Method for determining parameter combination of microbial fermentation process

Abstract

This application provides a method for determining a parameter combination for a microbial fermentation process. The method comprises (1) constructing a prediction model; (2) iteratively optimizing the parameter combination to be optimized using the prediction model; (3) conducting a microbial fermentation experiment using the recommended parameter combination; (4) updating the prediction model; and (5) repeating steps (1) to (4) until the final parameter combination for the microbial fermentation process is obtained. This method improves R&D efficiency, reduces experimental costs, and achieves precise and coordinated optimization of the flavor and functional characteristics of microbial fermentation products through intelligent and systematic means. The fermentation products developed using the method of this application can enhance unique flavor and texture, significantly improve their nutritional value, and enrich or release a variety of active substances, achieving multiple goals such as enhancing functional ingredients, improving bioavailability, reducing toxicity and increasing efficacy, removing harmful impurities, and improving taste. The method can be widely used in food development, skin care products, medical treatment and other fields.

Description

Technical Field

[0001] The present application relates to the field of microbial fermentation technology, and in particular, to a method for determining a parameter combination of a microbial fermentation process. Background Art

[0002] As a highly controllable biotransformation and value-added method, microbial directed fermentation plays a vital role in a variety of fields, including modern biomanufacturing, natural product development, functional raw material upgrading, utilization of traditional Chinese medicine resources, and health product innovation. Through microbial fermentation, not only can raw materials be given unique flavor and texture, but nutritional value can also be significantly improved, and a variety of active substances can be enriched or released, achieving multiple goals such as enhancing functional ingredients, improving bioavailability, reducing toxicity and increasing efficacy, removing harmful impurities, and improving taste. For example, in the development of fermented beverages, aroma, sweetness and sourness, and the content of bioactive substances can be significantly improved through process optimization; in the production of food additives (such as microbial acidulants), the specific organic acid spectrum can be precisely controlled, while giving the product additional functional properties; in the fermentation process of plants, traditional Chinese medicine, and natural medicinal resources, microorganisms can be directed to enrich functional ingredients such as polysaccharides, flavonoids, saponins, and phenols, while reducing the content of toxic ingredients or reducing the formation of irritating byproducts.

[0003] With the rise of smart manufacturing, the field of microbial fermentation is accelerating its transformation toward automation, digitization, and intelligentization. Leveraging cutting-edge technologies such as big data, the Internet of Things, and artificial intelligence, smart manufacturing enables digitalization, visualization, and refined control of the entire process—from raw materials and processes to process monitoring and product quality—providing a solid foundation for efficient, sustainable, and high-quality biomanufacturing.

[0004] However, the research and development of microbial directed fermentation processes generally faces the following challenges: on the one hand, microbial species and their combinations are extremely diverse, and different strains have complex metabolic differences in substrate conversion, active substance production, detoxification and synergy; on the other hand, the culture medium formula (such as carbon source, nitrogen source, plant extracts, traditional Chinese medicine, mineral elements, trace elements, functional precursors, etc.) and process parameters (such as temperature, pH, inoculation size, fermentation time, aeration and stirring, etc.) are numerous and interact with each other in a complex manner, resulting in the overall process space being highly multidimensional and nonlinear.

[0005] Traditional process development relies heavily on experience and extensive trial and error, resulting in low efficiency and difficulty achieving effective synergy and balance among multiple objectives (such as high yield and activity, ingredient enrichment, toxicity reduction, and flavor improvement). This is particularly true given the complex background of natural products and the multi-objective optimization requirements.

[0006] Rapid advances in artificial intelligence (AI), particularly in data mining, complex system modeling, and multi-objective optimization and decision-making, have revolutionized the process and functional optimization of microbial directed fermentation. Combined with intelligent manufacturing systems, AI can deeply explore the connections between raw materials, strains, processes, and products based on big data. It can intelligently predict and screen key variable combinations, efficiently explore process parameter space, and monitor and dynamically adjust the fermentation process in real time, collaboratively achieving the optimal balance of multiple objectives, including enhancing functional ingredients, increasing bioavailability, reducing toxicity and increasing efficacy, and improving flavor.

[0007] Therefore, developing an artificial intelligence-based microbial directed fermentation process and functional optimization method and applying it to the intelligent manufacturing platform will significantly improve the efficiency of high-value development of natural products and health products, and promote the upgrading of biomanufacturing to automation, intelligence, and greenness. It has important theoretical significance and broad application prospects. Summary of the Invention

[0008] The present application aims to solve at least one of the technical problems existing in the prior art to a certain extent. To this end, the present application provides a method for determining the parameter combination of a microbial fermentation process. The method improves R&D efficiency, reduces experimental costs, and realizes accurate and synergistic optimization of the flavor and functional characteristics of microbial fermentation products through intelligent and systematic means. The fermentation products developed using the method of the present application can increase unique flavor and texture, significantly improve their nutritional value, and enrich or release a variety of active substances, achieving multiple goals such as functional ingredient enhancement, bioavailability improvement, toxicity reduction and efficacy enhancement, removal of harmful impurities, and improved taste. It can be widely used in food development, skin care products, medical treatment and other fields.

