A rumen microecological balance index-based method for evaluating health status of ruminants
By constructing the rumen microecological balance index RMNBI and combining metagenomics and low-cost 16S sequencing data, the cross-scenario applicability and quantitative assessment of dairy cow rumen microbial community evaluation were solved, enabling accurate assessment and dynamic monitoring of dairy cow health status, applicable to different scale farming scenarios.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- 河南省种业发展中心
- Filing Date
- 2026-05-07
- Publication Date
- 2026-07-31
AI Technical Summary
Existing technologies for evaluating the rumen microbiome in dairy cows suffer from poor cross-scenario applicability, high testing costs, and a lack of quantitative assessment. They cannot consistently reflect the state of dysbiosis, distinguish between healthy, subclinical, and clinical disease states, and cannot achieve dynamic monitoring of disease severity and treatment effectiveness.
A method for evaluating the health status of ruminants based on the rumen microecological balance index was adopted. The rumen functional groups, interaction matrix, redundancy matrix, species-function mapping spectrum and environmental correction function were constructed by metagenomic sequencing data. The abundance of functional groups was predicted by combining low-cost 16S sequencing data, and the rumen microecological balance index RMNBI of dairy cows was calculated to achieve quantitative determination.
It consistently reflects the state of dysbiosis in different disease scenarios, reduces detection costs and cycles, can accurately distinguish the gradient states of healthy, subclinical, and clinical diseases, dynamically monitors the treatment effect of diseases, improves the accuracy and reliability of steady-state judgment results, and is adaptable to different scale farming scenarios.
Smart Images
Figure CN122493980A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the technical field of animal health testing in animal husbandry, and in particular to a method for evaluating the health status of ruminants based on the assessment of rumen microbial community homeostasis. Background Technology
[0002] The rumen is a digestive organ unique to ruminants, and it is home to a complex microbial community composed of bacteria, archaea, protozoa, and fungi, with a population as high as 1,000 per milliliter. These microbial cells work synergistically to convert plant fiber, which the host cannot directly utilize, into volatile fatty acids (VFAs), providing dairy cows with approximately 70%–80% of their energy. Numerous studies have shown that the composition and functional state of the rumen microbiome are closely related to various productive traits and health conditions in dairy cows; for example, microbial fermentation efficiency directly affects milk yield, milk fat percentage, and milk protein percentage. High-producing dairy cows typically have higher fiber degradation capabilities and more efficient VFA production pathways. Rumen microbiota dysbiosis is a core contributing factor to metabolic diseases such as subacute rumen acidosis (SARA) and ketosis. When easily fermentable carbohydrates are excessive, lactic acid-producing bacteria proliferate excessively, causing a sharp drop in pH and disrupting microbial homeostasis. Rumen microbes indirectly influence the risk of inflammatory diseases such as subclinical mastitis and endometritis by regulating intestinal barrier function and immune responses. Recent studies have found a correlation between rumen microbiota composition and the lifespan, conception rate, and culling risk of dairy cows. Furthermore, methanogenic bacteria in the rumen produce a significant amount of methane annually (accounting for 30%–40% of greenhouse gas emissions from livestock farming), while nitrogen emissions are also closely related to the activity of rumen protein-degrading bacteria. Therefore, accurate and quantitative assessment of the rumen microbial community homeostasis in dairy cows is of great significance for dairy cow health management, performance improvement, and the green development of livestock farming.
[0003] Currently, the evaluation and detection of the rumen microbiome in dairy cows mainly rely on the following three traditional technical methods: First, the microbial diversity evaluation method, represented by the Shannon index, judges the community status by measuring the species richness and evenness of the microbiome, and the industry generally regards the level of diversity as the basis for judging the stability of the microbiome; Second, the abundance detection method of specific functional microorganisms, which inversely infers rumen metabolic abnormalities and disease risks by monitoring the abundance changes of key species such as lactic acid-producing bacteria, fiber-degrading bacteria, and methanogens; Third, the rumen physicochemical index detection method, which directly measures parameters such as rumen fluid pH, ammonia nitrogen, and volatile fatty acids to indirectly reflect the fermentation status of microorganisms. With the development of sequencing technology, metagenomic sequencing can comprehensively analyze the functional composition of the microbiome, but it has the disadvantages of high detection cost and long cycle, making it difficult to adapt to large-scale routine screening in farms; 16S rRNA sequencing is inexpensive and fast, but it can only obtain species abundance information and cannot be directly converted into data for evaluating microbiome function and stability, thus its practical value has not been fully realized.
