level of related enzymes, and cost data in the fermentation process;

[0014] Standardize the various data to form a unified format data set;

[0015] Extract features from the data set by cluster analysis method, analyze the features by correlation analysis method, and evaluate the influence of each key metabolic node on the single-cell protein production target;

[0016] Based on the evaluation results, different weights are allocated to each key metabolic node; the weight allocation is based on the following factors:

[0017] Influence on yield: if a metabolic node has a significant impact on single-cell protein yield, a higher weight is allocated;

[0018] Influence on quality: if a metabolic node has a significant impact on single-cell protein quality, a higher weight is allocated;

[0019] Cost factor: if the optimization cost of a metabolic node is high, the weight is appropriately reduced;

[0020] Environmental impact: if the optimization of a metabolic node will result in more by-products, the weight is appropriately reduced.

[0021] Further, the weight allocation module comprises:

[0022] The weight definition module is configured to define the weight priority of each key metabolic node according to the importance of the production target;

[0023] a weight adjustment module configured to collect parameters in the fermentation process in real time and dynamically adjust the weights based on parameter changes.

[0024] Further, the weight definition module defines a weight priority for each key metabolic node, specifically:

[0025] Explicitly define the specific goals of single-cell protein production, and quantify the importance of each production goal according to actual production needs;

[0026] Based on the weight allocation factors, different weights are assigned to each production goal;

[0027] Based on the influence of each key metabolic node on the single-cell protein production goal, and combined with the importance of the production goal, the comprehensive weight of each key metabolic node is calculated;

[0028] The obtained comprehensive weight is output as the basis for subsequent metabolic regulation strategy generation.

[0029] Further, the intelligent analysis module comprises:

[0030] A sensitivity analysis module configured to perform sensitivity analysis on each key metabolic node to evaluate the influence of each key metabolic node on the single-cell protein production goal;

[0031] A flux analysis module configured to calculate the flux changes of different metabolic pathways based on the flux balance analysis method and the prediction results of the convolutional neural network model.

[0032] Further, the flux analysis module comprises:

[0033] A metabolic network model construction module for:

[0034] Obtaining known metabolic pathways and constructing a metabolic network model based on the metabolic pathways, and obtaining multi-dimensional constraint conditions for double-carbon source mixed fermentation metabolic regulation of single-cell protein production based on business protocols, wherein the constraint conditions include mass balance constraints, reaction irreversibility constraints, and flux boundary constraints;

[0035] Adding multi-dimensional constraint conditions in the metabolic network, and obtaining a target metabolic network model based on the addition results;

[0036] A metabolic regulation strategy formulation module for:

[0037] Determining the adjustable range of parameters under different metabolic pathways, and sequentially adjusting the parameters under different metabolic pathways based on the adjustable range of parameters;

[0038] Based on the target metabolic network model, the single variable adjustment result of each group is analyzed to obtain the flux corresponding to the single variable adjustment result of each group, and the flux under each metabolic pathway is summarized based on the parameter adjustment sequence;

[0039] The trained convolutional neural network model corrects the summary result to obtain the flux change of different metabolic pathways in the fermentation process, and determines the key metabolic nodes in different metabolic pathways based on the change trend of the flux change;

[0040] At the same time, the single cell protein production corresponding to the flux change result of different metabolic pathways is obtained, and the flux change and the single cell protein production are quantified based on the key metabolic nodes;

[0041] Based on the quantification result, a relationship function between different parameter values in each metabolic pathway and the single cell protein production is obtained, and different metabolic regulation strategies are generated based on the relationship function.

[0042] Further, the metabolic regulation module comprises:

[0043] The regulation verification module is configured to evaluate the optimization effect of the metabolic regulation strategy by comparing the fermentation results before and after the implementation of the metabolic regulation strategy, and if the regulation effect is not ideal, dynamically adjusting the metabolic regulation strategy and the fermentation parameters according to the feedback data;

[0044] The early warning feedback module is configured to real-time feedback the result to the user end for early warning prompt if the regulation effect is not ideal.

[0045] Further, the parameter acquisition unit comprises:

[0046] The parameter processing module is configured to clean and standardize the historical parameters as data samples of the convolutional neural network model;

[0047] The data storage module is configured to store the real-time collected fermentation parameters and the corresponding production target and metabolic regulation strategy to form a database.

[0048] Further, the analysis and regulation unit further comprises:

[0049] The model training module is configured to divide the data samples into a training set and a test set, which are respectively used for training and testing the convolutional neural network model, and the test passed convolutional neural network model is applied to practice;

[0050] The model optimization module is configured to periodically optimize and iterate the convolutional neural network model based on the data in the database.