[0009] Specifically, the technical solution of this application is as follows:

[0010] In a first aspect of the present application, the present application proposes a method for determining a parameter combination for a microbial fermentation process. According to an embodiment of the present application, the method includes: (1) using an existing experimental data set to construct a prediction model, wherein the existing experimental data set includes process parameters of multiple microbial fermentation processes and corresponding evaluation results, and the prediction model is used to predict the evaluation results of the microbial fermentation process, wherein the evaluation results are determined based on multiple evaluation indicators, and the multiple evaluation indicators include flavor characteristics and functional characteristics; (2) obtaining a parameter combination to be optimized for the fermentation process, and using the prediction model to iteratively optimize the parameter combination to be optimized to obtain a recommended parameter combination; (3) using the recommended parameter combination to conduct a microbial fermentation experiment to obtain a new evaluation result ; (4) adding the new evaluation result to the existing experimental data set to update the prediction model; and (5) repeating steps (1) to (4) until the final parameter combination of the microbial fermentation process is obtained; wherein the iterative optimization includes: (a) inputting the fermentation process parameter combination to be optimized into the prediction model to obtain the prediction evaluation result of the parameter combination to be optimized; (b) based on the prediction evaluation result, using the acquisition function to determine the new fermentation process parameters; (c) inputting the new fermentation process parameters into the prediction model, and repeating steps (a) and (b) for at least one cycle to obtain the recommended parameter combination.

[0011] This method enables precise control and coordinated optimization of fermentation process parameter combinations, significantly improving R&D efficiency and reducing the number of experiments and costs. Furthermore, it enhances our understanding of microbial fermentation and is highly modular and scalable, making it widely applicable to the development of a wide range of fermentation products, thereby comprehensively optimizing the performance and quality of microbial fermentation products.

[0012] According to an embodiment of the present application, the prediction model is a Gaussian process regression model or a random forest model.

[0013] According to an embodiment of the present application, the combination of parameters to be optimized for the fermentation process is composed of at least one variable of microbial strains and their compound ratios, culture medium components and fermentation process parameters, wherein the types of the variables include continuous, discrete and categorical variables; the combination of parameters to be optimized for the fermentation process is determined by the following steps: constructing a design space in the form of a Cartesian product based on the variables; and selecting the combination of parameters to be optimized for the fermentation process from the design space by Latin hypercube sampling.

[0014] According to an embodiment of the present application, the prediction evaluation result is determined based on the following steps:

[0015] Using the prediction model, obtaining the evaluation index prediction results corresponding to the fermentation process parameter combination to be optimized;

[0016] Based on the evaluation index prediction results, the multi-objective fuzzy score is determined as the prediction evaluation result according to the following formula:

[0017] , k is the number of evaluation indicators, y ij is the prediction result of the jth evaluation index under the i-th parameter combination, is the weight of each evaluation index, which satisfies ;

[0018] d j (y ij ) is the expected function for y ij The processing result,

[0019] Among them, the expected function ,

[0020] L j is the minimum acceptable value of the jth evaluation index,

[0021] T j is the ideal target value of the jth evaluation index,

[0022] r j is a predetermined shape parameter.

[0023] According to an embodiment of the present application, the acquisition function is an expected improvement function, a knowledge gradient function, or a confidence upper bound function.

[0024] According to an embodiment of the present application, before the existing experimental data is geometrically input into the prediction model, the process parameters are preprocessed in advance, and the preprocessing includes: filling missing values ​​with the mean method, median method or model prediction method, and / or normalizing continuous parameters to eliminate dimensional differences; for categorical variables, converting them into numerical variables using one-hot encoding.

[0025] In a second aspect, the present application provides a method for preparing a fermented product. According to an embodiment of the present application, the method comprises: obtaining a combination of fermentation process parameters for the fermented product according to the method described in the first aspect based on multiple optimization objectives; and performing a fermentation process using the combination of fermentation process parameters to obtain the fermented product.

[0026] According to an embodiment of the present application, the fermented product includes at least one of a microbial-derived acidulant, a mulberry pomace fermentation liquid, and a cordyceps sinensis fermentation filtrate.

