Soil health assessment methods and systems based on microbial communities

Abstract

This application discloses a method and system for soil health assessment based on microbial communities, relating to the field of soil health technology. This application integrates the network structure characteristics of microbial communities with basic soil physicochemical environmental information. By calculating network resilience and functional redundancy indices, it can deeply quantify the intrinsic stability and potential for resistance to disturbance of microbial communities from two key dimensions: species interaction stability and functional backup capacity. This enables precise diagnosis and effective tracing of soil health status. The resulting health index provides an intuitive and comprehensive health scale, facilitating the classification and comparison of different soil types and allowing for the tracing of dominant factors leading to a decline in health levels. This provides a clear scientific basis for developing precise and targeted soil restoration and management measures, and has significant application value for promoting sustainable agricultural development, ecological restoration projects, and land resource management.

Description

Technical Field

[0001] This application relates to the field of soil health technology, specifically to soil health assessment methods and systems based on microbial communities. Background Technology

[0002] Soil health is the cornerstone of maintaining ecosystem function and agricultural productivity. Traditional soil health assessments mainly rely on a series of physical and chemical properties (such as pH, organic matter, and nitrogen, phosphorus, and potassium content). While these methods can reflect basic soil fertility and environmental conditions, they neglect the core role of the vast and complex microbial community within the soil. Developing a comprehensive assessment method that integrates soil physicochemical environment with the structure and function of the microbial network has become an urgent need in the fields of precision agriculture and ecological management.

[0003] Existing technology, such as the invention application patent with publication number CN115293583A, discloses a soil health assessment system and method, relating to the field of soil health assessment technology. The system includes: an indicator acquisition module for acquiring indicator data of the target soil; the indicator data includes soil physicochemical components, indicative soil organisms, and crop yield; a soil health index determination module for inputting the indicator data into a soil health assessment comprehensive function, and then outputting the soil health index of the target soil through the comprehensive function; the comprehensive function includes an indicator scoring equation; the indicator scoring equation is determined based on measured indicator data and corresponding indicator scores; and a soil health level determination module for determining the soil health level of the target soil based on the soil health index and a preset threshold. This invention can unify soil health evaluation indicators, thereby improving the applicability and accuracy of soil health assessment.

[0004] Regarding the above-mentioned solutions, the inventors of this application have found that the above-mentioned technologies have at least the following technical problems: 1. Currently, there is a lack of identification of potential mutually beneficial, competitive or symbiotic effects through correlation, revealing the internal correlation structure of microbial communities, which cannot make up for the shortcomings of traditional diversity indices in reflecting species interactions; it cannot enhance the ecological explanatory power, cannot intuitively display topological features such as key species and module composition, and cannot help identify community assembly mechanisms and environmental response patterns.

[0005] 2. Currently, there is a lack of network topology analysis, no calculation of resilience through simulated node removal, and no assessment of functional redundancy index in conjunction with modular analysis; there is no quantitative assessment of community stability, no ability to simulate species loss scenarios, no direct reflection of the network structure's resistance to external disturbances, and no provision of new indicators for ecosystem resilience assessment.

[0006] 3. Currently, the lack of analysis of deviations in indicators makes it impossible to identify the root causes of health problems and guide targeted interventions; it does not support dynamic monitoring and prediction, and the lack of long-term data accumulation makes it impossible to reveal the trajectory of health evolution and changes in dominant factors, thus failing to provide trend warnings and a basis for strategy adjustments for soil restoration and protection. The absence of multi-dimensional information fusion fails to overcome the limitations of single-dimensional evaluation and cannot comprehensively characterize soil health; it cannot improve the ecological relevance of the index, lacks interaction factors, cannot reflect the coupling relationship between "microorganisms and the environment," and cannot make the index more closely reflect real ecological processes. Summary of the Invention

[0007] To address the aforementioned technical shortcomings, the purpose of this application is to provide a method and system for soil health assessment based on microbial communities.

[0008] To solve the above-mentioned technical problems, this application adopts the following technical solution: In the first aspect, this application provides a soil health assessment method based on microbial communities, which includes the following steps: S1, collecting target soil samples, obtaining microbial species composition and abundance data, and generating a standardized species abundance matrix; at the same time, obtaining preset key physicochemical property data of the soil, and generating a normalized physicochemical property vector.

[0009] S2. Based on the standardized species abundance matrix, the correlation between various microbial species is analyzed, and a microbial species co-occurrence network is constructed.

[0010] S3. Based on the microbial co-occurrence network, a topology analysis was performed to calculate the resilience and functional redundancy index of the microbial co-occurrence network.

[0011] S4. Based on the normalized physicochemical property vector, the comprehensive score of the soil's physicochemical properties is obtained through analysis.

[0012] S5. Based on the comprehensive scores of network resilience, functional redundancy index and physicochemical properties, analyze the deviations; then generate the soil microbial health index, and conduct soil health status assessment and source tracing.

[0013] Preferably, the step of obtaining microbial species composition and abundance data and generating a standardized species abundance matrix includes: collecting target soil samples, sequencing the target soil samples based on high-throughput sequencing technology to obtain original species abundance data; and standardizing the original species abundance data based on a preset standardization method to obtain a standardized species abundance matrix.

[0014] Preferably, the step of simultaneously acquiring preset key physicochemical property data of the soil and generating a normalized physicochemical property vector includes: acquiring preset key physicochemical property data of the target soil to obtain original physicochemical property data; and performing normalization processing on the original physicochemical property data based on a preset normalization algorithm to obtain a normalized physicochemical property vector.

[0015] Preferably, the step of analyzing the correlation between microbial species based on the standardized species abundance matrix and constructing a microbial species co-occurrence network includes: pairing each microbial species into pairs and recording them as microbial species pairs; and then using Spearman correlation analysis based on the standardized species abundance matrix to calculate the correlation between each microbial species pair, so as to obtain the correlation coefficient and corresponding significance value of each microbial species pair.