[0051] Further, the metabolic regulation module comprises:

[0052] a parameter determination module, configured to obtain initial cell concentration, initial methanol concentration and initial methane concentration in the fermenter based on an online monitoring technology;

[0053] a calculation module, configured to calculate cell concentrations at t1 and t2 respectively based on the initial cell concentration α, the initial methanol concentration β and the initial methane concentration θ in the fermenter, and calculate specific production rate during single-cell protein production based on the cell concentrations at t1 and t2, the specific steps comprising:

[0054] the cell concentrations at t1 and t2 are calculated according to the following formula:

[0055]

[0056] wherein, represents the cell concentration value in the fermenter at t1; represents the cell concentration value in the fermenter at t2; α represents the initial cell concentration value in the fermenter; μ represents the production rate coefficient, and the value range is (0.1, 0.8); β represents the initial methanol concentration value in the fermenter; θ represents the initial methane concentration value in the fermenter; represents the methanol concentration value in the fermenter at t1; represents the methane concentration value in the fermenter at t1; represents the methanol concentration value in the fermenter at t2; represents the methane concentration value in the fermenter at t2;

[0057] the specific production rate during single-cell protein production is calculated according to the following formula:

[0058]

[0059] wherein, represents the specific production rate during single-cell protein production; ρ represents the error factor, and the value range is (0.02, 0.06); t2 represents the specific time value for single-cell protein production in the fermenter; t1 represents the specific time value for single-cell protein production in the fermenter, and the value is less than t2; a strategy optimization module, configured to:

[0060] compare the calculated specific production rate with a preset specific production rate;

[0061] if the specific production rate is greater than or equal to the preset specific production rate, it is determined that the value of the environmental factor of the current fermentation meets the preset condition, and the single-cell protein production process is continuously monitored until the production is completed;

[0062] Otherwise, it is determined that the environmental factor value does not satisfy the preset condition, and the environmental factor value is iteratively adjusted until the specific production rate is greater than or equal to the preset specific production rate, wherein the environmental factor value includes a temperature value, a pH value, and a dissolved oxygen value.

[0063] Compared with the prior art, the present application has the following advantages:

[0064] 1. In the present application, by comprehensively analyzing various data related to historical parameters, the key metabolic nodes that significantly affect the yield and quality of single-cell protein production can be accurately identified, and appropriate weights can be assigned to them, thereby optimizing the production quality of single-cell protein in fermentation engineering and ensuring the realization of the production target. At the same time, according to the actual production target of fermentation engineering, the optimization cost of the key metabolic nodes is evaluated to avoid over-optimizing high-cost nodes, which not only improves the economic benefit of single-cell protein production and reduces the cost budget of fermentation engineering, but also enhances the flexibility of fermentation engineering and the scientificity of metabolic regulation decision-making.

[0065] 2. By constructing a metabolic network model according to known metabolic pathways and determining the multi-dimensional constraint conditions that need to be followed in metabolic regulation, the target metabolic network model is accurately and effectively constructed according to the metabolic network model and the multi-dimensional constraint conditions. Secondly, the parameters under different metabolic pathways are adjusted by a single variable, thereby effectively determining the flux under different conditions according to the target metabolic network model. Then, the trained convolutional neural network model is used to lock the flux change. Finally, the determined flux change and the single-cell protein production under the corresponding value are quantitatively analyzed to determine the relationship function between different parameter values in each metabolic pathway and the single-cell protein production, and then the metabolic regulation strategy is formulated according to the relationship function, ensuring the accuracy and reliability of the metabolic regulation strategy, and providing a reliable guarantee for the dual-carbon source mixed fermentation metabolic regulation of single-cell protein production.

[0066] 3. By calculating the cell concentration at different times and calculating the specific production rate during single-cell protein production according to the cell concentration, the metabolic regulation effect during single-cell protein production can be evaluated according to the calculated specific production rate, thereby facilitating timely metabolic regulation of environmental factors when the preset requirements are not met, and ensuring the efficiency of single-cell protein production. BRIEF DESCRIPTION OF DRAWINGS

[0067] Figure 1 The figure is a composition diagram of the dual-carbon source mixed fermentation metabolic regulation system for single-cell protein production of the present application. DETAILED DESCRIPTION

[0068] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0069] In order to solve the technical problems in the prior art that the separate utilization efficiency of methanol and methane is low, and the metabolic pathways of microorganisms for the two carbon sources are different, it is difficult to achieve efficient single-cell protein production, please refer to Figure 1 The embodiment provides the following technical solutions:

[0070] The double-carbon-source mixed fermentation metabolic regulation system for single-cell protein production comprises a parameter acquisition unit and an analysis and regulation unit.

[0071] The parameter acquisition unit is configured to integrate online monitoring technology in a fermentation tank, acquire temperature, pH, dissolved oxygen, methanol and methane concentration parameters in a fermentation process through a temperature sensor, a pH sensor, a dissolved oxygen sensor, a methanol and methane concentration sensor, and convert analog signals acquired into digital signals; the parameter acquisition unit comprises:

[0072] The parameter processing module is configured to perform cleaning and standardization processing on historical parameters, identify and remove abnormal values, supplement missing values using an interpolation method, and convert the numerical ranges of different parameters into the same range as data samples of a convolutional neural network model.