[0027] In a third aspect of the present application, an electronic device is provided. According to an embodiment of the present application, the device includes: a processor and a memory; the memory is configured to store a computer program; and the processor is configured to execute the computer program to implement the method described in the first aspect.

[0028] In a fourth aspect of the present application, a computer-readable storage medium is provided. According to an embodiment of the present application, the computer-readable storage medium stores computer instructions or a program, which, when executed on a computer, causes the method described in the first aspect to be executed.

[0029] In a fifth aspect of the present application, a computer program product is provided. According to an embodiment of the present application, the computer program product includes computer instructions, and when part or all of the computer instructions are run on a computer, the method described in the first aspect is executed.

[0030] The aforementioned electronic device, computer-readable storage medium, and computer program product provide a method for determining parameter combinations for a microbial fermentation process through the automatic execution of computer instructions, achieving efficient automation. Furthermore, the instruction-based nature of the method makes it widely applicable in various application scenarios.

[0031] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become obvious from the description below, or will be learned through practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0033] Figure 1 A schematic flow chart of a method for determining a parameter combination for a microbial fermentation process provided in one embodiment of the present application;

[0034] Figure 2 A schematic diagram of a device for determining a parameter combination for a microbial fermentation process provided in one embodiment of the present application;

[0035] Figure 3 A schematic diagram of an electronic device provided in accordance with an embodiment of the present application. DETAILED DESCRIPTION

[0036] The embodiments of the present application are described in detail below. The embodiments described below are exemplary and are only used to explain the present application, and should not be understood as limiting the present application.

[0037] It should be noted that the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, features defined as "first" or "second" may explicitly or implicitly include one or more of such features. Furthermore, in the description of this application, unless otherwise specified, "plurality" means two or more.

[0038] The endpoints of the ranges and any values ​​disclosed herein are not limited to the precise ranges or values, and these ranges or values ​​should be understood to include values ​​close to these ranges or values. For numerical ranges, the endpoints of each range, the endpoints of each range and individual point values, and the individual point values ​​can be combined with each other to obtain one or more new numerical ranges, which should be considered to be specifically disclosed herein.

[0039] In this document, the terms "include" or "comprising" are open expressions, that is, including the contents specified in this application, but not excluding other contents.

[0040] As used herein, the terms "optionally," "optional," or "optionally" generally mean that the subsequently described event or circumstance may but need not occur, and that the description includes instances where the event or circumstance occurs and instances where it does not.

[0041] As used herein, the term "module" or "unit" refers to a computer program or portion of a computer program that has a predetermined function and works together with other related components to achieve the predetermined goal. This function may be implemented in whole or in part using software, hardware (such as processing circuitry or memory), or a combination thereof. Similarly, a single processor (or multiple processors or memories) may be used to implement one or more modules or units. Furthermore, each module or unit may be part of an overall module or unit that incorporates the functionality of that module or unit.

[0042] In this article, the term "model" refers to an algorithm that automatically performs tasks such as prediction, classification, recognition, or decision-making by learning and analyzing input data. The model's learning process is based on statistical principles and data pattern recognition, using training datasets to adjust model parameters and optimize the model to improve its predictive or inference capabilities. Models can employ various algorithms and techniques, such as neural networks, support vector machines, decision trees, random forests, and deep learning. These models can be trained and optimized through supervised learning, unsupervised learning, or reinforcement learning. In practical applications, models can be used in a variety of fields, such as natural language processing, image recognition, pattern recognition, data mining, recommender systems, and predictive analytics. They have significant potential for large-scale data processing, automated decision-making, and intelligent systems. However, it should be noted that in specific applications, in-depth research on the features and models used for prediction is required to achieve satisfactory prediction results. Otherwise, various problems may arise, such as overfitting, underfitting, poor generalization, and low robustness. After extensive experimental verification, the inventors of this application discovered that by using existing microbial fermentation datasets to train machine learning models (such as a Gaussian process regression model under a Bayesian optimization framework), it is possible to learn the complex nonlinear mapping relationship between input parameters and various flavor / functional outputs, and to predict untested parameter combinations while providing an uncertainty assessment of the prediction results.

[0043] As used herein, the term "GCDC" stands for sodium glycochenodeoxycholate, a bile salt. During fermentation, if microorganisms have the ability to degrade this bile salt, the GCDC content in the fermentation system can be reduced, and the degradation rate can reflect the microorganism's ability to degrade GCDC.

[0044] As used herein, the term "TAC" refers to sodium taurodeoxycholate, a type of bile salt.

[0045] As used herein, the term "TCDC" refers to sodium taurochenodeoxycholate, a type of bile salt.

[0046] As used herein, the term "TDC" refers to sodium taurodeoxycholate, a type of bile salt.

