Method for predicting and controlling microbial flora in koji-making process and its application

Abstract

The present invention discloses a method for predicting and controlling microbial flora in the process of making Daqu and its application, belonging to the technical field of bioengineering. First, the core microbial community in the process of natural fermentation of medium-temperature Daqu is determined, and by constructing the interaction network between the core microorganisms and the environmental parameters of the fermentation process, a model of the relationship between the environmental parameters of the fermentation process and the change in microbial content is established using correlation, and the accuracy of the model is verified. Based on the Matlab curve fitting program, a biomass change model is constructed to regulate the content of core microorganisms in the fermentation process, thereby achieving the regulation of stability and the improvement of the quality of fermented foods.

Description

Technical Field

[0001] The present invention relates to a method for predicting and controlling the microbial flora in the koji-making process and its application, especially a method for predicting and controlling the microbial flora in the koji-making process based on a microbial biomass change model, belonging to the field of bioengineering technology. Background Art

[0002] Koji is the backbone of liquor. As the saccharifying and fermenting agent of Chinese liquor, it provides microorganisms, enzyme systems, nutrients and flavor substances for the liquor fermentation process. Approximately 70% of Chinese liquor and most vinegar brewing use koji as the fermenting agent. The fermentation process of koji is a natural fermentation process. Among them, Daqu is mainly divided into medium-temperature Daqu, medium-high temperature Daqu and high-temperature Daqu. Different types of koji are applied to the brewing of different flavor types of Chinese liquor. "Good koji makes good liquor", but the only factors that can be artificially controlled in the koji-making process are the raw material ratio and mixing moisture at the initial stage, and the whole process cannot be controlled. At the same time, the fermentation process of koji is a stacking fermentation process. Therefore, there are large differences in different fermentation chambers and different positions in the same fermentation chamber. It is a common phenomenon that the quality of koji in the same batch is uneven.

[0003] The koji-making cycle of koji is relatively long, usually one month or more, and it has the characteristics of a long production cycle, many influencing factors, mixed bacteria fermentation, and complex and diverse flavor substances. At present, in the research on koji, usually one or several functional microorganisms are added at the beginning of fermentation to increase or decrease the total amount of microorganisms or flavor substances / precursor substances in the fermentation process. However, the adjustment of the initial microbial structure still cannot enable us to accurately understand and control the changes of microorganisms and metabolites in the fermentation process. It only artificially changes the fermentation end point, so its practical value is relatively low.

[0004] At present, there is no model on the relationship between fermentation environment parameters or fermentation physical and chemical indexes and microorganisms, and it is impossible to predict and effectively control microorganisms through fermentation environment parameters or fermentation physical and chemical indexes. Therefore, it is very important to establish a method for accurately predicting and regulating microorganisms based on fermentation environment parameters or fermentation physical and chemical indexes, which is helpful for the better growth and metabolism of functional microorganisms and the improvement of koji quality. Summary of the Invention

[0005] [Technical Problem]

[0006] The technical problem to be solved by the present invention is that there are problems of unstable microbial growth and unstable production of key metabolites in the traditional fermentation process, but the change characteristics of microorganisms and metabolites in the fermentation process and their key influencing fermentation environment parameters or fermentation physical and chemical indexes are not clear, and the microorganisms and the fermentation process cannot be effectively controlled. For example, the change situation of microorganisms in the koji preparation process is not clear, which is not conducive to improving and stabilizing the quality of koji.

[0007] [Technical Solution]

[0008] The present invention provides a method for predicting and controlling the microbial flora in the koji-making process. This method first determines the core microbiota during the natural fermentation process of koji, constructs an interaction network between the core microbiota and the environmental parameters of the fermentation process (temperature in the fermentation chamber, air humidity in the fermentation chamber), establishes a model for the relationship between the environmental parameters of the fermentation process and the change in microbial content using correlation, verifies the accuracy of the model, and constructs a biomass change model based on the Matlab curve fitting program to regulate the content of the core microbiota in the fermentation process, thereby achieving the regulation of stability and the improvement of the quality of fermented foods.