[0073] The data storage module is configured to store real-time acquired fermentation parameters and corresponding production targets and metabolic regulation strategies to form a database, for subsequent data analysis and model training.

[0074] The analysis and regulation unit comprises a model training module, a model optimization module, an intelligent analysis module, a weight distribution module and a metabolic regulation module.

[0075] The model training module is configured to divide data samples into 80% training sets and 20% test sets, respectively, for training and testing of a convolutional neural network model, and apply the test-passed convolutional neural network model to practice; the convolutional neural network model structure is constructed, including a convolutional layer, a pooling layer, a full connection layer and the like; the training set is used to train the convolutional neural network model, the model parameters are optimized through a back propagation algorithm, and a loss function is minimized; and in the training process, the convolutional neural network model is verified using a verification set to avoid overfitting, and thus the convolutional neural network model structure and hyperparameters are adjusted according to the results of the verification set.

[0076] The model optimization module is configured to periodically optimize and iterate the convolutional neural network model based on the data in the database; the convolutional neural network model is retrained by using the latest updated fermentation parameters, the performance of the new and old models is compared, the model with better performance is selected as the new production model, and the optimized convolutional neural network model is verified to ensure that the performance of the convolutional neural network model meets the requirements.

[0077] The intelligent analysis module is configured to train the convolutional neural network model based on historical parameters, predict the flux changes of different metabolic pathways during the fermentation process based on the convolutional neural network model, identify key metabolic nodes, and generate different metabolic regulation strategies; the intelligent analysis module comprises:

[0078] The sensitivity analysis module is configured to perform sensitivity analysis on each key metabolic node to evaluate the influence of each key metabolic node on the single-cell protein production target; for example, by changing the activity of a certain key enzyme, the influence on the yield and quality of single-cell protein, such as protein content, purity, etc., is observed, and a sensitivity analysis report is generated to list the influence degree of each key metabolic node on the yield and quality of single-cell protein; the sensitivity analysis results are used as a basis for assigning weights to each key metabolic node.

[0079] The flux analysis module is configured to calculate the flux changes of different metabolic pathways based on the flux balance analysis method and the prediction results of the convolutional neural network model; according to the production target, such as maximizing single-cell protein yield, the objective function of flux balance analysis is set, and the constraint conditions of flux balance analysis are set according to historical parameters and prediction results, such as enzyme activity limitation, metabolite concentration limitation, etc.; the flux changes of different metabolic pathways are calculated using linear programming method, and a flux analysis report is generated to list the flux changes of different metabolic pathways, identify key metabolic nodes and potential bottlenecks; according to the flux analysis results, metabolic regulation suggestions are proposed, such as optimizing fermentation parameters and performing gene editing.

[0080] The above-mentioned beneficial effects: through sensitivity analysis, the influence of each key metabolic node on the yield and quality of single-cell protein can be accurately evaluated to provide a scientific basis for the development of metabolic regulation strategies; through flux analysis, key metabolic nodes and potential bottlenecks can be identified to optimize metabolic flux; the analysis results of the two are combined to provide a basis for optimizing fermentation parameters and metabolic engineering strategies, thereby improving the efficiency of fermentation engineering and the yield and quality of single-cell protein; at the same time, by optimizing key metabolic nodes, resource waste in fermentation engineering is reduced, and the production cost of fermentation engineering is reduced.

[0081] The weight distribution module is configured to assign different weights to each key metabolic node according to the importance of the single-cell protein production target, specifically:

[0082] acquire the same production target in the historical parameters, and its fermentation parameters, single-cell protein yield data, methanol and methane consumption rate and conversion rate, metabolite concentration, expression level of related genes in key metabolic pathways, activity of related enzymes and protein expression level, and cost data in the fermentation process;

[0083] standardize the various data to form a data set in a unified format; perform feature extraction on the data set by cluster analysis method, analyze the features by correlation analysis method, and evaluate the influence of each key metabolic node on the single-cell protein production target; for example: if the flux change of a certain metabolic node has a significant influence on the single-cell protein yield, and its flux is low, it means that the node may be a key bottleneck; if the flux change of a certain metabolic node has a significant influence on the quality of single-cell protein, it means that the node is crucial to product quality;

[0084] Based on the evaluation results, different weights are assigned to each key metabolic node; the weight assignment is based on the following factors: influence on yield: if a certain metabolic node has a significant influence on single-cell protein yield, a higher weight is assigned; influence on quality: if a certain metabolic node has a significant influence on the quality of single-cell protein, a higher weight is assigned; cost factor: if the optimization cost of a certain metabolic node is high, the weight is appropriately reduced; environmental impact: if the optimization of a certain metabolic node will result in more by-products, the weight is appropriately reduced.