[0047] Traditional methods rely on the accumulated experience of R&D personnel and a large number of trial and error experiments. Due to the numerous influencing factors and their complex interactions, finding the optimal parameter combination that meets the ideal flavor and functional properties is time-consuming, costly and inefficient, and is even more inefficient when balancing multiple optimization objectives.

[0048] In response to the above shortcomings, the inventors have made many improvements. The following is an explanation of the technical solution of this application:

[0049] In the first aspect of the present application, the present application proposes a method for determining a parameter combination of a microbial fermentation process. According to an embodiment of the present application, Figure 1 As shown, the method includes:

[0050] S110 (1) constructing a prediction model using an existing experimental data set, wherein the existing experimental data set includes process parameters of multiple microbial fermentation processes and corresponding evaluation results, wherein the prediction model is used to predict the evaluation results of the microbial fermentation process, wherein the evaluation results are determined based on multiple evaluation indicators, wherein the multiple evaluation indicators include flavor characteristics and functional characteristics;

[0051] S120 (2) obtaining a fermentation process parameter combination to be optimized, and iteratively optimizing the parameter combination to be optimized using the prediction model to obtain a recommended parameter combination;

[0052] S130 (3) conducting a microbial fermentation experiment using the recommended parameter combination to obtain a new evaluation result;

[0053] S140 (4) adding the new evaluation results to the existing experimental data set to update the prediction model; and

[0054] S150 (5) repeating steps (1) to (4) until the final parameter combination of the microbial fermentation process is obtained; wherein the iterative optimization includes: (a) inputting the fermentation process parameter combination to be optimized into the prediction model to obtain a prediction evaluation result of the parameter combination to be optimized; (b) based on the prediction evaluation result, determining new fermentation process parameters using an acquisition function; (c) inputting the new fermentation process parameters into the prediction model, and repeating steps (a) and (b) for at least one cycle to obtain the recommended parameter combination.

[0055] This method enables precise control and coordinated optimization of fermentation process parameter combinations, significantly improving R&D efficiency and reducing the number of experiments and costs. Furthermore, it enhances our understanding of microbial fermentation and is highly modular and scalable, making it widely applicable to the development of a wide range of fermentation products, thereby comprehensively optimizing the performance and quality of microbial fermentation products.

[0056] According to an embodiment of the present application, the prediction model is a Gaussian process regression model or a random forest model, thereby learning the complex nonlinear mapping relationship between input parameters and various flavor / functional outputs, and predicting untested parameter combinations.

[0057] According to an embodiment of the present application, the combination of parameters to be optimized for the fermentation process is composed of at least one variable of microbial strains and their compound ratios, culture medium components, and fermentation process parameters, wherein the types of the variables include continuous, discrete, and categorical variables; the combination of parameters to be optimized for the fermentation process is determined by the following steps: constructing a design space in the form of a Cartesian product based on the variables; and selecting the combination of parameters to be optimized for the fermentation process from the design space by Latin hypercube sampling. Thus, by constructing a design space in the form of a Cartesian product of variables such as microbial strains and their compound ratios, culture medium components, and fermentation process parameters, all possible parameter combinations can be fully covered. Subsequently, the Latin hypercube sampling method is used to scientifically select representative parameter combinations from the huge design space as candidate solutions to be optimized. The fermentation process parameter space is fully explored, and the number of experiments can be effectively reduced, the optimization efficiency can be improved, and a strong foundation can be provided for the subsequent use of predictive models for precise optimization.

[0058] According to an embodiment of the present application, the prediction evaluation result is determined based on the following steps:

[0059] Using the prediction model, obtaining the evaluation index prediction results corresponding to the fermentation process parameter combination to be optimized;

[0060] Based on the evaluation index prediction results, the multi-objective fuzzy score is determined as the prediction evaluation result according to the following formula:

[0061] , k is the number of evaluation indicators, y ij is the prediction result of the jth evaluation index under the i-th parameter combination, is the weight of each evaluation index, which satisfies ;

[0062] d j (y ij ) is the expected function for y ij The processing result,

[0063] Among them, the expected function ,

[0064] L j is the minimum acceptable value of the jth evaluation index,

[0065] T j is the ideal target value of the jth evaluation index,

[0066] r jThe multi-objective fuzzy scoring formula is then used to transform the result into a comprehensive prediction and evaluation result. This scoring formula can comprehensively reflect the overall performance of the parameter combination in terms of multiple flavor and functional indicators, thus providing a basis for subsequent optimization decisions.

[0067] According to some specific embodiments of the present application, the method uses a Bayesian optimization framework, and the objective formula is , obtain the parameter combination with the maximum value of multi-objective fuzzy score, X is a different parameter combination.