[0016] Based on the correlation coefficients and corresponding significance values ​​of each microbial species pair, the microbial species pairs are screened for significance to obtain a set of significantly related microbial species pairs; then, a microbial co-occurrence network is constructed to obtain a microbial co-occurrence network with microbial species as nodes and significant correlation relationships as edges.

[0017] Preferably, the step of using Spearman correlation analysis to calculate the correlation coefficient and corresponding significance value of each microbial species pair includes: calculating using the formula... The correlation coefficients of microbial species pairs were obtained. , Represented as the first The difference in abundance rank between two microbial species in a soil sample. This is represented by the number corresponding to the soil sample. , This is expressed as the total number of soil samples;

[0018] Through calculation formula Derive the test statistic Furthermore, in a degree of freedom of of Extract the significance value corresponding to the test statistic from the distribution.

[0019] Preferably, the topology analysis based on the microbial co-occurrence network to calculate the resilience and functional redundancy index of the microbial co-occurrence network includes: performing a random node removal simulation operation based on the microbial co-occurrence network and a preset random removal ratio to obtain several simulated network connectivity values, and performing a resilience quantification calculation operation to obtain the resilience of the microbial co-occurrence network.

[0020] Based on the microbial co-occurrence network, a modular structure analysis operation is performed to obtain the module partitioning results and modularity metric; based on the module partitioning results and the modularity metric, a functional redundancy index calculation operation is performed to obtain the functional redundancy index.

[0021] Preferably, the step of analyzing and obtaining the comprehensive score of soil physicochemical properties based on the normalized physicochemical property vector includes: obtaining the preset weight vector corresponding to each physicochemical property index in the normalized physicochemical property vector based on the preset weight allocation rules for each physicochemical property; and obtaining the comprehensive score of soil physicochemical properties by weighted summation based on the normalized physicochemical property vector and the weight vector.

[0022] Preferably, the step of analyzing deviations based on the network resilience, functional redundancy index, and comprehensive physicochemical property score includes: performing deviation judgment based on the values ​​of each of the network resilience, functional redundancy index, and comprehensive physicochemical property score, as well as preset optimal value ranges for each of the network resilience, functional redundancy index, and comprehensive physicochemical property score, to obtain a deviation judgment result; performing a branch decision operation based on the deviation judgment result; when the deviation judgment result indicates that the network resilience, functional redundancy index, or comprehensive physicochemical property score deviates from its corresponding optimal value range, performing in-depth source tracing analysis on the deviation index; generating and outputting a source tracing analysis report based on the result of the source tracing analysis operation; and triggering the generation of the soil microbial health index when none of them deviate from their respective optimal value ranges.

[0023] Preferably, the step of generating a soil microbial health index and assessing and tracing the soil health status includes: weighting and fusing the network resilience, the functional redundancy index, and the comprehensive score of the physicochemical properties through a preset health index fusion model and introducing interaction terms to obtain the soil microbial health index; and comparing and judging the soil microbial health index with a preset health level threshold range to obtain the soil health level judgment result.

[0024] In its second aspect, this application provides a system for soil health assessment based on microbial communities, comprising: preferably, a data generation module for collecting target soil samples, obtaining microbial species composition and abundance data, and generating a standardized species abundance matrix; and simultaneously obtaining preset key physicochemical property data of the soil, and generating a normalized physicochemical property vector.

[0025] The microbial species co-occurrence network construction module analyzes the correlation between various microbial species based on the standardized species abundance matrix and constructs a microbial species co-occurrence network.

[0026] The microbial species co-occurrence network analysis module performs topological structure analysis based on the microbial co-occurrence network and calculates the resilience and functional redundancy index of the microbial co-occurrence network.

[0027] The physical and chemical properties comprehensive score calculation module analyzes and obtains the comprehensive score of the soil's physical and chemical properties based on the normalized physical and chemical property vector.

[0028] The soil health status assessment and tracing module analyzes deviations based on the comprehensive scores of network resilience, functional redundancy index, and physicochemical properties; then it generates a soil microbial health index and conducts soil health status assessment and tracing.

[0029] The beneficial effects of this application are as follows: 1. The soil health assessment method and system based on microbial communities provided in this application integrate the network structure characteristics of microbial communities with basic soil physicochemical environmental information. By calculating network resilience and functional redundancy indices, it is possible to deeply quantify the intrinsic stability and potential for resisting disturbance of microbial communities from two key dimensions: species interaction stability and functional backup capacity; it achieves accurate diagnosis and effective source tracing of soil health status; the final generated health index provides an intuitive and comprehensive health scale, facilitating the classification and comparison of different soil grades, and allowing for the tracing of the dominant factors leading to a decline in health levels, thus providing a clear scientific basis for formulating precise and targeted soil restoration and management measures; it has significant application value for promoting sustainable agricultural development, ecological restoration projects, and land resource management.

[0030] 2. This application combines high-throughput sequencing technology with chemical analysis to simultaneously acquire microbial composition and key physicochemical properties data, and performs standardization and normalization processing respectively; this improves the comparability and integration of data, the standardization process eliminates the impact of sequencing depth differences on species abundance, and the normalization process solves the problem of inconsistent dimensions of different physicochemical indicators, laying the foundation for subsequent multi-source data fusion; it enhances the reliability of analytical results, the preprocessing reduces technical errors and extreme value interference, making microbial network construction and environmental assessment more representative; and it improves the applicability and scalability of the method, the workflow can flexibly adjust the target physicochemical properties and standardization methods to adapt to different types of soil or research objectives, providing a standardized data foundation for large-scale soil health monitoring.

[0031] 3. This application reveals the intrinsic interconnected structure of microbial communities, identifies potential mutualistic, competitive, or symbiotic interactions through correlations, and compensates for the inadequacy of traditional diversity indices in reflecting species interactions; it enhances ecological explanatory power, as the network structure can intuitively display topological features such as key species and module composition, which helps to identify community assembly mechanisms and environmental response patterns; it improves noise resistance and stability, eliminates spurious associations through saliency screening, reduces the interference of random errors on the network structure, makes the constructed network more biologically meaningful, and provides a reliable basis for subsequent stability and functional analysis.

