Method and system for evaluating the health status of hermetia illucens larvae microbial community and inoculation decision-making of microbial agent

Abstract

This invention relates to a method and system for assessing the health status of the gut microbiota and making decisions on inoculation with microbial agents in black soldier fly larvae. The method includes: acquiring biomarker data of black soldier fly larvae; normalizing the biomarker data according to mapping rules; calculating a microbiota health index based on a linear weighted model and the normalized values ​​of each biomarker; performing pattern matching between the normalized values ​​of each biomarker and a preset quantification threshold to determine the type of microbiota dysbiosis; and generating an inoculation strategy based on the comparison between the current microbiota health index and the preset threshold, the trend of the microbiota health index, and the type of microbiota dysbiosis. By fusing multi-source biomarker data through a linear weighted model, the previously invisible gut microbiota status is quantified into data, and this data can be dynamically updated, enabling farmers to monitor the health status of the larval microbiota in real time, shifting from traditional experience-based judgment and post-hoc sequencing analysis to quantitative and real-time scientific assessment.

Description

Technical Field

[0001] This invention relates to the fields of intelligent agriculture and microbial technology, and in particular to a method and system for assessing the health status of the gut microbiota of black soldier fly larvae and making decisions on inoculation with microbial agents. The method is used to quantitatively assess the health status of the gut microbiota of black soldier fly larvae and generate a precise inoculation strategy based on the type of dysbiosis. Background Technology

[0002] Black soldier fly larvae, as a highly efficient bioconverter of organic waste, have been widely used in recent years in fields such as agricultural waste treatment and resource utilization of kitchen waste. Studies have shown that a close symbiotic relationship exists between the gut microbiota of black soldier fly larvae and their host; the health of the microbiota structure directly affects the larvae's growth performance, disease resistance, and conversion efficiency. For example, beneficial bacteria (such as lactic acid bacteria and Bacillus) can promote gut health in black soldier fly larvae and inhibit the colonization of pathogenic microorganisms; while excessive proliferation of putrefactive bacteria (such as Escherichia coli and Clostridium perfringens) can lead to stunted larval growth and increased mortality. Therefore, inoculating with specific microbial agents to target and optimize the microbiota structure has become an important means of improving the efficiency of black soldier fly farming.

[0003] However, the application of microbial agents in the current black soldier fly farming process still has the following technical shortcomings:

[0004] 1. While existing technologies have achieved automated monitoring of aquaculture environmental parameters, such as the automated monitoring schemes disclosed in patents CN210275586U and CN118885040B, the monitoring targets are limited to macroscopic environmental indicators such as temperature, humidity, and ammonia concentration, and cannot detect the real-time health status of the gut microbiota of black soldier fly larvae. The addition of microbial agents often adopts a fixed cycle or fixed dosage strategy, lacking a dynamic feedback mechanism based on the microbiota status, leading to indiscriminate use of microbial agents and unstable effects.

[0005] 2. Currently, there are no quantitative indicators that can characterize the health status of the gut microbiota in black soldier fly larvae in real time. Whether the microbiota is healthy and the degree of imbalance usually relies on retrospective analysis methods such as post-hoc 16S rRNA gene sequencing. These methods are time-consuming, costly, and cannot provide immediate feedback, making it difficult to achieve real-time assessment and precise control of microbiota health in actual breeding processes.

[0006] 3. Different causes of microbial imbalance (such as excessive proliferation of putrefactive bacteria, insufficient beneficial bacteria, and decreased cellulose degradation function) require different microbial intervention strategies. However, existing detection methods cannot identify specific types of imbalance during the breeding process, and the selection of microbial agents lacks scientific basis. They often adopt a "one-size-fits-all" approach, which is not targeted and the intervention effect is not ideal.

[0007] 4. The system lacks intelligent decision-making capabilities based on the state of the microbial community, and cannot form a closed-loop control process of "state assessment → disorder diagnosis → inoculation with microbial agents". For example, the solution in patent CN118685328A only focuses on the question of "what strain to use", and fails to solve the strategic problems of "when to inoculate" and "what strain to inoculate".

[0008] Based on this, this case is proposed. Summary of the Invention

[0009] The purpose of this invention is to provide a method and system for assessing the health status of the gut microbiota of black soldier fly larvae and making decisions on inoculation with microbial agents. This method can quantitatively assess the health status of the gut microbiota of black soldier fly larvae, identify the type of microbiota imbalance, and generate a precise inoculation strategy based on this, so as to achieve intelligent and precise management of the black soldier fly farming process and improve farming efficiency and stability.