[0085] In one embodiment, it is assumed that if the flux change of Mdh methanol dehydrogenase has a significant influence on the yield of single-cell protein, and its flux is low, a higher weight is assigned, such as 0.6; if the flux change of Pmo methane monooxygenase has a significant influence on the quality of single-cell protein, but its optimization cost is high, a lower weight is assigned, such as 0.4; through the above steps, it can be accurately judged that a certain key metabolic node has the greatest influence on the production target of single-cell protein, and a reasonable weight is assigned to each key metabolic node, thereby providing a scientific basis for optimizing the fermentation process.

[0086] The weight assignment module comprises:

[0087] The weight definition module is configured to define the weight priority of each key metabolic node according to the importance of the production target; for example: if maximizing yield is the primary goal, the weight of the metabolic node directly related to yield (such as methanol dehydrogenase Mdh) can be set higher, specifically:

[0088] Clearly the specific goals of single cell protein production, according to the actual production needs, the importance of each production target is quantified; production goals include but are not limited to: maximize production: improve the yield of single cell protein; quality optimization: improve the purity of single cell protein or the content of specific amino acids; cost control: reduce production costs, including carbon source, energy and time cost; environmentally friendly: reduce by-product generation,

[0089] reduce environmental impact;

[0090] Based on the weight allocation factors, different weights are given to each production target; for example: maximize production: weight is 0.4; quality optimization: weight is 0.3; cost control: weight is 0.2; environmentally friendly: weight is 0.1;

[0091] Based on the influence of each key metabolic node on the production target of single cell protein, combined with the importance of production target, the comprehensive weight of each key metabolic node is calculated; specifically, through metabolic flux analysis, sensitivity analysis and experimental verification, the key metabolic nodes that have significant influence on the production target of single cell protein are identified; for example: Mdh methanol dehydrogenase, Pmo methane monooxygenase, Pyc pyruvate carboxylase and Adh ethanol dehydrogenase; through sensitivity analysis, the following key metabolic nodes have the following influence on the production target:

[0092] Mdh: the most influential on yield, weight is 0.6;

[0093] Pmo: has a greater impact on quality, weight is 0.4;

[0094] Pyc: has a greater impact on cost, weight is 0.3;

[0095] Adh: has a greater impact on the environment, weight is 0.2;

[0096] Considering the importance of production target, the final weight distribution is as follows:

[0097] Mdh: 0.4x0.6=0.24;

[0098] Pmo: 0.3x0.4=0.12;

[0099] Pyc: 0.2x0.3=0.06;

[0100] Adh: 0.1x0.2=0.02;

[0101] The comprehensive weight obtained is output as the basis for subsequent metabolic regulation strategy generation.

[0102] A weight adjustment module is configured to collect various parameters in the fermentation process in real time and dynamically adjust the weights based on the parameter changes. For example, if it is found that the flux change of a certain metabolic node has a greater impact on the yield, but the cost is higher, the weight of the metabolic node can be appropriately reduced to balance the yield and cost.

[0103] The above-mentioned beneficial effects: through comprehensive analysis of various data related to historical parameters, the key metabolic nodes that have a significant impact on the yield and quality of single-cell protein can be accurately identified, and appropriate weights are assigned to them, so as to optimize the production quality of single-cell protein in fermentation engineering and ensure the realization of the production target; at the same time, according to the production target of the actual fermentation engineering, the optimization cost of the key metabolic node is evaluated, and over-optimization of high-cost nodes is avoided, which not only improves the economic benefit of single-cell protein production and reduces the cost budget of fermentation engineering, but also enhances the flexibility of fermentation engineering and the scientificity of metabolic regulation decision-making.

[0104] A metabolic regulation module is configured to optimize the generated metabolic regulation strategy based on the production target and the weight, to obtain a metabolic regulation strategy suitable for the current production target. Specifically, the adjustment method includes: fermentation parameter adjustment: adjusting the fermentation parameters such as temperature, pH value, aeration rate, stirring speed, etc. to optimize the activity of the key metabolic node. Metabolic engineering modification: metabolic engineering modification is performed on the key metabolic node, such as gene editing, enzyme activity optimization, etc. to improve the metabolic efficiency. Weight adjustment: according to the real-time monitoring data, the weight of the key metabolic node is adjusted to better reflect the importance of the current production target.