[0068] According to some specific embodiments of the present application, the evaluation indicators include flavor characteristics and functional characteristics, wherein the flavor characteristics include: the concentration range of organic acids such as lactic acid and acetic acid, the target content or ratio of specific volatile flavor compounds such as esters, aldehydes and ketones, and the comprehensive score of sensory evaluation; the functional characteristics include pH value controlled within a specific range, maximization of the content of specific bioactive substances such as polyphenols, vitamins, and functional peptides, indicators of in vitro antioxidant activity or specific enzyme inhibition activity such as α-glucosidase inhibition rate, the number of viable bacteria in probiotic drinks or their bile salt tolerance in a simulated digestion environment.

[0069] According to an embodiment of the present application, the acquisition function is an expected improvement function, a knowledge gradient function, or a confidence upper bound function. Thus, the acquisition function is used to evaluate the sampling value of different potential experimental parameter combinations, thereby guiding the selection of the next experimental parameters. The formula is , X is a different parameter combination.

[0070] According to an embodiment of the present application, before inputting the existing experimental data set into the prediction model, the process parameters are preprocessed. This preprocessing includes: filling missing values ​​using the mean, median, or model prediction method; and / or normalizing continuous parameters to eliminate dimensionality differences. Categorical variables are converted to numeric variables using one-hot encoding. This meets the input requirements of the prediction model.

[0071] According to some specific embodiments of the present application, the preprocessing further includes: data type conversion, encoding text data into categorical variables, setting categorical variables , converted into: Thus, the input requirements of the prediction model are met.

[0072] According to some specific embodiments of the present application, the device 200 for determining the parameter combination of the microbial fermentation process is as follows: Figure 2As shown, the device 200 includes: an initialization module 210, a parameter optimization module 220, an experimental verification module 230, a model updating module 240, and an iterative feedback module 250.

[0073] An initialization module 210 is used to construct a prediction model; a parameter optimization module 220 is used to obtain a recommended parameter combination; an experimental verification module 230 is used to conduct a microbial fermentation experiment using the recommended parameter combination to obtain a new evaluation result; a model update module 240 is used to add the new evaluation result to the existing experimental data set to update the prediction model; and an iterative feedback module 250 is used to iterate modules 210~240 until the final parameter combination of the microbial fermentation process is obtained.

[0074] In a second aspect, this application provides a method for preparing a fermented product. According to an embodiment of this application, the method comprises: obtaining a combination of fermentation process parameters for the fermented product according to the method described in the first aspect based on multiple optimization objectives; and performing a fermentation process using the combination of fermentation process parameters to obtain the fermented product. Thus, the optimization method described in the first aspect is used to determine the ideal combination of fermentation process parameters for the fermented product. This combination is then used to perform a fermentation process, thereby achieving efficient preparation of the fermented product and ensuring that the product meets the expected standards for flavor and functional properties.

[0075] According to an embodiment of the present application, the fermented product includes at least one of a microbial acidulant, a mulberry pomace fermentation broth, and a Cordyceps sinensis fermentation filtrate. Therefore, the present application can be applied to optimization issues involving microbial culture and fermentation processes in various other fermented foods, probiotic preparations, skin care products, and medical applications.

[0076] In a third aspect of the present application, the present application provides an electronic device. Figure 3 , the electronic device 300 may be an execution device of the above method, but is not limited thereto. Figure 3 As shown, the electronic device 300 may include:

[0077] The memory 310 and the processor 320 are configured to store a computer program 330 and transmit the computer program 330 to the processor 320. In other words, the processor 320 can call and execute the computer program 330 from the memory 310 to implement the method in the embodiment of the present application.

[0078] For example, the processor 320 may be configured to execute the steps of the above method according to the instructions in the computer program 330 .

[0079] In some embodiments of the present application, the processor 320 may include but is not limited to:

[0080] General-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0081] In some embodiments of the present application, the memory 310 includes but is not limited to:

[0082] Volatile memory and / or non-volatile memory. Non-volatile memory can be read-only memory (ROM), programmable ROM (PROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct RAM bus random access memory (DR RAM).

[0083] In some embodiments of the present application, the computer program 330 may be divided into one or more modules, which are stored in the memory 310 and executed by the processor 320 to implement the method provided by the present application. The one or more modules may be a series of computer program instruction segments capable of implementing specific functions, and the instruction segments are used to describe the execution process of the computer program 330 in the electronic device.

[0084] like Figure 3 As shown, the electronic device 300 may further include:

[0085] The transceiver 340 may be connected to the processor 320 or the memory 310 .

[0086] The processor 320 may control the transceiver 340 to communicate with other devices. Specifically, the processor 320 may send information or data to other devices or receive information or data sent by other devices. The transceiver 340 may include a transmitter and a receiver. The transceiver 340 may further include one or more antennas.