[0010] To achieve the above objectives, the present invention provides the following technical solution:

[0011] A method for assessing the health status of the microbial community of black soldier fly larvae and making decisions on inoculation with microbial agents includes the following steps:

[0012] S1. Obtain biomarker data from the rearing environment and excrement of black soldier fly larvae;

[0013] S2. Determine the relationship between each biomarker data and gut health, and classify each biomarker data according to indicator type and data level;

[0014] S3. Define mapping rules based on the relationship between each biomarker data and microbial health, and normalize each biomarker data according to the mapping rules to obtain the normalized value of each biomarker.

[0015] S4. Determine the weighting coefficients for each biomarker;

[0016] S5. Based on the linear weighted model, input the normalized values ​​of each biomarker obtained in step S3 and the weight coefficients of each biomarker obtained in step S4, and calculate the microbial health index.

[0017] S6. Perform pattern matching between the normalized values ​​of each biomarker obtained in step S3 and their preset quantification thresholds to determine the type of dysbiosis.

[0018] S7. Based on the comparison results between the current microbial health index and its preset threshold, the trend of the microbial health index in the last N times, and the type of microbial imbalance obtained in step S6, generate a microbial agent inoculation strategy.

[0019] Furthermore, in step S1, the biomarker data includes temperature, humidity, ammonia concentration, sulfide concentration, amine concentration, short-chain fatty acid concentration, ATP bioluminescence value, and absorbance of cellulose characteristic peak.

[0020] Furthermore, in step S1, the biomarker data also includes one or more of the following: functional gene abundance, beneficial bacterial genera abundance, conditionally pathogenic bacterial genera abundance, and microbial diversity index.

[0021] Furthermore, the relationship between the biomarker data and gut microbiota health includes:

[0022] Indicators with suitable ranges: temperature and humidity;

[0023] Negative correlation indicators: ammonia concentration, sulfide concentration, amine concentration, absorbance of cellulose characteristic peak, and abundance of opportunistic pathogens;

[0024] Positively correlated indicators: short-chain fatty acid concentration, ATP bioluminescence value, abundance of functional genes, abundance of beneficial bacteria genera, and bacterial community diversity index;

[0025] The biomarker data are categorized by indicator type and data level as follows:

[0026] Real-time indicators and core data: temperature, humidity, ammonia concentration, sulfide concentration, amine concentration, and short-chain fatty acid concentration;

[0027] Semi-real-time indicators and core data: ATP bioluminescence value and absorbance of cellulose characteristic peak;

[0028] Semi-real-time indicator and augmented data: functional gene abundance;

[0029] Lagging indicators and calibrated data: abundance of beneficial bacteria genera, abundance of conditionally pathogenic bacteria genera, and bacterial community diversity index;

[0030] The real-time and semi-real-time indicators are used in the calculation of the microbial health index, while the lag indicators are only used for model training and validation.

[0031] Furthermore, in step S3, the mapping rules include:

[0032] For biomarkers that are positively or negatively correlated with the bacterial community of black soldier fly larvae, their normalized values Defined as:

[0033] ;

[0034] In the formula, For the detection data of the i-th biomarker, and These are the minimum and maximum values ​​of the biomarker under healthy aquaculture conditions, determined through historical data statistics;

[0035] Temperature normalized value The calculation formula is:

[0036] ;

[0037] In the formula, T is the collected temperature value. , These are the minimum and maximum values ​​within the suitable temperature range. The optimal temperature;

[0038] Humidity normalized value The calculation formula is:

[0039] ;

[0040] In the formula, RH is the collected humidity value. , These are the minimum and maximum values ​​within the suitable humidity range. The optimal humidity level.

[0041] Furthermore, the weight coefficients of each biomarker can be determined by one of the following methods: principal component analysis, analytic hierarchy process, multiple linear regression based on historical data and microbial health reference standards, or machine learning training.

[0042] Furthermore, the expression for the linear weighted model is as follows:

[0043] ;

[0044] In the formula, Given its corresponding weight coefficient, the clip function is used to restrict the calculation result to the interval [0,1]. This indicates the health index of the gut microbiota.

[0045] Furthermore, in step S6, the types of dysbiosis and the criteria for determining each type of dysbiosis are as follows:

[0046] When the normalized value of sulfides is ≤0.2 and the normalized value of amines is ≤0.3, it is determined to be an excessive proliferation of putrefactive bacteria.

[0047] When the normalized value of the absorbance of the characteristic peak of cellulose is ≤0.3, it is determined that the cellulose degradation function is insufficient.

[0048] When the normalized value of short-chain fatty acids is ≤0.4 and the normalized value of ATP bioluminescence is ≤0.4, it is determined that the overall level of beneficial bacteria is low.

[0049] When the normalized value of functional gene abundance is ≤0.5, it is judged as insufficient bacterial community function;

[0050] When the normalized value of the abundance of conditional pathogens is ≤0.2 and the normalized value of the community diversity index is ≤0.5, it is determined to be an invasion of pathogens.

[0051] When multiple gut microbiota imbalance criteria are met, it is classified as a complex imbalance.