[0009] The method for predicting and controlling the content of the microbial flora in koji-making includes the following steps:

[0010] (1) Analyze the microbial structure during the koji-making process based on high-throughput sequencing. At the same time, perform absolute quantification of the microbial flora during the fermentation process, and determine the core microbiota according to the microbial content and its functional contribution to the koji-making process. The core microbiota is used as the object of regulation during the koji-making process;

[0011] (2) Use the structural equation model to establish an interaction network among the fermentation environmental parameters, fermentation physical and chemical indexes, and microbial content. Determine the indexes for controlling the content of the microbial flora according to the significance of the interaction. After completing the construction of the structural equation model, perform model fitting calculations, and extract R 2 , path coefficients, significance, and model fitting degree (CMIN / DF, AIC, P value, etc.) in the result column. The value of CMIN / DF is preferably less than 3, and a value less than 5 is considered good model fitting. The P value is required to be less than 0.05;

[0012] (3) Conduct a normalization test on the data of the microbial content significantly controlled by the fermentation environmental parameters. After judging the normality of the data, use a univariate regression test to judge the correlation between the temperature in the fermentation chamber, the air humidity in the fermentation chamber, and the biomass of different genera of microorganisms. The fermentation environmental parameter factor with the strongest correlation (r > 0.6, P < 0.05) is used as the independent variable for predicting the biomass;

[0013] (4) Construct a biomass change model based on the Matlab curve fitting program

[0014] For data conforming to the normal distribution law, a Guassian model is used for model construction, and for data with other distribution laws, a linear model is used for model construction; among them,

[0015] The equation of the Gaussian model is: y = a × exp(-((x - b) / c) 2 ), where y is the microbial content, a, b, and c are constants, and x is the fermentation environmental parameter;

[0016] The equation of the unary linear regression model is: y = ax + b, where y is the microbial content, a and b are constants, and x is the fermentation environment parameter;

[0017] (5) During the fermentation process, the fermentation environment parameters are detected, the content of the core microorganisms is predicted according to the model established in step (4), and the content of the core microorganisms is adjusted to a content beneficial to koji making by regulating the fermentation environment parameters; The control is usually carried out at the beginning of fermentation and at different fermentation stages.

[0018] In the method for predicting and controlling the microbial flora in Daqu, the fermentation environment parameter refers to the temperature in the fermentation chamber and the air humidity in the fermentation chamber.

[0019] In the method for predicting and controlling the microbial content in Daqu, the data used for modeling is statistically tested.

[0020] In the method for predicting and controlling the microbial content in Daqu, the accuracy of the constructed model is also verified, and the parameters of the equation are adjusted according to the verification results.

[0021] In the method for predicting and controlling the microbial content in Daqu, the regulation of the fermentation environment parameters is based on non-contact system control.

[0022] The core microorganisms that can be controlled by the prediction model include Saccharomycopsis, Pichia, Wickerhamomyces, Saccharomyces, Lactobacillus, Bacillus, Kroppenstedtia, Aspergillus, and Rhizopus.

[0023] The prediction model is used for the indoor solid-state fermentation system, which can be either a natural fermentation system or under artificial control.

[0024] In the method for predicting and controlling the microbial content in Daqu, the Daqu can also be replaced by the traditional food fermentation process or the koji produced by the indoor fermentation method. The traditional food fermentation process is a traditional fermented food fermented in a fermentation chamber, and the koji produced by the indoor fermentation method is the koji for liquor production, the Daqu for vinegar production, and the koji for other fermented food production.

[0025] [Beneficial effects]

[0026] The present invention fully tracks the koji making process, samples in real time, quantitatively analyzes the microorganisms in the collected samples, examines the dispersion degree between the actual data and the model data, constructs a model, further evaluates the accuracy of the model and the control effectiveness, and predicts and controls the microbial flora in the fermentation process by controlling the fermentation environment parameters or the fermentation physical and chemical indexes. Description of the drawings

[0027] Figure 1 The action network of the fermentation environment parameters, physical and chemical indexes and microbial content in the Daqu fermentation process

[0028] Figure 2 Construction of the change model of the air humidity in the fermentation chamber and the genus Wickerhamomyces

[0029] Figure 3 Construction of the change model of the air humidity in the fermentation chamber and the genus Pichia