[0087] It should be understood that the various components in the electronic device 300 are connected via a bus system, wherein the bus system includes not only a data bus but also a power bus, a control bus and a status signal bus.

[0088] In a fourth aspect, the present application provides a computer-readable storage medium. Computer instructions or a program are stored thereon, which, when executed by a computer, enable the computer to perform the method of the aforementioned method embodiment. Alternatively, the present application also provides a computer program product containing instructions, which, when executed by a computer, enable the computer to perform the method of the aforementioned method embodiment.

[0089] In a fifth aspect, the present application provides a computer program product. The computer program product or computer program includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the method of the above-described method embodiment.

[0090] In other words, when implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions according to the embodiments of the present application are fully or partially generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer, or a data storage device such as a server or data center that integrates one or more available media. The available medium can be magnetic media (e.g., floppy disk, hard disk, magnetic tape), optical media (e.g., digital video disc (DVD)), or semiconductor media (e.g., solid-state drive (SSD)).

[0091] Those skilled in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0092] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the modules is merely a logical function division. In actual implementation, there may be other division methods, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0093] Modules described as separate components may or may not be physically separate, and components displayed as modules may or may not be physical modules, i.e., they may be located in one place or distributed across multiple network elements. Some or all of the modules may be selected to implement the present embodiment according to actual needs. For example, the functional modules in the various embodiments of the present application may be integrated into a single processing module, each module may exist physically separately, or two or more modules may be integrated into a single module.

[0094] Below in conjunction with embodiment, the scheme of the application will be explained. Those skilled in the art will appreciate that the following examples are merely for illustration of the application and should not be considered as limiting the scope of the application. Where specific techniques or conditions are not indicated in the examples, they are carried out according to the techniques or conditions described in the literature in this area or according to the product specifications. Reagents used or instruments not indicated by the manufacturer are conventional products that can be obtained commercially.

[0095] Example 1: Optimizing the development of microbial-derived acidulants with specific acid profiles and functional properties

[0096] The goal of this embodiment is to develop a novel microbial-derived acidulant that not only needs to achieve a predetermined overall acidity (measured by pH), but also needs to possess a specific organic acid profile (e.g., a high content of lactic acid to provide a mild sour taste, while strictly controlling the content of acids such as acetic acid that may impart irritation or unpleasant flavors, and possibly focusing on the synergistic effects of other specific organic acids such as butyric acid). Furthermore, it is also desired that the acidulant possess certain additional functional properties (e.g., lipid-lowering, bile-lowering, etc.).

[0097] To this end, the following fermentation process parameters were selected for combination:

[0098] Microbial strains: Five strains were investigated, including Lactobacillus plantarum VB165, Weizmannella coagulans VB320, Lactobacillus paracasei VB306, Lactobacillus rhamnosus VB255, and Bifidobacterium animalis VB315. Fermentation can be performed using a single strain or in combination with different ratios of two strains.

[0099] Lactobacillus plantarum VB165 was deposited in the General Microbiology Center of China Culture Collection Administration of Microorganisms on September 28, 2022, with the deposit number CGMCC NO.25839;

[0100] Weizmannella coagulans VB320 was deposited in the China Center for Type Culture Collection on April 3, 2024, with the deposit number CCTCC No: M 2024626;

[0101] Lactobacillus paracasei VB306 was deposited in the General Microbiology Center of China Culture Collection Administration on July 24, 2023, with the deposit number CGMCC No. 27999;

[0102] Lactobacillus rhamnosus VB255 was deposited in the General Microbiology Center of China Culture Collection Administration of Microorganisms on March 31, 2023, with the deposit number CGMCC No. 26968;

[0103] Animal Bifidobacterium VB315 was deposited in the General Microbiology Center of China Culture Collection Administration on August 14, 2023, with the deposit number CGMCC No.28159.

[0104] Culture medium components: carbon source addition: 1%~3%; different concentrations of yeast extract: 0~1.5%.

[0105] Fermentation process parameters: Fermentation temperature (range: 25~37°C).

[0106] Expected functions were set for lactate, acetate, butyrate, ethanol, lipase, and four bile salt-lowering abilities. Relative importance weights were assigned based on product development priorities, as shown in Table 1.

[0107] Table 1

[0108]

[0109] The optimization is performed using the method described in S110 to S150 of this application, wherein the Gaussian process regression model is selected as the proxy model, and the expected utility (Q-Noisy Expected Improvement, qEI) is selected as the acquisition function to guide the selection of the next round of experimental points in order to maximize the improvement of the multi-objective fuzzy score.