[0052] Furthermore, in step S7, the inoculation strategy for generating the microbial agent and the criteria for determining each inoculation strategy are as follows:

[0053] When the gut microbiota health index is >0.8, a no-vaccination instruction is generated;

[0054] When 0.6 < microbial health index ≤ 0.8 and the microbial health index shows a continuous downward trend over the last N times, a preventive vaccination instruction is generated, where N ≥ 3;

[0055] When 0.6 < microbial health index ≤ 0.8 and the microbial health index shows a non-continuous downward trend in the last N times, a non-vaccination instruction is generated;

[0056] When the gut microbiota health index is 0.4 ≤ 0.6, a therapeutic inoculation instruction is generated;

[0057] An emergency vaccination order is generated when the gut microbiota health index is <0.4;

[0058] Furthermore, when the generated instruction is a preventive vaccination instruction, a therapeutic vaccination instruction, or an emergency vaccination instruction, the type of dysbiosis identified in step S6 is also output.

[0059] A black soldier fly larvae microbial community health status assessment and microbial agent inoculation decision-making system based on the above method includes:

[0060] The data acquisition unit is used to acquire biomarker data from the breeding environment and excrement of black soldier fly larvae;

[0061] The classification unit is used to determine the relationship between each biomarker data and the health of the microbial community, and to classify each biomarker data according to the indicator type and data level;

[0062] The gut microbiota health index calculation unit is used to normalize the data of each biomarker, obtain the normalized value of each biomarker, and input it into the linear weighted model to calculate the gut microbiota health index.

[0063] The dysregulation type determination unit is used to perform pattern matching between the normalized values ​​of each biomarker and their respective preset quantification thresholds to determine the type of dysregulation.

[0064] The microbial inoculation strategy output unit is used to output a microbial inoculation strategy based on the comparison results between the current microbial health index and its preset threshold, the trend of the microbial health index in the last N times, and the type of microbial imbalance.

[0065] Compared with the prior art, the present invention has the following advantages:

[0066] 1. This invention proposes for the first time the quantitative indicator of "microbial health index". By integrating multi-source biomarker data such as temperature, humidity, volatile organic compounds, and ATP bioluminescence value through a linear weighted model, the previously invisible state of gut microbiota is mapped to a quantitative value in the range of 0 to 1. Moreover, this index can be dynamically updated with sensor data, enabling breeders to monitor the health status of larval microbiota in real time, shifting from traditional experience-based judgment and post-event sequencing analysis to quantitative and real-time scientific assessment.

[0067] 2. Based on the pattern matching results of the normalized values ​​of various biomarkers and preset quantification thresholds, this invention can identify a variety of specific dysregulation types, overcoming the "one-size-fits-all" defects of existing technologies. It provides a clear basis for subsequent selection of microbial agents, avoiding blind addition or incorrect intervention. At the same time, it combines the microbial health index and dysregulation type to form an inoculation strategy, changing the traditional extensive mode of fixed cycle or fixed dose. It can significantly improve the intervention effect while reducing the amount of microbial agents used.

[0068] 3. The system adopts a classified and hierarchical data collection design. Core data is collected in real time by automated equipment, while augmented data can be flexibly configured according to actual conditions. The system can operate normally under different data source configurations, balancing technological advancement and implementation costs. Attached Figure Description

[0069] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0070] Figure 1 This is a schematic diagram of the architecture of the black soldier fly larvae microbial community health status assessment and inoculation decision-making system in the embodiment.

[0071] Figure 2 This is a flowchart illustrating the method for assessing the health status of the microbial community of black soldier fly larvae and making decisions on inoculation with microbial agents in the embodiments.

[0072] Figure 3 This is a schematic diagram of the continuous monitoring and dynamic vaccination decision evaluation process in Scenario 3 of the embodiment. Detailed Implementation

[0073] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0074] This embodiment proposes a method for assessing the health status of the gut microbiota of black soldier fly larvae and making decisions on inoculation with microbial agents. By defining and calculating a microbiota health index, the health status of the gut microbiota of black soldier fly larvae is quantified. Based on this microbiota health index and combined with the identification of microbiota dysbiosis types, a precise microbial agent inoculation strategy is generated.

[0075] The microbial health index in this embodiment The results were obtained through the fusion of multi-source biomarker data and a linear weighted model was adopted. This model has the advantages of clear physical meaning, strong interpretability, and easy engineering implementation. The specific formula is as follows:

[0076] ;

[0077] In the formula, The normalized value is the original detection data of the i-th biomarker mapped to the interval [0,1]. is the weight coefficient of the i-th biomarker; the clip function is a general function in programming and mathematical operations, which restricts the calculation result to the interval [0,1].

[0078] Based on the above gut health indices, such as Figure 2 As shown, the method for assessing the health status of black soldier fly larvae microbial communities and making decisions on inoculation with microbial agents in this embodiment includes the following seven steps.