[0004] Although the importance of rumen microbiota is widely recognized in the industry, existing technologies have several insurmountable shortcomings in clinical applications: First, traditional diversity indices only focus on the species composition of the microbiota, ignoring core interactions such as cooperation and competition among microorganisms. In different disease scenarios such as ketosis and mastitis, diversity changes do not follow a uniform pattern, failing to consistently reflect the state of microbiota dysbiosis and exhibiting extremely poor cross-scenario applicability. Second, diagnostic methods based on the abundance of specific species lack generality and universality. The normal abundance of key microbial species varies significantly across different diets, lactation stages, and dairy cow breeds. A single judgment standard cannot be adapted to multiple scenarios and can only provide a rough assessment of diseases caused by a single factor. Third, existing technologies are all qualitative evaluations, lacking quantifiable indicators for assessing microbiota homeostasis. They cannot distinguish between healthy, subclinical, and clinical disease states, nor can they achieve assessment of disease severity and dynamic monitoring of treatment effects. Summary of the Invention
[0005] To address the technical problems of poor applicability to multiple scenarios, high detection costs, and lack of quantitative assessment in existing technologies, this invention proposes a method for evaluating the health status of ruminants based on the rumen microecological balance index. The method constructs rumen functional groups, interaction matrices, redundancy matrices, species-function mapping spectra, and environmental correction functions using metagenomic sequencing data. Then, it predicts the abundance of functional groups using low-cost 16S sequencing data, calculates the rumen microecological balance index (RMNBI) of dairy cows, and finally combines health thresholds to quantitatively determine the rumen microecological homeostasis.
[0006] To achieve the above objectives, the technical solution of the present invention is implemented as follows:
[0007] A method for evaluating the health status of ruminants based on the rumen microecological balance index, comprising the following steps:
[0008] S1: Obtain multiple sets of rumen fluid samples from dairy cows and perform metagenomic sequencing and rumen physicochemical index determination respectively. Obtain the first species abundance matrix from metagenomic sequencing; define multiple functional groups based on rumen metabolic pathways, and calculate the first functional group abundance matrix based on the abundance of metabolic pathways in each functional group; construct a species-functional group mapping spectrum based on non-negative least squares method with the mapping relationship between the first species abundance matrix and the first functional group abundance matrix as the objective.
[0009] S2: Calculate the correlation coefficients between functional groups based on the abundance matrix of the first functional group, and construct positive and negative interaction matrices based on the correlation coefficients between functional groups;
[0010] S3: Construct a functional redundancy matrix by calculating the similarity between functional groups in rumen metabolic pathways;
[0011] S4: Constructing environmental correction coefficients based on rumen physicochemical index data;
[0012] S5: 16S sequencing was performed on the rumen fluid samples from dairy cows to obtain the second species abundance matrix;
[0013] S6: Map the second species abundance matrix using the species-functional group mapping spectrum and normalize it to obtain the second functional group abundance matrix.
[0014] S7: Calculate the positive interaction strength of all functional group pairs based on the second functional group abundance matrix and the positive interaction matrix. Use the functional redundancy matrix as a correction factor and combine the second functional group abundance matrix and the negative interaction matrix to calculate the negative interaction strength of all functional group pairs.
[0015] S8: Calculate the original ecological balance index based on the positive and negative interaction strengths, and correct the original ecological balance index using the environmental correction coefficient to obtain the final RMNBI ecological balance index.
[0016] S9: Assess the health status of ruminants based on the health threshold of the RMNBI ecological balance index for healthy ruminants.