[0030] Figure 4 Construction of the change model of the air humidity in the fermentation chamber and the genus Saccharomycopsis

[0031] Figure 5 Construction of the change model of the fermentation acidity and the genus Lactobacillus

[0032] Figure 6 Construction of the change model of the temperature in the fermentation chamber and the genus Saccharomyces

[0033] Figure 7 Verification of the yeast genus content model in the fermentation process model and the distribution of its confidence interval Detailed implementation manners

[0034] Example 1: Analyzing the key driving factors of core microorganisms by structural equation model

[0035] (1) Based on high-throughput sequencing, analyze the microbial structure in the koji-making process. At the same time, perform absolute quantification on the microbial flora in the fermentation process, determine the core microorganisms according to the microbial content and their functional contributions to the koji-making process, and take the core microorganisms as the regulatory objects in the koji-making process;

[0036] In this example, the genus Saccharomycopsis, Pichia, Wickerhamomyces, Lactobacillus, Bacillus, Kroppenstedtia, Saccharomyces cerevisiae, Rhizopus, and Aspergillus are selected as core microorganisms.

[0037] (2) Use the structural equation model to establish the interaction network among the fermentation environment parameters, fermentation physical and chemical indexes, and microbial content, and determine the indexes for controlling the microbial flora content according to the interaction significance; after completing the construction of the structural equation model, perform model fitting calculation, and extract R 2 , path coefficients, significance, and model fitting degree (CMIN / DF, AIC, P value, etc.) in the result column. The value of CMIN / DF is preferably lower than 3, and a value below 5 is considered good model fitting. The P value is required to be less than 0.05;

[0038] As Figure 1 shown, after constructing the structural equation model, it is found that the air humidity and temperature in the fermentation chamber among the fermentation environment parameters indirectly control the microbial content through the fermentation physical and chemical indexes. Among them, the air humidity in the fermentation chamber controls the biomass of the genus Saccharomycopsis, Pichia, Wickerhamomyces, Aspergillus, and Rhizopus, the temperature in the fermentation chamber controls the biomass of the genus Lactobacillus and Saccharomyces, and the fermentation acidity controls the genus Lactobacillus.

[0039] (3) For the data of each microbial content significantly controlled by fermentation environment parameters, a normality test is performed. The normality test algorithm can adopt the one-sample K test, and modeling calculations based on the Gaussian model are performed on the data that conforms to the normal distribution (P < 0.05).

[0040] Table 1 Normality test of microbial data

[0041]

[0042] a. The test distribution is a normal distribution

[0043] b. Calculate according to the data

[0044] c. Riley's significance correction

[0045] Example 2: Model data processing and model selection

[0046] Through the Deive test, it is found that the fermentation environment parameters and the microorganisms they control show a normal distribution. Therefore, the Gaussian model is used to predict and model the microbial content.

[0047] (1) Construct a biomass change model based on the Matlab curve fitting program

[0048] The equation of the Gaussian model is: y = a × exp(-((x - b) / c) 2 ), where y is the microbial content, a, b, and c are constants, and x is the fermentation environment parameter.

[0049] (1)-1 Construction of the Gaussian model of the biomass of the genus Wickerhamomyces in Daqu based on the change of air humidity in the fermentation chamber

[0050] The biomass of the genus Wickerhamomyces shows a normal distribution relationship with the change of fermentation environment parameters. Therefore, the Gaussian equation is used to construct the model. Among them, a = 3.342×10 5 , b = 0.15, c = 0.94. It can be seen from the model that the biomass of Wickerhamomyces increases with the increase of air humidity in the fermentation chamber, and reaches the highest content of 3.34×10 5 copies / g when the air humidity in the fermentation chamber is 67.8%. After calculation by the unary linear regression equation, the P value is less than 0.001, and the model accuracy is relatively high.

[0051] (2)-2 Construction of the Gaussian model of the biomass of the genus Pichia in medium-temperature Daqu based on the change of air humidity in the fermentation chamber

[0052] The biomass of the genus Pichia shows a normal distribution relationship with the change of fermentation environment parameters. Therefore, the Gaussian equation is used to construct the model. Among them, a = 1.96×105 , b = 0.89, c = 1.44. It can be seen from the model that the air humidity in the fermentation chamber of 74.8% - 76.0% is most suitable for the growth of Pichia, and the biomass of Pichia is 1.96×10 5 copies / g when the air humidity in the fermentation chamber is 75.0%. Calculated by the linear regression equation, the P value is less than 0.001, and the model has high accuracy.