[0079] Step S1. Obtain biomarker data from the breeding environment and excrement of black soldier fly larvae. The biomarker data includes temperature, humidity, ammonia concentration, sulfide concentration, amine concentration, short-chain fatty acid concentration, ATP bioluminescence value, and absorbance of cellulose characteristic peak. It also includes one or more of the following: abundance of functional genes, abundance of beneficial bacteria genera, abundance of conditionally pathogenic bacteria genera, and bacterial community diversity index.

[0080] Step S2. Determine the relationship between each biomarker data and microbial health, and classify each biomarker data according to indicator type and data level, as shown in Table 1 and Table 2 below.

[0081] Table 1: Relationship between biomarker data and gut microbiota health

[0082]

[0083] Table 2: Classification of Biomarker Data by Indicator Type and Data Level

[0084]

[0085] Table 2 categorizes indicators by timeliness into real-time, semi-real-time, and lagging indicators, reflecting the update frequency and response speed of data in model calculations. Data levels are categorized by deployment necessity into core data and augmentation data, reflecting the degree of data dependence within the system and the method of data collection. Specifically, numbers 1-8 represent core data, which can be collected in real-time by automated equipment; number 9 represents augmentation data, and numbers 10-12 represent calibration data, which can be selectively deployed as needed.

[0086] The following is an explanation of the indicator types:

[0087] 1) Real-time indicators: Data is updated frequently (on a second / minute basis), with a lag time much shorter than the system decision-making cycle (typically ≥1 hour). Each model calculation uses the latest real-time values ​​for dynamic process control. These include: temperature, humidity, ammonia concentration, sulfide concentration, amine concentration, and short-chain fatty acid concentration.

[0088] 2) Semi-real-time indicators: Data update frequency is moderate (hourly / daily), with a lag time comparable to the decision-making cycle. Between two detections, the most recent detection value is used as the current input for model calculation, ensuring the continuity of model calculation. These include: ATP bioluminescence value, absorbance of cellulose characteristic peaks, and abundance of functional genes.

[0089] 3) Lag Indicators: Data update frequency is low (daily / weekly), with lag times much longer than the decision-making cycle. These indicators are not used in the calculation of real-time gut microbiota health indices, but only for model training, calibration, and phase validation. They include: abundance of beneficial bacterial genera, abundance of conditionally pathogenic bacterial genera, and gut microbiota diversity index (obtained via 16S sequencing).

[0090] The following is an explanation of the data hierarchy:

[0091] 1) Core Data: The essential basic data for system operation is collected in real time by automated equipment without manual intervention. This includes environmental parameters (temperature, humidity, ammonia), VOCs (sulfides, amines, short-chain fatty acids), ATP bioluminescence values, and absorbance of cellulose characteristic peaks. These data are the main inputs for calculating the gut microbiota health index, and their high collection frequency supports real-time dynamic control.

[0092] 2) Augmented Data: Optional supplementary data to improve assessment accuracy or serve as a model calibration benchmark. Data collection may require manual assistance. This includes functional gene abundance (qPCR detection), genus-specific abundance, and microbial diversity indices (16S sequencing). The system can still function normally using core data even when augmented data is missing; deploying augmented data can further improve assessment accuracy.

[0093] 3) Calibration data: During the system debugging phase, it is used to train the weight coefficients and establish the mapping relationship between real-time / semi-real-time indicators and the health status of the microbial community; During the system operation phase, it does not participate in the calculation of the real-time microbial community health index, ensuring the real-time response capability of the system. It is sampled and verified regularly (e.g., weekly) to evaluate the accuracy of the current microbial community health index and to fine-tune the weights based on the verification results.

[0094] Step S3. Define mapping rules based on the relationship between each biomarker data and gut microbiota health, and normalize the biomarker data according to the mapping rules to obtain the normalized values ​​of each biomarker. The mapping rules include rules for positive / negative correlation indicators and rules for indicators with appropriate intervals.

[0095] For biomarkers that are positively or negatively correlated with the bacterial community of black soldier fly larvae, their normalized values Defined as:

[0096] ;

[0097] In the formula, For the detection data of the i-th biomarker, and These are the minimum and maximum values ​​of the biomarker under healthy aquaculture conditions, determined through historical data statistics.

[0098] For indicators such as temperature and humidity that have suitable ranges, their normalized values ​​decay linearly based on the degree of deviation from the optimal value. Temperature normalized value. The calculation formula is:

[0099] ;

[0100] In the formula, T is the collected temperature value. , These are the minimum and maximum values ​​within the suitable temperature range. The optimal temperature. For example, based on literature research, the optimal temperature for the growth of black soldier fly larvae is set as... =28℃, the suitable temperature range is [20℃, 35℃], substituting into the equation, we get:

[0101] when hour, ;

[0102] when hour, .