[0017] Furthermore, multiple functional groups are defined based on rumen metabolic pathways, including:
[0018] Based on the key role of metabolic pathways in rumen energy metabolism, the detectability of metabolic pathways in metagenomic data, and the known associations between metabolic pathways and dairy cow production performance, functional groups are defined using the MetaCyc standardized metabolic pathway database as a reference. At least the following functional groups are defined: fiber degradation, starch degradation, methanogenesis, propionic acid production, lactic acid metabolism, protein degradation, and sulfate reduction.
[0019] Furthermore, the first functional group abundance matrix is calculated based on the abundance of metabolic pathways in each functional group, including: for each sample, aggregating the abundance of all metabolic pathways in each functional group to obtain an initial functional group abundance matrix; and normalizing the abundance vector of each functional group in the initial functional group abundance matrix to obtain the first functional group abundance matrix. N is the number of samples, and K is the number of functional groups.
[0020] Furthermore, taking the mapping relationship between the first species abundance matrix and the first functional group abundance matrix as the objective, a species-functional group mapping spectrum is constructed based on the non-negative least squares method, including:
[0021] For each functional group, establish a mapping relationship between species abundance and functional group abundance: , This represents the abundance vector of the N samples corresponding to the k-th functional group. Let be the contribution weight vector of species s to functional group k. This is random error;
[0022] For each functional group, predict the abundance of the functional group. With actual functional group abundance The optimization objective is to minimize the sum of squared errors, while applying non-negative weights. The constraints are solved using the non-negative least squares algorithm to obtain the contribution weight vector of each species to each functional group;
[0023] The contribution weight vector of each species to each functional group is normalized so that the sum of the contributions of each species to all functional groups is 1, thus obtaining the species-functional group mapping spectrum. S represents the number of microbial species.
[0024] Furthermore, when calculating the correlation coefficient between functional groups based on the abundance matrix of the first functional group, the Spearman rank correlation coefficient is used to characterize the correlation coefficient between functional groups: ,in, , Let a and b represent the abundance vectors of the N samples corresponding to the a-th and b-th functional groups, respectively. This represents the function for calculating the Spearman rank correlation coefficient.
[0025] Construct positive and negative interaction matrices based on the correlation coefficients between functional groups, including:
[0026] Construct a positive interaction matrix: ;
[0027] Construct a negative interaction matrix: .
[0028] Furthermore, a functional redundancy matrix is constructed by calculating the similarity between functional groups in rumen metabolic pathways, including:
[0029] The similarity between functional groups a and b in rumen metabolic pathways was calculated using Jaccard similarity. The redundancy matrix is composed of the similarity between multiple functional group pairs. The formula is:
[0030] ;
[0031] in, The similarity between functional groups a and b in rumen metabolic pathways.
[0032] Furthermore, the method for constructing environmental correction coefficients based on rumen physicochemical index data is as follows:
[0033] ;
[0034] in, Let m be the measured value of the m-th environmental factor. For the m-th environmental factor, This is a sensitivity coefficient used to control the rate of decrease of the correction coefficient when deviating from the optimal value. For the weighting coefficients, satisfying , This is the product symbol.
[0035] Furthermore, the formula for mapping the second species abundance matrix using the species-functional group mapping spectrum is as follows:
[0036] ;
[0037] in, The second species abundance matrix is the result of prediction. This is the second species abundance matrix obtained from the samples to be tested.
[0038] Furthermore, the method for calculating the positive interaction strength of all functional group pairs based on the second functional group abundance matrix and the positive interaction matrix is as follows:
[0039] ;
[0040] in, Let be the strength of the positive interaction between functional group a and functional group b. , These represent the functional group abundance of the k-th functional group of the i-th test sample in the second species abundance matrix;
[0041] The method for calculating the negative interaction strength of all functional group pairs, using the functional redundancy matrix as a correction factor and combining the second functional group abundance matrix and the negative interaction matrix, is as follows:
[0042] ;
[0043] in, The negative interaction strength between functional group a and functional group b.
[0044] Furthermore, the method for calculating the original ecological balance index based on positive and negative interaction strengths is as follows:
[0045] ;
[0046] in, Let be the original ecological balance index of the i-th sample to be tested. It is a very small positive number;
[0047] The method for correcting the original ecological balance index using environmental correction coefficients is as follows:
[0048] ;
[0049] in, Let i be the ecological balance index of the i-th sample to be tested. is the environmental correction coefficient for the i-th sample to be tested.