[0032] 4. This application introduces network topology analysis, calculates resilience by simulating node removal, and evaluates functional redundancy index by combining modular analysis; it quantitatively assesses community stability, simulates species loss scenarios, and directly reflects the network structure's resistance to external disturbances, providing a new indicator for ecosystem resilience assessment; it quantifies the community's functional backup potential based on the similarity of species functions within modules, bridging the research gap between structural redundancy and functional redundancy; and it provides early warning indicators. Network structure indicators are sensitive to environmental changes and can reflect system vulnerability before species loss, providing a forward-looking basis for soil health management.

[0033] 5. The application enables precise source tracing of problems. By analyzing deviations in indicators, the main causes of health issues (such as structural fragility, insufficient redundancy, or environmental degradation) can be identified, guiding targeted interventions. It supports dynamic monitoring and prediction; long-term data accumulation can reveal the trajectory of health evolution and changes in dominant factors, providing trend warnings and a basis for strategy adjustment for soil restoration and protection. It achieves multi-dimensional information fusion, overcoming the limitations of single-dimensional evaluation and comprehensively characterizing soil health. It enhances the ecological relevance of the index; interaction terms reflect the coupling relationship between microorganisms and the environment, making the index closer to real ecological processes. It strengthens the sensitivity and discriminative power of the assessment; weighted fusion highlights the contribution of key indicators, making the health index more responsive to changes in different soil conditions, suitable for gradient comparison and long-term monitoring. Attached Figure Description

[0034] To more clearly illustrate the technical solutions in the embodiments of this application 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 some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0035] Figure 1 This is a flowchart illustrating the implementation steps of the method described in this application.

[0036] Figure 2 This is a schematic diagram of the system structure connection of this application. Detailed Implementation

[0037] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0038] Please see Figure 1 As shown, this application provides a soil health assessment method based on microbial communities in the first aspect, including: S1, collecting target soil samples, obtaining microbial species composition and abundance data, and generating a standardized species abundance matrix; and simultaneously obtaining preset key physicochemical property data of the soil and generating a normalized physicochemical property vector.

[0039] In a specific example, the step of acquiring microbial species composition and abundance data and generating a standardized species abundance matrix includes: collecting target soil samples, sequencing the target soil samples using high-throughput sequencing technology to obtain raw species abundance data; and standardizing the raw species abundance data using a preset standardization method to obtain a standardized species abundance matrix.

[0040] It should be noted that the high-throughput sequencing technology involves amplifying and sequencing the total microbial DNA extracted from the target soil sample. By comparing and classifying the sequenced gene sequences, the number of sequence reads of each microbial species (such as OTUs, operational taxonomic units) in the sample is counted, generating the original species abundance data. This data can comprehensively reflect the species composition and relative abundance information of the microbial community.

[0041] It should be noted that the original species abundance data is a two-dimensional matrix, where the rows represent different soil samples and the columns represent different microbial species (such as OTUs). Each element in the matrix represents the number of sequence reads of a species in a sample, which is used to quantitatively describe the relative abundance of the species in the sample and forms the basis for subsequent community structure analysis.

[0042] It should be noted that the standardization process described above uses a summation standardization method to process the original species abundance data of each sample. The specific process is as follows: First, the total number of sequence reads of all microbial species in each soil sample is calculated; then, the original number of reads of each microbial species in each soil sample is divided by the total number of reads in that sample; finally, the relative abundance percentage of each microbial species in that soil sample is obtained to eliminate the incomparability of data between soil samples caused by different sequencing depths.

[0043] It should be noted that the standardized species abundance matrix is ​​a two-dimensional matrix obtained after summation and standardization. Its rows represent soil samples and columns represent microbial species. Each element in the matrix is ​​the relative abundance of the microbial species in the soil sample (between 0 and 1). This standardized species abundance matrix eliminates the differences in sequencing depth between soil samples, making the microbial composition of different samples comparable. It is a key input for subsequent Spearman correlation analysis and construction of co-occurrence networks.

[0044] In a specific example, the step of simultaneously acquiring preset key physicochemical property data of the soil and generating a normalized physicochemical property vector includes: acquiring preset key physicochemical property data of the target soil to obtain raw physicochemical property data; and performing normalization processing on the raw physicochemical property data based on a preset normalization algorithm to obtain a normalized physicochemical property vector.

[0045] It should be noted that the preset key physicochemical property data, including soil pH, organic matter content, total nitrogen, available phosphorus, available potassium, and water content—data that directly affect the structure and function of the microbial community—are obtained through laboratory chemical analysis or on-site sensor measurements of the target soil samples. The raw physicochemical property data is a two-dimensional matrix, where rows represent soil samples and columns represent different physicochemical property indicators. Each element in the matrix is ​​the raw measurement value of a specific physicochemical property in a particular sample. This data quantifies the physical and chemical environmental conditions of the soil.

[0046] It should be noted that the normalization process employs a maximum-minimum normalization method to process the data in each column (i.e., each physicochemical property) of the original physicochemical property data, mapping physicochemical property data with different dimensions and numerical ranges to a unified interval of [0, 1]. Furthermore, the implementation process of the normalization algorithm is as follows: For each physicochemical property index, firstly, all soil samples are traversed to find the maximum and minimum values ​​of a certain physicochemical property index across all soil samples; then, based on the original values ​​of each soil sample for that physicochemical property index, the difference between the original value and the minimum value is calculated; finally, this difference is divided by the difference between the maximum and minimum values ​​to obtain the normalized value of that physicochemical property index; and thus, the normalized values ​​of each physicochemical property index are obtained. This process linearly scales the index data of all samples to between 0 and 1, eliminating the influence of differences in dimensions and absolute values ​​between different physicochemical properties.