[0103] Humidity normalized value The calculation formula is:

[0104] ;

[0105] In the formula, RH is the collected humidity value. , These are the minimum and maximum values ​​within the suitable humidity range.

[0106] The optimal humidity is defined as follows. For example, based on literature research, the optimal humidity for the growth of black soldier fly larvae is set as... =65%, the suitable humidity range is [40%, 80%], substituting, we get:

[0107] when hour, ;

[0108] when hour, .

[0109] Step S4. Determine the weighting coefficients for each biomarker, which can be done using one of the following methods:

[0110] Principal component analysis determines the weight of each indicator based on the principal component contribution rate of historical data.

[0111] The analytic hierarchy process (AHP) constructs a judgment matrix based on expert experience and calculates weights.

[0112] The experimental data training method involves collecting multiple batches of aquaculture data, simultaneously acquiring core data detection values ​​and 16S sequencing data (including microbial diversity index, beneficial bacteria abundance, and conditionally pathogenic bacteria abundance) as calibration data. Using the core data as independent variables and microbial health calibration data (such as the diversity index) as the dependent variable, multiple linear regression or machine learning methods are employed to determine the weights of each core data point. This method ensures that the real-time calculated microbial health index remains consistent with the calibration data, while the calibration data itself is not involved in the real-time calculation. For example, an example weight configuration based on the core data source is shown in Table 3 below:

[0113] Table 3: Example Weight Configuration Table Based on Core Data Source

[0114]

[0115] In Table 3, when deploying enhanced data functional gene abundance, the weights can be redistributed among all indicators, with the sum of the weights of all indicators being 1. For example, the weights of the above 8 core indicators can be reduced proportionally (e.g., multiplied by 0.9 each), and the released 10% weight can be allocated to functional gene abundance. The actual weights can be adjusted according to the specific application scenario to ensure that the sum of the weights of all indicators is 1.

[0116] Step S5. Based on the linear weighted model, input the normalized values ​​of each biomarker obtained in step S3 and the weight coefficients of each biomarker obtained in step S4 to calculate the microbial health index.

[0117] Step S6. Perform pattern matching between the normalized values ​​of each biomarker obtained in step S3 and their preset quantification thresholds to determine the type of dysbiosis.

[0118] Table 4: Preset quantification thresholds for normalized values ​​of each biomarker (determined through statistical analysis of experimental data; adjustments can be made based on different insect ages, feed types, and seasons in practical applications).

[0119]

[0120] Table 5: Types of dysbiosis and criteria for determining each type of dysbiosis

[0121]

[0122] It should be noted that without the deployment of augmentation and calibration data, the system only identifies the type of disorder based on core data, and does not determine the type of disorder such as insufficient microbial function or pathogen invasion.

[0123] Step S7. Based on the comparison results between the current microbial health index and its preset threshold, the trend of the microbial health index in the last three times, and the type of microbial imbalance obtained in step S6, generate a microbial agent inoculation strategy.

[0124] Table 6: Inoculation Decision Table

[0125]

[0126] The continuous decline in the trend described in the table refers to the three (or more) most recent records from the system. The value decreases from far to near.

[0127] It should be noted that the gut microbiota health index is a comprehensive score, while the type of gut microbiota imbalance is determined based on a single indicator threshold; the two may not be consistent. When a high level (>0.6) indicates dysbiosis, the system tends not to intervene if there is no worsening trend, allowing the microbial community to recover naturally. When the decision is to administer prophylactic, therapeutic, or emergency vaccination, the system simultaneously outputs the identified type of microbial dysbiosis (e.g., excessive proliferation of putrefactive bacteria, overall low levels of beneficial bacteria, insufficient microbial function, pathogen invasion, complex dysbiosis, etc.) for reference by the external microbial agent inoculation system. The specific type and dosage of the microbial agent are determined automatically by the external microbial agent inoculation system based on the type of dysbiosis.

[0128] The following are practical application examples.

[0129] This system was deployed in a black soldier fly breeding workshop. Each breeding box measures 1m x 2m, and approximately 20,000 larvae were initially introduced. The system configuration is as follows:

[0130]

[0131] Scenario 1: Microbial health assessment using only core data

[0132] This scenario demonstrates the operation of using core data and deploying automated equipment (environmental sensors, VOCs sensor array, and automatic insect feces sampling and analysis module), without configuring a gene detection module, and employing a linear weighted model for microbial health assessment.

[0133] Steps S1-S3. Calculate the biomarker detection values ​​and normalized values ​​(rounded to two decimal places, the same below) according to the normalization calculation formulas of the aforementioned biomarkers.