[0050] The beneficial effects of this invention are as follows:
[0051] This invention abandons the traditional approach of judging the homeostasis of the microbial community solely based on species richness and evenness. Instead, it quantifies the cooperative and competitive interactions of microbial functional groups at the metabolic function level. It can consistently reflect the state of microbial dysbiosis in different disease scenarios such as ketosis, mastitis, and subacute rumen acidosis, thus solving the core problems of traditional indices failing across scenarios and lacking a unified judgment standard.
[0052] A functional redundancy matrix is constructed based on the overlap of metabolic pathways to biologically correct the competitive relationships of functional groups, effectively avoiding misjudgments of interaction relationships caused by functional overlap, and making the steady-state assessment results more consistent with the actual metabolic state of the dairy cow's rumen.
[0053] By adopting a two-stage strategy of one-time metagenomic modeling and routine application of 16S sequencing, only the initial investment in metagenomic sequencing to build the core model is required. Routine testing can be completed by using low-cost 16S sequencing to calculate the index, which greatly reduces the testing cost and cycle. This solves the industry pain points of metagenomic sequencing being difficult to apply on a large scale and 16S sequencing being unable to be used for functional homeostasis evaluation.
[0054] The constructed RMNBI index can accurately distinguish between healthy, subclinical, and clinical disease states. The index value is directly related to the severity of the disease. It can quantitatively assess the degree of rumen microecological imbalance and dynamically monitor the recovery effect after disease treatment, filling the technical gap in the field where the microbiota homeostasis cannot be quantitatively assessed.
[0055] By incorporating core physicochemical indicators such as rumen pH, ammonia nitrogen, and volatile fatty acids into the environmental correction function, the coupling calibration of microbial interaction data and the rumen internal environment is achieved, avoiding the one-sidedness of pure microbiome data evaluation and significantly improving the accuracy and reliability of steady-state determination results.
[0056] With rumen core metabolic function as the evaluation core, it is not limited by factors such as dairy cow breed, diet type, lactation stage, and pasture environment. One standard can be adapted to different scale farming scenarios, solving the problems of poor universality and limited scenarios of traditional species-specific diagnostic methods. Attached Figure Description
[0057] 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 some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0058] Figure 1 This is a flowchart of the rumen microecological balance index-based method for evaluating the health status of ruminants according to the present invention.
[0059] Figure 2 This is an RMNBI trend chart of ketosis-stricken dairy cows in an embodiment of the present invention.
[0060] Figure 3 This is an RMNBI trend chart for the robustness test of mastitis in this embodiment of the invention. Detailed Implementation
[0061] 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.
[0062] Methods for evaluating the health status of ruminants based on rumen microecological balance index, such as Figure 1 As shown, this includes the model building phase and the application phase.
[0063] During the model building phase, the following are included:
[0064] S1. Obtain multiple groups of rumen fluid samples from dairy cows and perform metagenomic sequencing and rumen physicochemical index determination respectively. Obtain the species abundance matrix of microorganisms in each sample group (first species abundance matrix) from metagenomic sequencing. Define multiple functional groups based on rumen metabolic pathways, and calculate the first functional group abundance matrix based on the abundance of metabolic pathways in each functional group. Using the mapping relationship between the first species abundance matrix and the first functional group abundance matrix as the target, construct a species-functional group mapping spectrum based on non-negative least squares method. This step is to obtain core data for modeling, define functional groups and build species-functional group mapping relationships. By collecting sample data from multiple scenarios, dividing functional groups, calculating standardized functional group abundances and constructing mapping spectra, a rumen microbial functional evaluation system is established, laying a solid foundation for low-cost 16S functional prediction.