[0105] In one embodiment, it is assumed that the production target includes: yield maximization (weight 0.4), quality optimization (weight 0.3), cost control (weight 0.2), and environmental friendliness (weight 0.1); the key metabolic nodes include: Mdh methanol dehydrogenase, Pmo methane monooxygenase, Pyc pyruvate carboxylase, and Adh ethanol dehydrogenase; the initial metabolic regulation strategy is applied to the fermentation process, and the fermentation data before and after the implementation are recorded; the fermentation data, including single-cell protein yield, biomass, metabolite concentration, fermentation parameters, etc. are collected; by comparing the data before and after the implementation, the effect of the regulation strategy is evaluated; for example: calculating the indicators of yield improvement, quality improvement, cost reduction, etc.; according to the comprehensive performance indicators, the effect of the regulation strategy is evaluated; if the regulation effect is not ideal, the next step is entered; according to the feedback data, the fermentation parameters and metabolic engineering modification scheme are adjusted; for example: adjusting the fermentation temperature, pH value, aeration rate, etc. or performing gene editing or enzyme activity optimization on the key metabolic node, until a metabolic regulation strategy suitable for the current production target is obtained, so as to improve the efficiency and quality of single-cell protein production and reduce the cost of fermentation engineering.

[0106] The metabolic regulation module includes:

[0107] The regulation verification module is configured to evaluate the optimization effect of the metabolic regulation strategy by comparing the fermentation results before and after the implementation of the metabolic regulation strategy; if the regulation effect is not ideal, the metabolic regulation strategy and the fermentation parameters are dynamically adjusted according to the feedback data to ensure that the fermentation process meets the optimization target; specifically, the generated metabolic regulation strategy is applied to the fermentation process, and the fermentation parameters and production results before and after the implementation are recorded, including single-cell protein yield, biomass, metabolite concentration, etc.; the data before and after the implementation are compared to evaluate the effect of the regulation strategy, and the main indicators include yield improvement, quality improvement, cost reduction, etc. If the regulation effect is not ideal, the regulation strategy and the fermentation parameters are adjusted according to the feedback data; for example, the fermentation temperature, pH value, aeration rate and other parameters are adjusted, or the key metabolic nodes are further modified by metabolic engineering.

[0108] The early warning feedback module is configured to real-time feedback the result to the user end for early warning prompt when the regulation effect is not ideal; specifically, the early warning conditions are set, such as the regulation effect not reaching the expected target, specifically, the yield improvement being less than 10%; when the regulation effect is not ideal, the early warning information is real-time feedback through the user end interface or notification system, such as short message, email, etc.; the user takes measures in time according to the early warning information, such as adjusting the fermentation operation or redesigning the regulation strategy.

[0109] The above content achieves the beneficial effects: by dynamically adjusting the metabolic regulation strategy and the fermentation parameters, it is ensured that the fermentation process always meets the optimization target, and the production efficiency and product quality of the fermentation engineering are improved; combined with real-time monitoring and early warning mechanism, the situation of ideal regulation effect can be found and adjusted in time, and the production risk of the fermentation engineering is reduced; at the same time, according to different production targets and real-time data, the metabolic regulation strategy is flexibly adjusted, thereby enhancing the adaptability and flexibility of the system.

[0110] Working principle: by real-time monitoring the key parameters in the fermentation process, different weights are given to each production target according to the specific target of single-cell protein production; combined with the influence of each key metabolic node on the production target, the comprehensive weight of each key metabolic node is calculated; based on the production target and the weight, the generated metabolic regulation strategy is optimized to obtain a metabolic regulation strategy suitable for the current production target; thereby the dynamic optimization of the fermentation engineering can be realized, the production efficiency of the fermentation engineering is improved, the product quality of the single-cell protein is improved, the cost of the fermentation engineering is reduced, and the environmental impact is reduced.

[0111] The embodiment provides a double-carbon-source mixed fermentation metabolic regulation system for single-cell protein production, and the flux analysis module comprises:

[0112] The metabolic network model construction module is used for:

[0113] acquire a known metabolic pathway, and construct a metabolic network model based on the metabolic pathway, and acquire multi-dimensional constraint conditions in a double-carbon source mixed fermentation metabolic regulation process of single-cell protein production based on a business protocol, wherein the constraint conditions include mass balance constraints, reaction irreversibility constraints, and flux boundary constraints;

[0114] add the multi-dimensional constraint conditions in the metabolic network, and obtain a target metabolic network model based on an addition result;

[0115] a metabolic regulation strategy formulation module, configured to:

[0116] determine a parameter adjustable range under different metabolic pathways, and sequentially perform single-variable adjustment on parameters under different metabolic pathways based on the parameter adjustable range;

[0117] analyze each set of single-variable adjustment results based on the target metabolic network model, obtain fluxes corresponding to each set of single-variable adjustment results, and aggregate fluxes under each metabolic pathway based on a parameter adjustment sequence;

[0118] correct the aggregation results by using a trained convolutional neural network model, obtain flux changes of different metabolic pathways in the fermentation process, and determine key metabolic nodes under different metabolic pathways based on a change trend of the flux changes;

[0119] Meanwhile, acquire single-cell protein production amounts corresponding to the flux change results of different metabolic pathways, and quantify the flux changes and the single-cell protein production amounts based on the key metabolic nodes;

[0120] obtain a relationship function between different parameter values and the single-cell protein production amounts in each metabolic pathway based on the quantification results, and generate different metabolic regulation strategies based on the relationship function.