[0134]

[0135] Step S4. Based on the contribution of each indicator to the health of the gut microbiota, the weight coefficients are determined using the analytic hierarchy process (AHP). The weights of the eight indicators collected in this scenario are determined as follows:

[0136] ;

[0137] Step S5. Substitute the normalized values ​​of each biomarker and its corresponding weight values ​​in this scenario into the gut microbiota health index calculation model:

[0138] ;

[0139] Step S6. Identification of dysbiosis type:

[0140] sulfides

[0141] amines

[0142] This satisfies the conditions for excessive proliferation of putrefactive bacteria;

[0143] Short-chain fatty acids

[0144] ATP The overall level of beneficial bacteria is not met;

[0145] Absorbance of characteristic peaks of cellulose This does not meet the condition of insufficient cellulose degradation function.

[0146] The overall assessment is: excessive proliferation of putrefactive bacteria.

[0147] Step S7. Vaccination Decision:

[0148] =0.53, falling within the 0.4~0.6 range, indicating a therapeutic vaccination level; the disorder type is excessive proliferation of putrefactive bacteria.

[0149] Output results:

[0150] Vaccination decision: therapeutic vaccination; dysregulation type: excessive proliferation of putrefactive bacteria.

[0151] Scenario 2: Microbial health assessment using core and augmented data

[0152] This scenario demonstrates how, after adding gene testing enhancement data to the core data, a linear weighted model is used for evaluation.

[0153] Steps S1-S3. Biomarker detection values ​​and normalized values ​​are shown in the table below:

[0154]

[0155] Step S4. The core data in this scenario has the same weighting coefficients as in Scenario 1.

[0156] Step S5. Normalized values ​​of each biomarker Substituting the weighting coefficients into the gut microbiota health index calculation model, the weighted sum of the core data in this scenario is calculated:

[0157]

[0158] Functional gene abundance Correction (optional enhancement of data in calculation): Functional gene abundance is added as the 9th indicator to the model, with a weight of 0.10. The total weight of the core data is adjusted to 0.90, i.e.:

[0159]

[0160] Step S6. Identification of dysbiosis type:

[0161] sulfides amines The conditions for excessive proliferation of putrefactive bacteria are not met.

[0162] Short-chain fatty acids ATP The overall level of beneficial bacteria is not met;

[0163] Absorbance of characteristic peaks of cellulose This does not meet the condition of insufficient cellulose degradation function;

[0164] Functional gene abundance This satisfies the condition of insufficient bacterial community function.

[0165] The overall assessment is: insufficient gut microbiota function.

[0166] Step S7. Vaccination Decision Making:

[0167] =0.75, falling within the 0.6~0.8 range, based on historical monitoring data, The level has decreased twice consecutively, and the vaccination level is preventive vaccination; the disorder type is insufficient gut microbiota function.

[0168] Output results:

[0169] Vaccination decision: Prophylactic vaccination; Dysfunction type: Insufficient gut microbiota function.

[0170] Scenario 3: Continuous monitoring and dynamic vaccination decision evaluation

[0171] This scenario demonstration system uses core data to continuously monitor the data within 24 hours according to the actual data collection frequency. Temperature, humidity, and ammonia concentration are automatically collected every 5 minutes; sulfide concentration, amine concentration, and short-chain fatty acid concentration are automatically collected every 30 minutes; ATP bioluminescence value and cellulose characteristic peak absorbance are collected every 4 hours. The system performs an imbalance type determination and inoculation decision every 4 hours and outputs the results. Monitoring data at non-decision times are used as process data display.

[0172] The process for this scenario can be found in the appendix. Figure 3 , Figure 3 The real-time cycle refers to the detection cycle of temperature, humidity, and ammonia concentration (5 minutes); the semi-real-time cycle 1 refers to the detection cycle of the absorbance of the characteristic peak of cellulose (4 hours); and the semi-real-time cycle 2 refers to the detection cycle of sulfide concentration, amine concentration, and short-chain fatty acid concentration (30 minutes).

[0173] Steps S1-S2. The raw detection data of biomarkers are shown in the table below:

[0174]

[0175] Steps S3-S5. Calculate the values ​​of each biomarker and the gut microbiota health index according to the normalized calculation formulas for the aforementioned biomarkers. The core data in this scenario have the same weighting coefficients as in Scenario 1. The normalized values ​​of each biomarker and the gut microbiota health index at different times are shown in the table below:

[0176]

[0177] Note: In the above table Represents the normalized value of temperature. The normalized value representing humidity. The normalized value representing the ammonia concentration. The normalized value representing the sulfide concentration. The normalized value representing the concentration of amines. Normalized values ​​representing the concentration of short-chain fatty acids. The normalized value representing the bioluminescence value of ATP. The normalized value represents the absorbance of the characteristic peak of cellulose, and the same applies below.

[0178] Step S6. According to the aforementioned rules for determining the types of dysbiosis, the types of dysbiosis at each time point are shown in the table below.

[0179]

[0180] Note: In the table above, “√” indicates that the threshold condition is met; “×” indicates that the threshold condition is not met.