[0065] In this embodiment of the application, metagenomic sequencing data and rumen physicochemical index data of multiple groups of bovine rumen fluid samples are obtained. The species abundance matrix of microorganisms in each group of samples is obtained from the metagenomic sequencing data, including:
[0066] Three hundred and forty representative rumen samples (covering different health states, dietary types, and lactation stages) were collected. Metagenomic sequencing (Illumina NovaSeq 6000, sequencing depth ≥10 Gb / sample) was performed on each sample using Kraken2 and Bracken bioinformatics software to obtain the species abundance matrix of microorganisms in each sample group. N represents the number of samples, S represents the number of microorganisms, and the sum of each row is 1. At the same time, rumen physicochemical indicators (pH, ammonia nitrogen, VFA concentration, etc.) are measured.
[0067] In this embodiment of the application, multiple functional groups are defined based on the rumen metabolic pathway, including:
[0068] Based on the characteristics of rumen metabolic function and using the MetaCyc standardized metabolic pathway database as a reference, seven functional groups were defined. The criteria for functional group classification included: the key role of the metabolic pathway in rumen energy metabolism, the detectability of the metabolic pathway in metagenomic data, and the known associations between the metabolic pathway and dairy cow production performance. These included:
[0069]
[0070] In this embodiment of the application, the calculation of the first functional group abundance matrix based on the abundance of metabolic pathways in each functional group includes:
[0071] For each sample, the abundance of all metabolic pathways in each functional group is aggregated to obtain an initial functional group abundance matrix. The abundance vector of each functional group in the initial functional group abundance matrix is normalized to obtain the first functional group abundance matrix.
[0072] The formula is as follows:
[0073] Functional group abundance aggregation:
[0074] ;
[0075] in, Let be the abundance of the functional group of the k-th functional group for the i-th sample. For the kth functional group, For the i-th sample, the th The abundance of each metabolic pathway;
[0076] Functional group abundance normalization:
[0077] ;
[0078] Here, K represents the number of functional groups. The functional group abundance vector of each sample is normalized so that the sum of its components is 1, facilitating subsequent cross-sample comparisons. The final output is the functional group abundance matrix. Where N is the number of samples.
[0079] In this embodiment of the application, the mapping relationship between the first species abundance matrix and the first functional group abundance matrix is taken as the objective, and a species-functional group mapping spectrum is constructed based on the non-negative least squares method, including:
[0080] First, for each functional group, establish a mapping relationship between species abundance and functional group abundance: ,in This represents the abundance vector (N) of the N samples corresponding to the k-th functional group. 1), The weight vector (S) corresponding to the kth functional group 1) represents the proportion of species s' contribution to functional group k. This is random error.
[0081] Furthermore, for each functional group, the abundance of the functional group is predicted. With actual functional group abundance The optimization objective is to minimize the sum of squared errors, while applying non-negative weights. The constraints are solved using the non-negative least squares algorithm to obtain the contribution weight of each species to the functional group; the formula is as follows:
[0082] ;
[0083] Among them, the non-negativity constraint ensures that the contribution of a species to a functional group is non-negative, which is consistent with biological intuition.
[0084] Furthermore, the contribution weight vector for each species to each functional group is... Row normalization is performed so that the sum of the species' contributions to all functional groups is 1. In this way, for any species, its total abundance is weighted and allocated to each functional group.
[0085] Finally, the species-functional group mapping spectrum was obtained. ,satisfy For any species Established.
[0086] S2: Calculate the correlation coefficients between functional groups based on the abundance matrix of the first functional group, and construct positive and negative interaction matrices according to the correlation coefficients. This step quantifies the cooperative and competitive ecological interactions of functional groups. By calculating Spearman rank correlations and constructing positive and negative interaction matrices, standardized quantification of microbial community interactions is achieved, adapting to the steady-state evaluation of multiple disease scenarios.
[0087] In this embodiment of the application, the calculation of the correlation coefficient between functional groups based on the first functional group abundance matrix includes:
[0088] First, the Spearman rank correlation coefficient between functional groups is calculated to reflect the covariation trend of their abundance across samples. The Spearman correlation coefficient is insensitive to outliers and is suitable for microbiome data. The formula is:
[0089] ;
[0090] in, The Spearman rank correlation coefficient represents the relationship between functional groups a and b. This represents the function for calculating the correlation coefficient. This indicates that the Spearman rank correlation method is used. , Let a and b represent the abundance vectors of the a-th and b-th functional groups, respectively.