[0047] It should be noted that the normalized physicochemical property vector refers to a one-dimensional vector composed of the values ​​of all preset key physicochemical property indicators of a soil sample after normalization. The normalized physicochemical property vector comprehensively represents the soil environmental conditions of the soil sample after dimensionless processing.

[0048] This application combines high-throughput sequencing technology with chemical analysis to simultaneously acquire microbial composition and key physicochemical properties data, which are then standardized and normalized. This enhances data comparability and integration; standardization eliminates the impact of sequencing depth differences on species abundance, and normalization resolves the issue of inconsistent dimensions of different physicochemical indicators, laying the foundation for subsequent multi-source data fusion. It also improves the reliability of analytical results; the preprocessing reduces technical errors and extreme value interference, making microbial network construction and environmental assessment more representative. Furthermore, it enhances the applicability and scalability of the method; the workflow allows for flexible adjustment of target physicochemical properties and standardization methods to adapt to different soil types or research objectives, providing a standardized data foundation for large-scale soil health monitoring.

[0049] S2. Based on the standardized species abundance matrix, the correlation between various microbial species is analyzed, and a microbial species co-occurrence network is constructed.

[0050] In a specific example, the step of analyzing the correlation between microbial species based on the standardized species abundance matrix and constructing a microbial species co-occurrence network includes: pairing each microbial species into pairs and recording them as microbial species pairs; and then using Spearman correlation analysis based on the standardized species abundance matrix to calculate the correlation between each microbial species pair, so as to obtain the correlation coefficient and corresponding significance value of each microbial species pair.

[0051] Based on the correlation coefficients and corresponding significance values ​​of each microbial species pair, the microbial species pairs are screened for significance to obtain a set of significantly related microbial species pairs; then, a microbial co-occurrence network is constructed to obtain a microbial co-occurrence network with microbial species as nodes and significant correlation relationships as edges.

[0052] It should be noted that the set of significantly correlated microbial species pairs is a data structure that records all microbial species pairs that pass the significance test and their corresponding correlation coefficients.

[0053] It should be noted that the construction of the microbial co-occurrence network is based on the set of significantly related species pairs. Each unique microbial species is abstracted as a node, and each significantly related relationship (whether positive or negative) is abstracted as an edge connecting the two nodes. Further, this operation generates a network graph data structure, where the set of nodes represents the microbial species members participating in the significantly co-occurring relationships, and the set of edges represents the statistically validated significant associations between these species. This network intuitively represents the structured association patterns within the microbial community.

[0054] It should be noted that the aforementioned microbial co-occurrence network is a graph model composed of nodes and edges (visualizing and structuring complex multidimensional species association data into a network topology). In this network, nodes represent microbial species (such as OTUs) and are the basic building blocks of the network; edges represent statistically validated significant correlations between microbial species, and connecting edges indicates that two microbial species exhibit statistically significant patterns of synergistic abundance changes in multiple soil samples. This network serves as the foundational input data for subsequent topological structure analysis.

[0055] In a specific example, the calculation of the correlation between microbial species pairs using Spearman correlation analysis to obtain the correlation coefficient and corresponding significance value of each microbial species pair includes: calculating using the formula... The correlation coefficients of microbial species pairs were obtained. , Represented as the first The difference in abundance rank between two microbial species in a soil sample. This is represented by the number corresponding to the soil sample. , This represents the total number of soil samples.

[0056] Through calculation formula Derive the test statistic Furthermore, in a degree of freedom of of Extract the significance value corresponding to the test statistic from the distribution.

[0057] It should be noted that Spearman correlation analysis is an operation that calculates nonparametric statistical correlation based on the abundance vectors of each microbial species pair in the standardized species abundance matrix. This operation first converts the relative abundance value of each species in all samples into a rank, i.e., its position number after being arranged in ascending order, generating a species abundance rank vector. Then, based on the abundance rank vectors of paired species, the Spearman rank correlation coefficient is calculated. This coefficient quantifies the degree of consistency in the abundance trends of two species; its value range is [-1, 1], with positive values ​​indicating positive correlation and negative values ​​indicating negative correlation. A larger absolute value indicates a stronger correlation. Degrees of freedom Essentially, the degrees of freedom represent the number of independent information items contained in the soil sample data when estimating the overall correlation coefficient. In a model that calculates the correlation coefficient based on two variables (i.e., the abundance vectors of two microbial species) and performs a t-test, the degrees of freedom are calculated as the number of soil samples. Subtract 2.

[0058] It should be noted that the correlation coefficient is an indicator quantifying whether the abundance trends of two microbial species are synchronized (revealing potential ecological relationships such as symbiosis, competition, or non-interference between species). The significance value is a probability value that measures whether the observed correlation is statistically significant, used to determine whether the inter-species association is significant rather than accidental; the smaller the significance value, the lower the probability that the observed correlation is caused by random error, thus providing more reason to reject the null hypothesis and determine that the correlation is statistically significant. Further, the significance screening is based on comparing the significance value with a preset significance threshold. This operation determines whether the calculated significance value for each microbial species pair is less than the preset threshold; if it is less, the association between the microbial species pair is determined to be statistically significant, and it is included in the set of significantly related species pairs; otherwise, it is excluded. This operation filters out those inter-species relationships with statistical significance, eliminating spurious associations that may be caused by noise or random fluctuations, ensuring that the relationship basis for subsequent network construction is reliable.

[0059] This application reveals the intrinsic interconnected structure of microbial communities, identifying potential mutualistic, competitive, or symbiotic interactions through correlations, thus compensating for the shortcomings of traditional diversity indices in reflecting species interactions. It enhances ecological explanatory power, as the network structure visually displays topological features such as key species and module composition, aiding in the identification of community assembly mechanisms and environmental response patterns. Furthermore, it improves noise resistance and stability by eliminating spurious associations through saliency screening, reducing the interference of random errors on the network structure, making the constructed network more biologically meaningful, and providing a reliable basis for subsequent stability and functional analyses.

[0060] S3. Based on the microbial co-occurrence network, a topology analysis was performed to calculate the resilience and functional redundancy index of the microbial co-occurrence network.