[0181] Step S7. In accordance with the aforementioned inoculation strategy, the inoculation decision for this scenario is shown in the table below.

[0182]

[0183] After determining the type of dysbiosis and making an inoculation decision at each time point, the results will be output.

[0184] Scenario 1 through Scenario 3, through progressive examples, verify the feasibility and effectiveness of the method in this embodiment under different data configurations and monitoring conditions: Scenario 1, relying solely on core data, completes health index calculation, disorder identification, and inoculation decision-making, demonstrating the system's basic operational capabilities and low-cost deployment advantages; Scenario 2, by introducing augmented data, achieves more accurate diagnosis and decision adjustments, showcasing the system's scalability and potential for accuracy improvement; Scenario 3, by continuously and dynamically monitoring and capturing the changing trends of the microbial community state, verifies the system's support capabilities for early warning and tiered decision-making. These three scenarios together constitute a complete verification chain from static to dynamic, from basic to augmented, comprehensively demonstrating the engineering practicality, adaptability, and reliability of this invention in black soldier fly larval rearing.

[0185] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0186] Based on the above method, this embodiment also proposes a black soldier fly larvae microbial community health status assessment and microbial agent inoculation decision system, including:

[0187] The data acquisition unit is used to acquire biomarker data from the breeding environment and excrement of black soldier fly larvae;

[0188] The classification unit is used to determine the relationship between each biomarker data and the health of the microbial community, and to classify each biomarker data according to the indicator type and data level;

[0189] The gut microbiota health index calculation unit is used to normalize the data of each biomarker, obtain the normalized value of each biomarker, and input it into the linear weighted model to calculate the gut microbiota health index.

[0190] The dysregulation type determination unit is used to perform pattern matching between the normalized values ​​of each biomarker and their respective preset quantification thresholds to determine the type of dysregulation.

[0191] The microbial inoculation strategy output unit is used to output a microbial inoculation strategy based on the comparison results between the current microbial health index and its preset threshold, the trend of the microbial health index in the last N times, and the type of microbial imbalance.

[0192] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification. Furthermore, the above embodiments only illustrate several implementation methods of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the present invention. For those skilled in the art, several modifications and improvements can be made without departing from the concept of the present invention, and these all fall within the protection scope of the present invention.

Claims

1. A method for assessing the health status of the microbial community of black soldier fly larvae and making decisions on inoculation with microbial agents, characterized in that, Includes the following steps: S1. Obtain biomarker data from the rearing environment and excrement of black soldier fly larvae; S2. Determine the relationship between each biomarker data and gut health, and classify each biomarker data according to indicator type and data level; S3. Define mapping rules based on the relationship between each biomarker data and microbial health, and normalize each biomarker data according to the mapping rules to obtain the normalized value of each biomarker. S4. Determine the weighting coefficients for each biomarker; S5. Based on the linear weighted model, input the normalized values ​​of each biomarker obtained in step S3 and the weight coefficients of each biomarker obtained in step S4, and calculate the microbial health index. S6. Perform pattern matching between the normalized values ​​of each biomarker obtained in step S3 and their preset quantification thresholds to determine the type of dysbiosis. S7. Based on the comparison results between the current microbial health index and its preset threshold, the trend of the microbial health index in the last N times, and the type of microbial imbalance obtained in step S6, generate a microbial agent inoculation strategy.

2. The method for assessing the health status of black soldier fly larvae microbial communities and making decisions on inoculation with microbial agents as described in claim 1, characterized in that, In step S1, the biomarker data includes temperature, humidity, ammonia concentration, sulfide concentration, amine concentration, short-chain fatty acid concentration, ATP bioluminescence value, and absorbance of cellulose characteristic peak.

3. The method for assessing the health status of black soldier fly larvae microbial communities and making decisions on inoculation with microbial agents as described in claim 2, characterized in that, In step S1, the biomarker data may also include one or more of the following: functional gene abundance, beneficial bacterial genera abundance, conditionally pathogenic bacterial genera abundance, and microbial diversity index.

4. The method for assessing the health status of black soldier fly larvae microbial communities and making decisions on inoculation with microbial agents as described in claim 3, characterized in that, The relationship between the biomarker data and gut health includes: Indicators with suitable ranges: temperature and humidity; Negative correlation indicators: ammonia concentration, sulfide concentration, amine concentration, absorbance of cellulose characteristic peak, and abundance of opportunistic pathogens; Positively correlated indicators: short-chain fatty acid concentration, ATP bioluminescence value, abundance of functional genes, abundance of beneficial bacteria genera, and bacterial community diversity index; The biomarker data are categorized by indicator type and data level as follows: Real-time indicators and core data: temperature, humidity, ammonia concentration, sulfide concentration, amine concentration, and short-chain fatty acid concentration; Semi-real-time indicators and core data: ATP bioluminescence value and absorbance of cellulose characteristic peak; Semi-real-time indicator and augmented data: functional gene abundance; Lagging indicators and calibrated data: abundance of beneficial bacteria genera, abundance of conditionally pathogenic bacteria genera, and bacterial community diversity index; The real-time and semi-real-time indicators are used in the calculation of the microbial health index, while the lag indicators are only used for model training and validation.