[0110] After a total of four rounds of iterative optimization, five new experimental points were generated and executed in each round based on the recommendations. The recommended parameter combinations shown in Table 2 were obtained.

[0111] Table 2

[0112]

[0113] The validation experiments were repeated using the recommended parameter combinations shown in Table 2, and the resulting fermentation broth characteristics are shown in Table 3.

[0114] Table 3

[0115]

[0116] In this way, a microbial acidulant with good performance in multiple preset flavor and functional indicators was successfully prepared.

[0117] Example 2: Method for improving flavor and producing acid by fermentation of mulberry pomace

[0118] The goal of this example is to significantly increase the acidity of mulberry pomace fermentation liquid, creating a pleasant, lactic acid-dominated sour flavor profile and improving its original sweet, non-acidic taste. At the same time, while achieving flavor improvement, the minimum number of microbial strains (or the lowest effective inoculum size) is used to improve fermentation efficiency and reduce potential costs.

[0119] A mulberry pomace-based liquid was prepared with water, mulberry pomace and mulberry filtrate, and a microbial fermentation experiment was carried out using the mulberry pomace-based liquid.

[0120] To this end, the following fermentation process parameters were selected for combination:

[0121] Microbial strains: Five strains were investigated, including Lactobacillus plantarum VB165, Weizmannella coagulans VB320, Lactobacillus paracasei VB306, Lactobacillus rhamnosus VB255, and Bifidobacterium animalis VB315. Fermentation can be performed using a single strain or in combination with different proportions of two strains.

[0122] Culture medium components: carbon source addition 0~3%;

[0123] Fermentation process parameters: fermentation time 48 h~96 h; fermentation temperature 25℃, 30℃, 37℃.

[0124] Expected functions were set for lactic acid and acetic acid, respectively. Relative importance weights were assigned based on product development priorities, as shown in Table 4.

[0125] Table 4

[0126]

[0127] The optimization is performed using the method described in S110 to S150 of this application, wherein the Gaussian process regression model is selected as the proxy model, and the expected utility (Q-Noisy Expected Improvement, qEI) is selected as the acquisition function to guide the selection of the next round of experimental points in order to maximize the improvement of the multi-objective fuzzy score.

[0128] After five rounds of iterative optimization and 42 mulberry pomace fermentation experiments, the recommended parameter combinations shown in Table 5 were obtained.

[0129] Table 5

[0130]

[0131] The validation experiments were repeated using the recommended parameter combinations shown in Table 5, and the resulting fermentation broth characteristics are shown in Table 6.

[0132] Table 6

[0133]

[0134] Thus, a mulberry pomace fermentation liquid that meets multiple conditions was successfully prepared.

[0135] Example 3: Method for Dual Optimization of Safety and Functionality of Cordyceps Sinensis Fermentation Filtrate

[0136] Cordyceps sinensis fermentation filtrate is rich in various bioactive ingredients and has broad potential for medicinal and skincare applications. However, its application in topical products is limited by two factors: first, the fermentation process produces high levels of endotoxins (initial detection value is 4.69 EU / mg), which poses a potential risk of irritation and sensitization; second, its relatively low inhibition rate of hyaluronidase (initial value is 65.16%), limiting its moisturizing, anti-inflammatory, and anti-aging properties.

[0137] The goal of this example is to reduce the high level of endotoxins produced during the fermentation process and to increase the hyaluronidase inhibition rate, thereby achieving both safety and functionality. A microbial fermentation experiment was conducted using Cordyceps sinensis liquid.

[0138] To this end, the following fermentation process parameters were selected for combination:

[0139] Microbial strain: Bacillus subtilis natto VB205 was used for fermentation, and different inoculation amounts were investigated.

[0140] Bacillus natto VB205 was deposited in the China General Microbial Culture Collection Center on September 5, 2022, with the deposit number CGMCC No. 25652.

[0141] Culture medium components: different concentrations of yeast extract powder; different concentrations of glycerol;

[0142] Fermentation process parameters: fermentation temperature; initial pH; ventilation volume; fermentation time.

[0143] The expectation functions were set based on endotoxin content and hyaluronidase inhibition rate. Relative importance weights were assigned based on product priority, as shown in Table 7.

[0144] Table 7

[0145]

[0146] The optimization is performed using the method described in S110 to S150 of this application, wherein the Gaussian process regression model is selected as the proxy model, and the expected utility (Q-Noisy Expected Improvement, qEI) is selected as the acquisition function to guide the selection of the next round of experimental points in order to maximize the improvement of the multi-objective fuzzy score.

[0147] After four rounds of iterative optimization, a total of 20 Cordyceps sinensis fermentation experiments were performed, and the recommended parameter combinations shown in Table 8 were obtained.