[0061] In a specific example, the topology analysis based on the microbial co-occurrence network to calculate the resilience and functional redundancy index of the microbial co-occurrence network includes: performing a random node removal simulation operation based on the microbial co-occurrence network and a preset random removal ratio to obtain several simulated network connectivity values, and performing a resilience quantification calculation operation to obtain the resilience of the microbial co-occurrence network.

[0062] Based on the microbial co-occurrence network, a modular structure analysis operation is performed to obtain the module partitioning results and modularity metric; based on the module partitioning results and the modularity metric, a functional redundancy index calculation operation is performed to obtain the functional redundancy index.

[0063] It should be noted that the random node removal simulation operation is based on the microbial co-occurrence network and simulates the impact of random species loss on the overall network structure. The operation first randomly selects and removes a corresponding proportion of nodes from the network node set according to a preset random removal ratio, and at the same time removes all edges connected to these nodes to generate a damaged network after random species loss. Then, the network connectivity index (such as global efficiency or average path length) of the damaged network is calculated. The network connectivity index quantifies the efficiency of information or material transmission between network nodes. Furthermore, the above random selection, removal and calculation process is repeated a preset number of times (e.g., 1000 times) to obtain several simulated network connectivity values ​​(reflecting the changes in network connectivity under multiple random node removal events).

[0064] It should be noted that the resilience quantification calculation operation is a quantification process that evaluates the network's ability to maintain its connectivity function against random disturbances based on the several simulated network connectivity values. Specifically, this operation uses a preset algorithm to summarize the network connectivity changes obtained from multiple simulations into a scalar value, namely the resilience of the microbial co-occurrence network (the higher the resilience value of the microbial co-occurrence network, the better its overall connectivity is maintained when facing multiple simulations of random species loss). Further, through the calculation formula... The resilience of the microbial co-occurrence network was calculated. , This represents the number corresponding to each repetition of the random node removal simulation operation. , This represents the total number of repetitions of the random node removal simulation operation. This represents the initial connectivity value of the original microbial co-occurrence network. Represented as in the first The connectivity value of the damaged network obtained after a random node removal simulation operation.

[0065] It should be noted that the modular structure analysis operation is a computational process based on the microbial co-occurrence network, identifying the internal structure of the microbial co-occurrence network and quantifying its modularity. The nodes of the microbial co-occurrence network are divided such that nodes within the same module are tightly connected, while connections between different modules are relatively sparse, thus obtaining the module division result, i.e., the module label to which each microbial co-occurrence network node belongs. Simultaneously, this operation calculates the modularity metric of the microbial co-occurrence network. The specific calculation process is as follows: multiplying by one-half of the total weight of all edges in twice the microbial co-occurrence network by the difference between the actual connection weight between each microbial co-occurrence network node and the expected connection weight under a random network, and then multiplying by the sum of the differences between pairs of microbial co-occurrence network nodes belonging to the same module, finally yields the modularity metric.

[0066] It should be noted that the functional redundancy index calculation operation is a process of quantifying the functional backup capacity within a microbial community based on the module partitioning results and the modularity metric. This operation assesses the redundancy level by analyzing the connectivity patterns of species within a module. Further, the calculation formula... Determine the functional redundancy index A higher functional redundancy index indicates a stronger potential functional backup capability of the community. This represents the module number corresponding to the microbial co-occurrence network. , This represents the total number of modules in the microbial co-occurrence network. This represents the node number corresponding to each node within a module of a microbial co-occurrence network. , This represents the total number of nodes within a module of a microbial co-occurrence network. Represented as nodes Intra-module degree (node) (Number of connections between other nodes in the same module of the microbial co-occurrence network). Represented as nodes Total degree (nodes) The number of connections to all other nodes (including those in the same module and those in different modules) in the co-occurrence network with microorganisms.

[0067] This application introduces network topology analysis to calculate resilience by simulating node removal and combines modular analysis to assess functional redundancy index; it quantitatively assesses community stability by simulating species loss scenarios, directly reflecting the network structure's resistance to external disturbances, providing a new indicator for ecosystem resilience assessment; it quantifies the community's functional backup potential based on the functional similarity of species within modules, bridging the research gap between structural and functional redundancy; and it provides early warning indicators. Network structure indicators are sensitive to environmental changes and can reflect system vulnerability before species loss, providing a forward-looking basis for soil health management.

[0068] S4. Based on the normalized physicochemical property vector, the comprehensive score of the soil's physicochemical properties is obtained through analysis.

[0069] In a specific example, the step of analyzing and obtaining the comprehensive score of soil physicochemical properties based on the normalized physicochemical property vector includes: obtaining the preset weight vector corresponding to each physicochemical property index in the normalized physicochemical property vector based on the preset weight allocation rules for each physicochemical property; and obtaining the comprehensive score of soil physicochemical properties by weighted summation based on the normalized physicochemical property vector and the weight vector.

[0070] It should be noted that obtaining the weight vector involves assigning a preset weight value to each physicochemical property indicator (such as pH value and organic matter content) in the normalized physicochemical property vector based on a preset weighting rule. This weight value is a real number between 0 and 1, representing the relative importance of that physicochemical property indicator in the comprehensive evaluation of basic soil environmental conditions, and its sum is 1. The weighting rule is based on prior knowledge and experience in soil science, used to ensure that the contribution of different properties is reasonably quantified. Furthermore, the weight vector is a one-dimensional array with the same length as the normalized physicochemical property vector, and each element in the array corresponds sequentially to the weight coefficient of a physicochemical property indicator.

[0071] It should be noted that the weighted summation calculation is based on the normalized physicochemical property vector and the weight vector, and performs a linear weighted combination calculation process. This process multiplies each normalized physicochemical property index value by its corresponding weight coefficient to obtain a weighted value. Then, all weighted values ​​are summed to generate a scalar value, i.e., the comprehensive physicochemical property score. This operation integrates multi-dimensional soil physicochemical property information into a single, comparable comprehensive score, used to comprehensively characterize the soil's physical and chemical environmental foundation (integrating information from all preset key physicochemical property indicators to obtain a quantitative score for overall assessment and comparison of the basic environmental conditions of different soil samples); a higher comprehensive physicochemical property score generally indicates better basic physicochemical environmental conditions of the soil.