[0053] (2)-3 Construction of the Gaussian model of the biomass of Saccharomycopsis in medium-temperature Daqu based on the change of air humidity in the fermentation chamber

[0054] The biomass of Saccharomycopsis shows a normal distribution relationship with the change of fermentation environment parameters, so the Gaussian equation is used to construct the model. Among them, a = 6.02×10 6 , b = -0.15, c = 0.46. It can be seen from the model that the air humidity in the fermentation chamber of 67.0% - 70.0% is most suitable for the growth of Saccharomycopsis, and the biomass of Saccharomycopsis is 6.02×10 6 copies / g when the air humidity in the fermentation chamber is 68.7%. Calculated by the linear regression equation, the P value is less than 0.001, and the model has high accuracy.

[0055] (2)-4 Construction of the Gaussian model of the biomass of Saccharomyces in medium-temperature Daqu based on the change of temperature in the fermentation chamber

[0056] The biomass of Saccharomyces shows a normal distribution relationship with the change of fermentation environment parameters, so the Gaussian equation is used to construct the model. a = 5.91×10 4 , b = -0.01, c = 1.67. It can be seen from the model that the temperature in the fermentation chamber of 26.2℃ - 28.0℃ is most suitable for the growth of Saccharomyces, and the biomass of Saccharomyces is 5.91×10 4 copies / g when the temperature in the fermentation chamber is 27.1℃. Calculated by the linear regression equation, the P value is less than 0.05, and the model has high accuracy.

[0057] (2)-5 Construction of the Gaussian model of the biomass of Lactobacillus in high-temperature Daqu based on the change of temperature in the fermentation chamber

[0058] The biomass of Lactobacillus shows a normal distribution relationship with the change of fermentation acidity, so the Gaussian equation is used to construct the model. Among them, a = 1.48×10 6 , b = 0.576, c = 0.579. When the acidity is 0.8 - 1.1 mmol / L, it is most suitable for the growth of Saccharomyces, and the biomass is the highest at 0.9 acidity value, which is 1.48×10 6 copies / g. Calculated by the linear regression equation, the P value is less than 0.001, and the model has high accuracy.

[0059] Example 3: Verification of the change in the content of the genus Saccharomyces based on the change in the temperature of the fermentation chamber

[0060] Referring to the methods of Examples 1 and 2, a Gaussian model was constructed for the biomass of the genus Saccharomyces in medium-temperature Daqu based on the change in the temperature of the fermentation chamber. The model is as Figure 6 shown. The model was verified. The verification method was to regulate the temperature of the fermentation chamber by using the relationship between the biomass of the genus Saccharomyces reflected by the model and the temperature of the fermentation chamber, so as to achieve the purpose of regulating the biomass of the genus Saccharomyces. The data used for modeling were collected in the first 15 days of Daqu fermentation (after 15 days of fermentation, the moisture content in Daqu was lower than 15%, and the biological growth activity was greatly reduced).

[0061] The Daqu fermentation process was 28 days, and the product temperature was not higher than 50°C. Sampling and microbial quantification were carried out on the fermentation process in the first 15 days of Daqu fermentation. The sampling positions were three samples at the centers of the upper, middle, and lower layers of the koji pile.

[0062] As Figure 6 shown in the model, the content of the genus Saccharomyces in Daqu was relatively low below 22°C. At the 0th day of the fermentation process, when the temperature of the fermentation chamber was 21°C, the content of the genus Saccharomyces in the sample was 11537 copies / g. The temperature of the fermentation chamber was adjusted to 24°C using a temperature control system. At the 2nd day of fermentation, the temperature of the fermentation chamber stabilized at 24°C. At this time, the content of the genus Saccharomyces in the sample was 3.57×10 4 copies / g. The temperature was further increased. At the 4th - 5th days of fermentation, the temperature of the fermentation chamber stabilized at 25 - 29°C. The content of the genus Saccharomyces at this stage was 6.71×10 4 -8.37×10 4 copies / g, which was significantly higher than the highest content of the genus Saccharomyces in the natural fermentation koji room without regulation (5.91×10 4 copies / g), achieving a good improvement.