[0121] In this embodiment, the metabolic network model is constructed according to the metabolic pathway, and is used to analyze the metabolism in the double-carbon source mixed fermentation metabolic regulation process of single-cell protein production, and specifically includes determining flux changes under each metabolic pathway.

[0122] In this embodiment, the business protocol is known in advance, and is used to represent standards or parameter requirements that need to be followed in the double-carbon source mixed fermentation metabolic regulation of single-cell protein production.

[0123] In this embodiment, the target metabolic network model refers to a result obtained after adding the multi-dimensional constraint conditions in the metabolic network, that is, a tool that can directly analyze the metabolic process.

[0124] In this embodiment, the parameter adjustable range refers to a parameter adjustment interval corresponding to different metabolic pathways, for example, a temperature adjustment interval and a pH value adjustment interval.

[0125] In this embodiment, single variable adjustment refers to adjusting the parameter value of each one of the metabolic pathways, so as to determine the flux of different metabolic pathways under different conditions.

[0126] In this embodiment, the single cell protein production refers to the specific amount of single cell protein production under different flux values.

[0127] In this embodiment, quantifying the flux change and the single cell protein production refers to determining the specific quantitative relationship between the flux and the single cell protein production, i.e., determining the correlation between the unit flux change and the unit single cell protein production, i.e., obtaining the relationship function.

[0128] In this embodiment, generating different metabolic regulation strategies based on the relationship function is determined according to the obtained relationship function, i.e., the flux can be determined according to the single cell protein production, and then the specific parameter value under different metabolic pathways can be determined, so as to determine the metabolic regulation strategy.

[0129] The working principle and beneficial effects of the above technical solution are: by constructing a metabolic network model according to the known metabolic pathways and determining the multi-dimensional constraint conditions that need to be followed in the metabolic regulation process, the target metabolic network model is accurately and effectively constructed according to the metabolic network model and the multi-dimensional constraint conditions. Secondly, the parameters under different metabolic pathways are adjusted by single variable, so as to effectively determine the flux under different conditions according to the target metabolic network model, and then realize the locking of the flux change according to the trained convolutional neural network model. Finally, the determined flux change and the single cell protein production under the corresponding value are quantitatively analyzed, the relationship function between different parameter values in each metabolic pathway and the single cell protein production is determined, and then the metabolic regulation strategy is formulated according to the relationship function, which ensures the accuracy and reliability of the metabolic regulation strategy, and provides a reliable guarantee for the double-carbon source mixed fermentation metabolic regulation of single cell protein production.

[0130] The present embodiment provides a double-carbon source mixed fermentation metabolic regulation system for single cell protein production, the metabolic regulation module comprises:

[0131] The parameter determination module is used for obtaining the initial cell concentration, the initial methanol concentration and the initial methane concentration in the fermenter based on online monitoring technology.

[0132] The computing module is configured to calculate the cell concentration at the time t1 and the time t2 based on the initial cell concentration α and the initial methanol concentration β and the initial methane concentration θ in the fermenter, and calculate the specific production rate during the single-cell protein production based on the cell concentration at the time t1 and the time t2, and the specific steps include:

[0133] The cell concentration at the time t1 and the time t2 is calculated according to the following formula:

[0134]

[0135]

[0136] wherein, represents the cell concentration value in the fermenter at the time t1; represents the cell concentration value in the fermenter at the time t2; α represents the initial cell concentration value in the fermenter; μ represents the production rate coefficient, and the value range is (0.1, 0.8); β represents the initial methanol concentration value in the fermenter; θ represents the initial methane concentration value in the fermenter; represents the methanol concentration value in the fermenter at the time t1; represents the methane concentration value in the fermenter at the time t1; represents the methanol concentration value in the fermenter at the time t2; represents the alkane concentration value in the fermenter at the time t2;

[0137] The specific production rate during the single-cell protein production is calculated according to the following formula:

[0138]

[0139] wherein, represents the specific production rate during the single-cell protein production; ρ represents the error factor, and the value range is (0.02, 0.06); t2 represents the specific time value for the single-cell protein production in the fermenter; t1 represents the specific time value for the single-cell protein production in the fermenter, and the value is less than t2; the strategy optimization module is configured to:

[0140] compare the calculated specific production rate with the preset specific production rate;

[0141] if the specific production rate is greater than or equal to the preset specific production rate, it is determined that the environmental factor value of the current fermentation meets the preset condition, and the single-cell protein production process is continuously monitored until the production is completed;

[0142] Otherwise, it is determined that the environment factor value does not satisfy the preset condition, and the environment factor value is iteratively adjusted until the specific production rate is greater than or equal to the preset specific production rate, wherein the environment factor value includes a temperature value, a pH value and a dissolved oxygen value.