[0072] S5. Based on the comprehensive scores of network resilience, functional redundancy index and physicochemical properties, analyze the deviations; then generate the soil microbial health index, and conduct soil health status assessment and source tracing.

[0073] In a specific example, the analysis of deviations based on the network resilience, functional redundancy index, and comprehensive physicochemical property score includes: performing deviation judgment based on the values ​​of each of the network resilience, functional redundancy index, and comprehensive physicochemical property score, as well as preset optimal value ranges for each of the network resilience, functional redundancy index, and comprehensive physicochemical property score, to obtain a deviation judgment result; performing a branch decision operation based on the deviation judgment result; when the deviation judgment result indicates that the network resilience, functional redundancy index, or comprehensive physicochemical property score deviates from its corresponding optimal value range, performing in-depth source tracing analysis on the deviation index; generating and outputting a source tracing analysis report based on the result of the index source tracing analysis operation; and triggering the generation of the soil microbial health index when none of them deviate from their respective optimal value ranges.

[0074] It should be noted that the deviation judgment operation is a process of comparing the values ​​of the network resilience, functional redundancy index, and physicochemical property comprehensive scores with their respective preset optimal value ranges. Specifically, this operation checks whether the value of each indicator falls within its preset optimal value range (for example, the optimal value range for network resilience may be [0.7, 1.0]). If the value of an indicator is lower or higher than its corresponding optimal value range, it is determined that the indicator has deviated; if the values ​​of all indicators are within their corresponding optimal value ranges, it is determined that none have deviated. The deviation judgment results are used to identify the possible main causes of soil health problems. If there is an indicator deviation, it indicates that the soil health problem may be directly caused by the instability of the microbial network structure, insufficient functional backup, or unsuitable physicochemical environment represented by the abnormal indicator; if none deviate, it indicates that a more comprehensive health status assessment is needed. Furthermore, the branch decision operation is based on the deviation judgment result and controls the logical selection process of the subsequent processing flow (when any core indicator is detected to be abnormal, the process will switch to the in-depth source analysis branch for the abnormal indicator to quickly locate the root cause of the problem); only when all core indicators are within the normal range will the subsequent comprehensive health index calculation and level determination branch be triggered, thereby avoiding invalid or misleading comprehensive assessments in the case of abnormal indicators and improving the efficiency and pertinence of the assessment process.

[0075] It should be noted that this source tracing analysis aims to quickly locate the underlying causes of the abnormal indicators, specifically through the following steps: Based on the deviation judgment results, identify the types of abnormal indicators that have deviated; the types of abnormal indicators include: network resilience deviation, functional redundancy index deviation, physicochemical property comprehensive score deviation, or a combination thereof; according to the type of abnormal indicator, perform the corresponding source tracing analysis operation: A1. If the type of abnormal indicator is network resilience deviation, then perform network resilience source tracing analysis: retrieve the original data used to calculate network resilience, including the microbial co-occurrence network, and the damaged network connectivity values ​​recorded after each simulation in the random node removal simulation operation. Analyze the degree distribution changes of nodes (microbial species) in the microbial co-occurrence network, the absence of key microbial species (such as high-connectivity nodes), and, combined with the downward trend of the damaged network connectivity values, locate the main microbial species or microbial species groups that cause the increase in the overall network connectivity vulnerability. A2. If the type of abnormal indicator is functional redundancy index deviation, then perform functional redundancy index source tracing analysis: retrieve the original data used to calculate the functional redundancy index, including the module partitioning results and modularity measurement values. Analyze the sparsity of internal node connections within each functional module and the abnormal enhancement of connections between modules to identify specific microbial functional modules with significantly reduced functional backup capabilities. A3. If the abnormal indicator type is a deviation in the comprehensive score of physicochemical properties, then perform a source tracing analysis of the comprehensive score of physicochemical properties: retrieve the original physicochemical property data and the weight vector. Compare the original measured values ​​of each physicochemical property indicator (such as pH value, organic matter content) with their respective optimal ranges, and combine them with their preset weights to quantify the contribution of each indicator to the deviation of the comprehensive score, thereby locating one or more abnormal physicochemical property parameters with the greatest impact.

[0076] It should be noted that the process of generating the source tracing analysis report is a summary and formatted output of the aforementioned specific analysis results. This report not only lists abnormal indicators, but more importantly, clarifies the most likely direct causes of the abnormalities (such as "the absence of key nitrifying bacteria genus X leads to decreased network resilience", "the internal connection of denitrification functional module Y is broken, resulting in insufficient functional redundancy", and "soil pH value below the suitable range is the main limiting factor"). Its significance lies in transforming complex health assessment results into specific and actionable diagnostic conclusions that can be understood and implemented by agricultural or environmental managers, completing the entire analytical chain from "identifying the problem" to "locating the root cause".

[0077] In a specific example, the step of generating a soil microbial health index and assessing and tracing the soil health status includes: weighting and fusing the network resilience, the functional redundancy index, and the comprehensive score of the physicochemical properties through a preset health index fusion model and introducing interaction terms to obtain the soil microbial health index; and comparing and judging the soil microbial health index with a preset health level threshold range to obtain the soil health level judgment result.

[0078] It should be noted that the health index fusion model is a pre-defined mathematical model used to integrate the network resilience, functional redundancy index, and physicochemical property comprehensive score, which represent different dimensions, into a single comprehensive health index. This model comprehensively considers the independent contribution of each indicator to overall health, as well as the synergistic or antagonistic effects between indicators, through weighted fusion and the introduction of interaction terms. Further, the process of calculating the soil microbial health index through weighted fusion and the introduction of interaction terms is as follows: First, calculate the network resilience multiplied by the first weighting coefficient, the functional redundancy index multiplied by the second weighting coefficient, and the physicochemical property comprehensive score multiplied by the third weighting coefficient; then calculate the product of network resilience and functional redundancy index, and multiply it by the interaction term coefficient to obtain the interaction contribution value; finally, add the aforementioned three weighted results to the interaction contribution value to obtain the final soil microbial health index.