[0091] In this embodiment of the application, constructing a positive interaction matrix and a negative interaction matrix based on the correlation coefficient between functional groups includes:
[0092] First, construct the positive interaction matrix:
[0093] ;
[0094] in, This represents the positive interaction matrix between functional groups a and b. If the abundances of the two functional groups are positively correlated (… This indicates that they tend to coexist and may form a cooperative relationship through mechanisms such as cross-feeding and metabolic complementarity. This correlation coefficient is taken as the strength of positive interaction; if the correlation is negative or zero, the strength of positive interaction is set to 0.
[0095] Furthermore, construct a negative interaction matrix:
[0096] ;
[0097] in, This represents the negative interaction matrix between functional groups a and b. If the abundances of the two functional groups are negatively correlated ( This indicates that they tend to be mutually exclusive, possibly competing for the same substrate or having an inhibitory effect. The absolute value of the correlation coefficient is taken as the negative interaction strength; if it is positively correlated or zero, the negative interaction strength is set to 0.
[0098] Ultimately, the positive interaction matrix Negative interaction matrix All diagonal elements are 0.
[0099] S3: Construct a functional redundancy matrix by calculating the similarity between functional groups along rumen metabolic pathways. This step aims to align with the physiological characteristics of rumen functional backup and calibrate the competitive relationships among functional groups. A functional redundancy matrix is constructed by calculating pathway overlap using Jaccard similarity, avoiding misjudgments of competitive relationships and improving the biological rationality of interaction strength calculations.
[0100] In this embodiment of the application, a functional redundancy matrix is constructed by calculating the similarity between functional groups in the rumen metabolic pathway, including:
[0101] The similarity between functional groups a and b in rumen metabolic pathways was calculated using Jaccard similarity. The formula is:
[0102] ;
[0103] in, This represents the similarity between functional groups a and b in rumen metabolic pathways, specifically the proportion of pathways shared by the two functional groups relative to their total number of pathways. This value ranges from 0 to 1, reflecting the degree of overlap in metabolic function between the two functional groups. If the two functional groups share a large number of metabolic pathways (i.e.,...), then... The value close to 1 indicates that they are highly redundant in function, and even if their abundance is negatively correlated, their competitive relationship may be buffered by "functional backup". Therefore, in subsequent calculations, As a correction factor for negative interactions, the redundancy matrix is... The diagonal elements are 0.
[0104] S4: Construct environmental correction coefficients based on rumen physicochemical index data. This step couples rumen physicochemical internal environmental factors to calibrate the steady-state evaluation results. By converting physicochemical indices into correction coefficients and integrating the effects of multi-factor stress, environmental fluctuation interference is eliminated, improving the accuracy of the evaluation results.
[0105] In this embodiment of the application, the method for constructing the environmental correction coefficient based on rumen physicochemical index data is as follows:
[0106] ;
[0107] in, Let m be the measured value of the m-th environmental factor. For the m-th environmental factor, The sensitivity coefficient controls the rate of decrease of the correction coefficient when it deviates from the optimal value. For the weighting coefficients, satisfying , The product sign ensures that the overall correction coefficient decreases significantly when any factor deviates severely.
[0108] Taking pH as a single factor as an example:
[0109] ;
[0110] The formula maps pH values to a correction factor in the (0,1] interval. The optimal pH for rumen health (usually 6.5). The sensitivity coefficient controls the rate of decrease of the correction coefficient when the pH deviates from the optimum value. When the pH deviates from the optimum value, the correction coefficient decreases, indicating that environmental stress reduces the health of the ecological network.
[0111] During the application period, it includes:
[0112] S5: Obtain 16S sequencing data from the rumen fluid sample of the dairy cow to be tested. From the 16S sequencing data, obtain the species abundance matrix (second species abundance matrix) of the microorganisms in the sample. This step is a low-cost way to obtain species abundance data from the sample. By standardizing the 16S sequencing data and aligning the mapped species, the detection cost and cycle time are significantly reduced, making it suitable for large-scale, routine screening in dairy farms.