[0143] In this embodiment, the initial cell concentration refers to the cell concentration value contained in the production environment provided for the production of single-cell protein in the fermenter, i.e., the value when the substrate concentration has not yet participated in the metabolic process.

[0144] In this embodiment, the initial methanol concentration and the initial methane concentration refer to the methanol concentration and the methane concentration added in the fermenter before metabolic regulation.

[0145] In this embodiment, the specific production rate is a measure of the production rate of single-cell protein under certain conditions.

[0146] In this embodiment, the preset specific production rate is set in advance and is a reference basis for measuring whether the specific production rate for the production of single-cell protein is qualified, and can be adjusted.

[0147] The working principle and beneficial effects of the above technical solution are: by calculating the cell concentration at different times, and calculating the specific production rate of single-cell protein production according to the cell concentration, the metabolic regulation effect of single-cell protein production can be evaluated according to the calculated specific production rate, so that the metabolic regulation of the environment factor can be timely performed when the preset requirement is not met, and the efficiency of single-cell protein production is ensured.

[0148] It should be noted that, in this document, relational terms such as first and second and the like can only be used to distinguish one entity or action from another entity or action, without necessarily requiring or implying any such actual relationship or order between such entities or actions. Moreover, the terms "comprising", "having", or any other variant thereof are intended to cover non-exclusive inclusions, such that processes, methods, articles, or apparatuses including a series of elements not only include those elements, but also include other elements not explicitly listed or inherent to such processes, methods, articles, or apparatuses.

[0149] Although embodiments of the present application have been shown and described, it will be understood by those skilled in the art that changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the application, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A dual carbon source mixed fermentation metabolic regulation system for single cell protein production, characterized in that, The system comprises a parameter acquisition unit and an analysis and regulation unit, the analysis and regulation unit comprising an intelligent analysis module, a weight allocation module and a metabolic regulation module; The parameter acquisition unit is configured to integrate online monitoring technology in the fermenter, and to acquire temperature, pH, dissolved oxygen, methanol and methane concentration parameters in the fermentation process through temperature sensors, pH sensors, dissolved oxygen sensors, methanol and methane concentration sensors; The intelligent analysis module is configured to train a convolutional neural network model based on historical parameters, to predict flux changes in different metabolic pathways during fermentation based on the convolutional neural network model, to identify key metabolic nodes, and to generate different metabolic regulation strategies; the intelligent analysis module comprises: A sensitivity analysis module configured to perform sensitivity analysis on each key metabolic node to evaluate the influence of each key metabolic node on the single-cell protein production target; A flux analysis module configured to calculate flux changes in different metabolic pathways based on the flux balance analysis method combined with the prediction results of the convolutional neural network model; the flux analysis module comprises: A metabolic network model construction module for obtaining known metabolic pathways and constructing a metabolic network model based on the metabolic pathways, and obtaining multi-dimensional constraint conditions for double-carbon source mixed fermentation metabolic regulation of single-cell protein production based on a business protocol, wherein the constraint conditions include mass balance constraints, reaction irreversibility constraints and flux boundary constraints; adding the multi-dimensional constraint conditions in the metabolic network, and obtaining a target metabolic network model based on the addition results; A metabolic regulation strategy formulation module for determining the parameter adjustable range under different metabolic pathways, and sequentially adjusting the parameters under different metabolic pathways based on the parameter adjustable range; analyzing each set of single variable adjustment results based on the target metabolic network model to obtain the flux corresponding to each set of single variable adjustment results, and summarizing the flux under each metabolic pathway based on the parameter adjustment sequence; the trained convolutional neural network model corrects the summary results to obtain the flux changes in different metabolic pathways during fermentation, and determines the key metabolic nodes under different metabolic pathways based on the change trend of the flux changes; At the same time, the single-cell protein production amount corresponding to the flux change results of different metabolic pathways is obtained, and the flux change and the single-cell protein production amount are quantified based on the key metabolic nodes; a relationship function between different parameter values and the single-cell protein production amount in each metabolic pathway is obtained based on the quantification results, and different metabolic regulation strategies are generated based on the relationship function; The weight allocation module is configured to allocate different weights to each key metabolic node according to the importance of the single-cell protein production target; The metabolic regulation module is configured to optimize the generated metabolic regulation strategies based on the production target and the weight to obtain a metabolic regulation strategy suitable for the current production target; the metabolic regulation module comprises: A parameter determination module for obtaining initial cell concentration, initial methanol concentration and initial methane concentration in the fermenter based on online monitoring technology; The calculation module is used to calculate the initial cell concentration in the fermenter. and initial methanol concentration and initial methane concentration Calculate separately Time and The bacterial concentration at a given time, and based on Time and The specific steps for calculating the specific production rate of single-cell protein production at a given time cell concentration include: The cell concentration at time t is calculated according to the following equation The time t at which the cell concentration is determined The cell concentration at time t is calculated according to the following equation ； ； in, express The cell concentration value in the fermenter at that time; express The cell concentration value in the fermenter at that time; This represents the cell concentration value at the initial moment in the fermenter. This represents the production yield coefficient, and its value ranges from (0.1 to 0.8). This indicates the initial methanol concentration in the fermenter; This indicates the initial methane concentration in the fermenter; Indicates in The methanol concentration in the fermenter at that time; Indicates in The methane concentration in the fermenter at that time; Indicates in The methanol concentration in the fermenter at that time; Indicates in The alkyl concentration value in the fermenter at that time; The specific production rate during single-cell protein production is calculated according to the following formula: ； wherein, represents the specific production rate during the production of single cell protein; represents an error factor, and has a value range of (0.02, 0.06); represents a specific time value during the production of single cell protein in the fermenter; represents a specific time value during the production of single cell protein in the fermenter, and has a value less than ; The strategy optimization module is configured to compare the calculated specific production rate with a preset specific production rate; if the specific production rate is greater than or equal to the preset specific production rate, it is determined that the value of the environmental factor of the current fermentation satisfies the preset condition, and the single-cell protein production process is continuously monitored until the production is completed; otherwise, it is determined that the value of the environmental factor does not satisfy the preset condition, and the value of the environmental factor is iteratively adjusted until the specific production rate is greater than or equal to the preset specific production rate, wherein the value of the environmental factor includes a temperature value, a pH value, and a dissolved oxygen value.