[0079] It should be noted that the acquisition of the first, second, and third weighting coefficients is as follows: B1, collect a historical soil sample dataset, which includes network resilience values, functional redundancy index values, and comprehensive physicochemical property scores of multiple historical samples. B2, for each indicator in the historical soil sample dataset, plot the change curve of each indicator with the number of samplings as the horizontal axis and the indicator value as the vertical axis. B3, calculate the integral of the change curve of each indicator within a preset sampling number interval to obtain the network resilience integral, functional redundancy index integral, and comprehensive physicochemical property score integral. B4, normalize the three integral values ​​so that their sum is 1. The normalized proportions are set as the first, second, and third weighting coefficients, respectively.

[0080] It should be noted that the aforementioned proportional allocation method based on historical data is an operation to quantify the long-term contribution of various indicators to the health of the soil microbial community (by analyzing a large amount of historical health data, objectively measuring the relative importance of indicators in different dimensions (structural stability, functional backup capacity, environmental basis) in maintaining the overall health status). The more stable an indicator is in historical data, and the larger the area (integral) under its change curve, the more significant the contribution of the stability of that dimension to maintaining the health status, and therefore it should be given a higher weight.

[0081] It should be noted that the comparison and determination operation is a logical process that compares the calculated value of the soil microbial health index with multiple pre-set health level threshold intervals. This operation sequentially matches the soil microbial health index value with threshold intervals representing different health levels (such as "healthy," "sub-healthy," or "unhealthy"). Furthermore, the health level threshold intervals are numerical ranges pre-defined based on the statistical distribution of soil microbial health indices from a large number of historical healthy soil samples or expert experience. For example, a soil microbial health index greater than or equal to 0.8 is considered "healthy," 0.5 less than or equal to a soil microbial health index less than 0.8 is considered "sub-healthy," and a soil microbial health index less than 0.5 is considered "unhealthy." The soil health level determination result is a classification label output by this matching operation. Its physical meaning lies in providing an intuitive and graded qualitative evaluation of the overall health status of the soil's microbial community and environment, providing a direct decision-making basis for soil management and remediation.

[0082] This application enables precise source tracing of problems. By analyzing deviations in indicators, it can identify the main causes of health issues (such as structural fragility, insufficient redundancy, or environmental degradation), guiding targeted interventions. It supports dynamic monitoring and prediction; long-term data accumulation can reveal the trajectory of health evolution and changes in dominant factors, providing trend warnings and a basis for strategy adjustment for soil restoration and protection. It achieves multi-dimensional information fusion, overcoming the limitations of single-dimensional evaluation and comprehensively characterizing soil health. It enhances the ecological relevance of the index; interaction terms reflect the coupling relationship between microorganisms and the environment, making the index closer to real ecological processes. It strengthens the sensitivity and discriminative power of the assessment; weighted fusion highlights the contribution of key indicators, making the health index more responsive to changes in different soil conditions, suitable for gradient comparison and long-term monitoring.

[0083] Please see Figure 2 As shown, in a second aspect, this application provides a system for a soil health assessment method based on microbial communities.

[0084] The system 100 of the soil health assessment method based on microbial communities described in this invention can be installed in an electronic device. Depending on the functions implemented, the system 100 may include a data generation module 101, a microbial species co-occurrence network construction module 102, a microbial species co-occurrence network analysis module 103, a physicochemical property comprehensive score calculation module 104, and a soil health status assessment and tracing module 105.

[0085] The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and are stored in the memory of the electronic device.

[0086] In this embodiment, the functions of each module / unit are as follows: The data generation module is used to collect target soil samples, obtain microbial species composition and abundance data, and generate a standardized species abundance matrix; at the same time, it obtains preset key physicochemical property data of the soil and generates a normalized physicochemical property vector.

[0087] The microbial species co-occurrence network construction module analyzes the correlation between various microbial species based on the standardized species abundance matrix and constructs a microbial species co-occurrence network.

[0088] The microbial species co-occurrence network analysis module performs topological structure analysis based on the microbial co-occurrence network and calculates the resilience and functional redundancy index of the microbial co-occurrence network.

[0089] The physical and chemical properties comprehensive score calculation module analyzes and obtains the comprehensive score of the soil's physical and chemical properties based on the normalized physical and chemical property vector.

[0090] The soil health status assessment and tracing module analyzes deviations based on the comprehensive scores of network resilience, functional redundancy index, and physicochemical properties; then it generates a soil microbial health index and conducts soil health status assessment and tracing.

[0091] The soil health assessment method and system based on microbial communities provided in this application integrate the network structure characteristics of microbial communities with basic soil physicochemical environmental information. By calculating network resilience and functional redundancy indices, it can deeply quantify the intrinsic stability and potential for resistance to disturbance of microbial communities from two key dimensions: species interaction stability and functional backup capacity. This enables precise diagnosis and effective tracing of soil health status. The resulting health index provides an intuitive and comprehensive health scale, facilitating the classification and comparison of different soil types and allowing for the tracing of dominant factors leading to a decline in health levels. This provides a clear scientific basis for formulating precise and targeted soil restoration and management measures, and has significant application value for promoting sustainable agricultural development, ecological restoration projects, and land resource management.