5. The method for assessing the health status of black soldier fly larvae microbial communities and making decisions on inoculation with microbial agents as described in claim 4, characterized in that, In step S3, the mapping rules include: For biomarkers that are positively or negatively correlated with the bacterial community of black soldier fly larvae, their normalized values Defined as: ； In the formula, For the detection data of the i-th biomarker, and These are the minimum and maximum values ​​of the biomarker under healthy aquaculture conditions, determined through historical data statistics; Temperature normalized value The calculation formula is: ； In the formula, T is the collected temperature value. , These are the minimum and maximum values ​​within the suitable temperature range. The optimal temperature; Humidity normalized value The calculation formula is: ； In the formula, RH is the collected humidity value. , These are the minimum and maximum values ​​within the suitable humidity range. The optimal humidity level.

6. The method for assessing the health status of black soldier fly larvae microbial communities and making decisions on inoculation with microbial agents as described in claim 1, characterized in that, The weight coefficients of each biomarker can be determined by one of the following methods: principal component analysis, analytic hierarchy process, multiple linear regression based on historical data and microbial health reference standards, or machine learning training.

7. The method for assessing the health status of black soldier fly larvae microbial communities and making decisions on inoculation with microbial agents as described in claim 1, characterized in that, The expression for the linear weighted model is as follows: ； In the formula, Given its corresponding weight coefficient, the `clip` function is used to restrict the calculation result to the interval [0,1]. This indicates the health index of the gut microbiota.

8. The method for assessing the health status of black soldier fly larvae microbial communities and making decisions on inoculation with microbial agents as described in claim 4, characterized in that, In step S6, the types of dysbiosis and the criteria for determining each type of dysbiosis are as follows: When the normalized value of sulfides is ≤0.2 and the normalized value of amines is ≤0.3, it is determined to be an excessive proliferation of putrefactive bacteria. When the normalized value of the absorbance of the characteristic peak of cellulose is ≤0.3, it is determined that the cellulose degradation function is insufficient. When the normalized value of short-chain fatty acids is ≤0.4 and the normalized value of ATP bioluminescence is ≤0.4, it is determined that the overall level of beneficial bacteria is low. When the normalized value of functional gene abundance is ≤0.5, it is judged as insufficient bacterial community function; When the normalized value of the abundance of conditional pathogens is ≤0.2 and the normalized value of the community diversity index is ≤0.5, it is determined to be an invasion of pathogens. When multiple gut microbiota imbalance criteria are met, it is classified as a complex imbalance.

9. The method for assessing the health status of black soldier fly larvae microbial communities and making decisions on inoculation with microbial agents as described in claim 4, characterized in that, In step S7, the inoculation strategy for generating the microbial agent and the criteria for determining each inoculation strategy are as follows: When the gut microbiota health index is >0.8, a no-vaccination instruction is generated; When 0.6 < microbial health index ≤ 0.8 and the microbial health index shows a continuous downward trend over the last N times, a preventive vaccination instruction is generated, where N ≥ 3; When 0.6 < microbial health index ≤ 0.8 and the microbial health index shows a non-continuous downward trend in the last N times, a non-vaccination instruction is generated; When the gut microbiota health index is 0.4 ≤ 0.6, a therapeutic inoculation instruction is generated; An emergency vaccination order is generated when the gut microbiota health index is <0.4; Furthermore, when the generated instruction is a preventive vaccination instruction, a therapeutic vaccination instruction, or an emergency vaccination instruction, the type of dysbiosis identified in step S6 is also output.

10. A system for assessing the health status of black soldier fly larvae microbial communities and making decisions on inoculation with microbial agents based on the method described in any one of claims 1 to 9, characterized in that, include: The data acquisition unit is used to acquire biomarker data from the breeding environment and excrement of black soldier fly larvae; The classification unit is used to determine the relationship between each biomarker data and the health of the microbial community, and to classify each biomarker data according to the indicator type and data level; The gut microbiota health index calculation unit is used to normalize the data of each biomarker, obtain the normalized value of each biomarker, and input it into the linear weighted model to calculate the gut microbiota health index. The dysregulation type determination unit is used to perform pattern matching between the normalized values ​​of each biomarker and their respective preset quantification thresholds to determine the type of dysregulation. The microbial inoculation strategy output unit is used to output a microbial inoculation strategy based on the comparison results between the current microbial health index and its preset threshold, the trend of the microbial health index in the last N times, and the type of microbial imbalance.

Images