2. The dual carbon source mixed fermentation metabolic regulation system for single-cell protein production according to claim 1, characterized in that: The weight distribution module assigns different weights to each key metabolic node according to the importance of the single-cell protein production target, specifically: Obtain the same production target in the historical parameters, and the fermentation parameters, single-cell protein yield data, consumption rate and conversion rate of methanol and methane, metabolite concentration, expression level of related genes in the key metabolic pathway, activity of related enzymes and protein expression level, and cost data in the fermentation process; Standardize a plurality of data to form a data set in a unified format; Feature extraction is performed on the data set by cluster analysis method, and the features are analyzed by correlation analysis method to evaluate the influence of each key metabolic node on the single-cell protein production target; Based on the evaluation result, different weights are assigned to each key metabolic node; the weight distribution is based on the following factors: Influence on yield: if a metabolic node has a significant influence on single-cell protein yield, a higher weight is assigned; Influence on quality: if a metabolic node has a significant influence on single-cell protein quality, a higher weight is assigned; Cost factor: if the optimization cost of a metabolic node is high, the weight is appropriately reduced; Environmental impact: if the optimization of a metabolic node leads to the generation of more by-products, the weight is appropriately reduced.

3. The dual-carbon source mixed fermentation metabolic regulation system for single-cell protein production according to claim 2, characterized in that: The weight distribution module includes: The weight definition module is configured to define the weight priority of each key metabolic node according to the importance of the production target; The weight adjustment module is configured to collect parameters in the fermentation process in real time and dynamically adjust the weight based on the parameter changes.

4. The dual-carbon source mixed fermentation metabolic regulation system for single-cell protein production according to claim 3, characterized in that: The weight definition module defines the weight priority of each key metabolic node, specifically: Quantify the importance of each production target according to actual production needs; Based on the weight distribution factors, different weights are assigned to each production target; Based on the influence of each key metabolic node on the single-cell protein production target, and combined with the importance of the production target, the comprehensive weight of each key metabolic node is calculated; The obtained comprehensive weight is output as the basis for generating subsequent metabolic regulation strategies.

5. The dual carbon source mixed fermentation metabolic regulation system for single-cell protein production according to claim 1, wherein: The metabolic regulation module includes: The regulation verification module is configured to evaluate the optimization effect of the metabolic regulation strategy by comparing the fermentation results before and after the implementation of the metabolic regulation strategy; if the regulation effect is not ideal, the metabolic regulation strategy and the fermentation parameters are dynamically adjusted according to the feedback data; The early warning feedback module is configured to real-time feedback the result to the user end for early warning prompt if the regulation effect is not ideal.

6. The dual carbon source mixed fermentation metabolic regulation system for single-cell protein production according to claim 1, wherein: The parameter acquisition unit includes: The parameter processing module is configured to clean and standardize the historical parameters as data samples of the convolutional neural network model. The data storage module is configured to store the real-time collected fermentation parameters, the corresponding production targets and metabolic regulation strategies to form a database.

7. The dual-carbon source mixed-fermentation metabolic regulation system for single-cell protein production according to claim 1, wherein: The analysis and regulation unit further comprises: The model training module is configured to divide the data samples into a training set and a test set, which are respectively used for training and testing the convolutional neural network model, and to apply the test-passed convolutional neural network model to practice. The model optimization module is configured to periodically optimize and iterate the convolutional neural network model based on the data in the database.

Images