[0092] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0093] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0094] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0095] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0096] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0097] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A soil health assessment method based on microbial communities, characterized in that, include: S1. Collect target soil samples, obtain microbial species composition and abundance data, and generate a standardized species abundance matrix; Simultaneously, the preset key physicochemical property data of the soil are acquired, and normalized physicochemical property vectors are generated; S2. Based on the standardized species abundance matrix, the correlation between various microbial species is analyzed, and a microbial species co-occurrence network is constructed. S3. Based on the microbial co-occurrence network, topological structure analysis was performed to calculate the resilience and functional redundancy index of the microbial co-occurrence network. S4. Based on the normalized physicochemical property vector, the comprehensive score of the soil's physicochemical properties is obtained through analysis; S5. Based on the comprehensive scores of network resilience, functional redundancy index and physicochemical properties, analyze the deviations; then generate the soil microbial health index, and conduct soil health status assessment and source tracing.

2. The soil health assessment method based on microbial community according to claim 1, characterized in that, The process of acquiring microbial species composition and abundance data and generating a standardized species abundance matrix includes: Target soil samples were collected and sequenced using high-throughput sequencing technology to obtain raw species abundance data. The raw species abundance data were then standardized using a pre-defined standardization method to obtain a standardized species abundance matrix.

3. The soil health assessment method based on microbial community according to claim 1, characterized in that, The simultaneous acquisition of preset key physicochemical property data of the soil and the generation of normalized physicochemical property vectors include: Obtain preset key physicochemical property data of the target soil to obtain raw physicochemical property data; based on a preset normalization algorithm, normalize the raw physicochemical property data to obtain a normalized physicochemical property vector.

4. The soil health assessment method based on microbial community according to claim 1, characterized in that, Based on the standardized species abundance matrix, the correlations among various microbial species are analyzed, and a microbial species co-occurrence network is constructed, including: Each microbial species is paired up and recorded as a microbial species pair; then, based on the standardized species abundance matrix, Spearman correlation analysis is used to calculate the correlation between each microbial species pair to obtain the correlation coefficient and corresponding significance value of each microbial species pair. Based on the correlation coefficients and corresponding significance values ​​of each microbial species pair, the microbial species pairs are screened for significance to obtain a set of significantly related microbial species pairs; then, a microbial co-occurrence network is constructed to obtain a microbial co-occurrence network with microbial species as nodes and significant correlation relationships as edges.

5. The soil health assessment method based on microbial community according to claim 4, characterized in that, The calculation of the correlation between each microbial species pair using Spearman correlation analysis to obtain the correlation coefficient and corresponding significance value of each microbial species pair includes: Through calculation formula The correlation coefficients of microbial species pairs were obtained. , Represented as the first The difference in abundance rank between two microbial species in a soil sample. This is represented by the number corresponding to the soil sample. , This is expressed as the total number of soil samples; Through calculation formula Derive the test statistic Furthermore, in a degree of freedom of of Extract the significance value corresponding to the test statistic from the distribution.

6. The soil health assessment method based on microbial community according to claim 1, characterized in that, The topology analysis based on the microbial co-occurrence network calculates the resilience and functional redundancy index of the microbial co-occurrence network, including: Based on the microbial co-occurrence network and a preset random removal ratio, a random node removal simulation operation is performed to obtain several simulated network connectivity values. Then, a resilience quantification calculation operation is performed to obtain the resilience of the microbial co-occurrence network. Based on the microbial co-occurrence network, a modular structure analysis operation is performed to obtain the module partitioning results and modularity metric; based on the module partitioning results and the modularity metric, a functional redundancy index calculation operation is performed to obtain the functional redundancy index.

7. The soil health assessment method based on microbial community according to claim 1, characterized in that, The comprehensive score of soil physicochemical properties, derived from the normalized physicochemical property vector, includes: Based on the preset weight allocation rules for various physicochemical properties, the preset weight vectors corresponding to each physicochemical property index in the normalized physicochemical property vector are obtained; based on the normalized physicochemical property vector and the weight vector, the comprehensive score of the soil's physicochemical properties is obtained by weighted summation.

8. The soil health assessment method based on microbial community according to claim 1, characterized in that, The analysis of deviations based on the comprehensive score of network resilience, functional redundancy index, and physicochemical properties includes: Based on the values ​​of the network resilience, the functional redundancy index, and the comprehensive score of the physicochemical properties, and the preset optimal value ranges for each of the network resilience, functional redundancy index, and comprehensive score of the physicochemical properties, deviation judgment is performed to obtain deviation judgment results. Based on the deviation judgment results, a branch decision operation is executed. When the deviation judgment results indicate that the network resilience, functional redundancy index, or comprehensive score of the physicochemical properties deviates from its corresponding optimal value range, a deep source tracing analysis is performed on the deviation index. Based on the results of the source tracing analysis operation, a source tracing analysis report is generated and output. When none of them deviate from their respective optimal value ranges, the generation of the soil microbial health index is triggered.

9. The soil health assessment method based on microbial community according to claim 1, characterized in that, The process then generates a soil microbial health index and assesses and traces the source of soil health status, including: Based on the network resilience, the functional redundancy index, and the comprehensive score of the physicochemical properties, a weighted fusion is performed through a preset health index fusion model, and an interaction term is introduced to obtain the soil microbial health index. Based on the soil microbial health index and the preset health level threshold range, a comparison and judgment are made to obtain the soil health level judgment result.

10. A system for implementing the soil health assessment method based on microbial communities according to any one of claims 1-9, characterized in that, include: The data generation module is used to collect target soil samples, obtain microbial species composition and abundance data, and generate a standardized species abundance matrix; at the same time, it obtains preset key physicochemical property data of the soil and generates normalized physicochemical property vectors. The microbial species co-occurrence network construction module analyzes the correlation between various microbial species based on the standardized species abundance matrix and constructs a microbial species co-occurrence network. The microbial species co-occurrence network analysis module performs topological structure analysis based on the microbial co-occurrence network and calculates the resilience and functional redundancy index of the microbial co-occurrence network. The physical and chemical properties comprehensive score calculation module analyzes and obtains the comprehensive score of the soil's physical and chemical properties based on the normalized physical and chemical property vector; The soil health status assessment and tracing module analyzes deviations based on the comprehensive scores of network resilience, functional redundancy index, and physicochemical properties; then it generates a soil microbial health index and conducts soil health status assessment and tracing.

Images