[0063] As Figure 7 shown, a correlation analysis was carried out on the sample data and the model data of the fermentation process, and it was found that the model calculation data was significantly correlated with the content of the genus Saccharomyces in the samples in the verification experiment, conforming to the model regulation law.

[0064] Although the present invention has been disclosed above with preferred embodiments, it is not intended to limit the present invention. Anyone familiar with this technology can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention should be defined by the claims.

Claims

1. A method for predicting and controlling the microbial flora in the koji-making process, characterized in that, It includes the following steps: (1) Determine the core microorganisms according to the microbial content during the koji-making process and their effects on the koji-making process. The core microorganisms serve as the regulation targets during the koji-making process; (2) Use the structural equation model to establish the interaction network among fermentation environment parameters, fermentation physical and chemical indexes, and microbial content, and determine the key indexes for controlling the content of microbial flora according to the significance of the interactions; (3) Conduct a normalization test on the data of the microbial content significantly controlled by each fermentation environment parameter or fermentation physical and chemical index. After judging the data normality, use a univariate regression test to judge the correlation between the indexes obtained in step (2) and the biomass of different genera of microorganisms. The fermentation environment parameter or fermentation physical and chemical index with the strongest correlation is used as the independent variable for predicting the biomass; (4) Construct a biomass change model based on the Matlab curve fitting program For data conforming to the normal distribution law, a Guassian model is used for model construction, and for data with other distribution laws, a linear model is used for model construction; among them, The equation of the Gaussian model is: y = a × exp(-((x - b) / c)) 2 ), where y is the microbial content, a, b, and c are constants, and x is the fermentation environment parameter; The equation of the univariate linear regression model is: y = ax + b, where y is the microbial content, a and b are constants, and x is the fermentation environment parameter; (5) During the fermentation process, detect the fermentation environment parameters or fermentation physical and chemical indexes, predict the content of the core microorganisms according to the model established in step (4), and make the content of the core microorganisms reach a content beneficial to koji-making by regulating the fermentation environment parameters or fermentation physical and chemical indexes; The core microorganisms include: bacteria, yeasts, and molds.

2. The method for predicting and controlling the microbial flora in the koji-making process according to claim 1, characterized in that, After completing the construction of the structural equation model in step (2), perform model fitting calculation, and the P value is required to be less than 0.

05.

3. The method for predicting and controlling the microbial flora in the koji-making process according to claim 1, characterized in that, The fermentation environment parameters refer to the temperature in the fermentation chamber and the air humidity in the fermentation chamber.

4. The method for predicting and controlling the microbial flora in the koji-making process according to claim 1, wherein The regulation in step (5) is carried out at the beginning of fermentation and at different fermentation stages.

5. The method for predicting and controlling the microbial flora in the koji-making process according to claim 1, characterized in that, The data used for modeling is statistically tested.

6. The method for predicting and controlling the microbial flora in the koji-making process according to claim 1, wherein Step (4) also verifies the accuracy of the constructed model and adjusts the parameters of the equation according to the verification results.

7. The method for predicting and controlling the microbial flora in the koji-making process according to claim 1, wherein The regulation of the fermentation environment parameters is based on non-contact system control.

8. The method for predicting and controlling the microbial flora in the koji-making process according to claim 1, wherein, The koji can also be replaced with a traditional food fermentation process or koji produced by an indoor fermentation method.

9. The method for predicting and controlling the microbial flora in the koji-making process according to claim 8, characterized in that, The traditional food fermentation process is a traditional fermented food fermented in a fermentation chamber. The koji produced by the indoor fermentation method is koji for liquor production, koji for vinegar production, and koji for other fermented food production.

10. The method for predicting and controlling the microbial flora in the koji-making process according to claim 1, characterized in that, The core microorganisms include: Saccharomycopsis, Pichia, Wickerhamomyces, Lactobacillus, Bacillus, Kroppenstedtia, Rhizopus, and Aspergillus.

Images