[0148] Table 8

[0149]

[0150] The validation experiment was repeated using the recommended parameter combination shown in Table 8, and the characteristics of the Cordyceps sinensis fermentation filtrate obtained are shown in Table 9.

[0151] Table 9

[0152]

[0153] This resulted in a set of Cordyceps fermentation process parameters that balance safety and functionality. Validation experiments demonstrated that the endotoxin level in the fermentation broth dropped from an initial 4.69 EU / mg to 1.12 EU / mg, and the hyaluronidase inhibition rate increased to 89.17%, significantly exceeding the initial level. This achievement lays a solid foundation for the application of Cordyceps fermentation products in high-end skincare and topical medical applications.

[0154] In the description of this specification, the reference terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and features of different embodiments or examples without contradiction.

[0155] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for determining a parameter combination for a microbial fermentation process, characterized in that: include: (1) constructing a prediction model using an existing experimental data set, wherein the existing experimental data set includes process parameters of multiple microbial fermentation processes and corresponding evaluation results, wherein the prediction model is used to predict the evaluation results of the microbial fermentation process, wherein the evaluation results are determined based on multiple evaluation indicators, wherein the multiple evaluation indicators include flavor characteristics and functional characteristics; (2) obtaining a fermentation process parameter combination to be optimized, and iteratively optimizing the parameter combination to be optimized using the prediction model to obtain a recommended parameter combination; (3) Conducting microbial fermentation experiments using the recommended parameter combination to obtain new evaluation results; (4) Supplementing the new evaluation results to the existing experimental data set to update the prediction model; and (5) Repeat steps (1) to (4) until the final parameter combination of the microbial fermentation process is obtained; Wherein, the iterative optimization includes: (a) inputting the fermentation process parameter combination to be optimized into the prediction model to obtain a prediction evaluation result of the parameter combination to be optimized; (b) determining new fermentation process parameters using an acquisition function based on the prediction and evaluation results; (c) inputting the new fermentation process parameters into the prediction model, and repeating steps (a) and (b) for at least one cycle to obtain the recommended parameter combination; The prediction model is a Gaussian process regression model or a random forest model.

2. The method according to claim 1, wherein The fermentation process parameter combination to be optimized is composed of at least one variable of microbial strains and their compound ratios, culture medium components, and fermentation process parameters, wherein the types of the variables include continuous, discrete, and categorical variables; the fermentation process parameter combination to be optimized is determined by the following steps: constructing a design space in the form of a Cartesian product based on the variables; and The fermentation process parameter combination to be optimized is selected from the design space by Latin hypercube sampling.

3. The method according to claim 1, characterized in that The prediction evaluation results are determined based on the following steps: Using the prediction model, obtaining the evaluation index prediction results corresponding to the fermentation process parameter combination to be optimized; Based on the evaluation index prediction results, the multi-objective fuzzy score is determined as the prediction evaluation result according to the following formula: , k is the number of evaluation indicators, y ij is the prediction result of the jth evaluation index under the i-th parameter combination, is the weight of each evaluation index, which satisfies ; d j (y ij ) is the expected function for y ij The processing result, Among them, the expected function , L j is the minimum acceptable value of the jth evaluation index, T j is the ideal target value of the jth evaluation index, r j is a predetermined shape parameter.

4. The method according to claim 1, wherein The acquisition function is an expected improvement function, a knowledge gradient function or a confidence upper bound function.

5. The method according to claim 1, wherein Before inputting the existing experimental data set into the prediction model, the process parameters are pre-processed, and the pre-processing includes: Fill missing values ​​using mean, median, or model prediction, and / or For continuous parameters, normalization is performed to eliminate dimensional differences; For categorical variables, one-hot encoding is used to convert them into numerical variables.

6. A method for preparing a fermented product, characterized in that: include: Based on multiple optimization objectives, obtaining a fermentation process parameter combination of the fermentation product according to the method according to any one of claims 1 to 5; The fermentation process parameter combination is adopted to carry out fermentation treatment to obtain the fermentation product.

7. The method according to claim 6, characterized in that The fermentation product comprises at least one of a microbial-derived acidulant, a mulberry pomace fermentation liquid and a cordyceps sinensis fermentation filtrate.

8. An electronic device, characterized in that: include: processor and memory; The memory is used to store computer programs; The processor is configured to execute the computer program to implement the method according to any one of claims 1 to 5.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions or programs, and when the computer instructions or programs are run on a computer, the method according to any one of claims 1 to 5 is executed.

10. A computer program product, characterized in that The computer program product includes computer instructions, and when part or all of the computer instructions are run on a computer, the method according to any one of claims 1 to 5 is executed.

Images