[0113] In this embodiment, the 16S sequencing data of the new sample is processed using the DADA2 tool or the QIIME2 platform to obtain the abundance table of the species to be tested. , The number of samples is denoted as , and the species names are aligned with the species names in the mapping spectrum M.
[0114] S6: The second species abundance matrix is mapped and normalized using a species-functional group mapping spectrum to obtain the predicted functional group abundance matrix (second functional group abundance matrix). To transform 16S species abundance data into standardized functional group abundance data, abundance is predicted and normalized based on a pre-built mapping spectrum, enabling rapid functional prediction and solving the problem that 16S sequencing cannot assess functional homeostasis.
[0115] In this embodiment of the application, the formula is:
[0116] Functional group abundance prediction:
[0117] ;
[0118] in, For the predicted second functional group abundance matrix, the predicted first... The first sample to be tested The abundance of each functional group is , Let be the measured abundance of the s-th species in the i-th new sample.
[0119] S7: Based on the second functional group abundance matrix and the positive interaction matrix, calculate the positive interaction strength of all functional group pairs. Using the functional redundancy matrix as a correction factor, combine the second functional group abundance matrix and the negative interaction matrix to calculate the negative interaction strength of all functional group pairs. This step obtains the overall cooperation and true competition strength of the sample after redundancy correction. It weights and calculates the total positive interaction strength and incorporates redundancy correction to calculate the total negative interaction strength, integrating complex interactions into core values to provide accurate input for the balance index calculation.
[0120] In this embodiment of the application, the method for calculating the positive interaction strength of all functional group pairs based on the second functional group abundance matrix and the positive interaction matrix is as follows:
[0121] ;
[0122] in, Let be the strength of the positive interaction between functional group a and functional group b. , Let be the function group abundances of the k-th function group for the i-th test sample in the second species abundance matrix. For the i-th test sample, calculate the abundance of all function group pairs. positive interaction strength Multiply by the product of the abundance of both sides Then sum them up. The product is... This reflects the ecological principle that only existing functional groups can interact—if the abundance of a certain functional group is zero, then all interactions it participates in contribute zero.
[0123] In this embodiment of the application, the method for calculating the negative interaction strength of all functional group pairs based on the negative interaction matrix of the second functional group abundance matrix and the functional redundancy matrix is as follows:
[0124]
[0125] in, Let be the negative interaction strength between function group a and function group b. Similarly, calculate the negative interaction strength for all function group pairs, but multiply by the redundancy matrix. If the functions of the two functional groups are highly redundant ( If the degree of functional overlap is close to 1, then its competitive weight is amplified; if the degree of functional overlap is low ( If the abundance is close to 0, then even if the abundance is negatively correlated, its competitive weight is weakened.
[0126] S8: Calculate the original ecological balance index based on the positive and negative interaction strengths, and then correct the original ecological balance index using an environmental correction coefficient to obtain the final RMNBI ecological balance index. This step is to construct a standardized quantitative rumen microecological balance index. By calculating the original balance index and coupling it with the environmental correction coefficient, the standardized RMNBI index is generated, which intuitively quantifies the degree of microbial community homeostasis and imbalance.
[0127] In this embodiment of the application, the method for calculating the original ecological balance index based on positive and negative interaction strengths is as follows:
[0128] ;
[0129] in, Let be the original ecological balance index of the i-th sample to be tested. Calculate the relative difference between positive and negative interactions, with a range of . When positive interaction dominates, When negative interactions dominate, When the two are in balance, Add a very small positive number to the denominator. (like This prevents the denominator from being zero.
[0130] In this embodiment of the application, the method for correcting the original ecological balance index using an environmental correction coefficient is as follows:
[0131] ;
[0132] in, Let i be the ecological balance index of the i-th sample to be tested. Let be the environmental correction coefficient for the i-th sample. If the sample has physicochemical index data, then... Calculate according to formula S4 in step S4; if not, then .
[0133] The meaning of the ecological balance index:
[0134] Negative interactions dominate, the community is in a state of competitive balance, indicating good health;
[0135] Positive interactions dominate, the community is in a state of excessive cooperation, indicating potential risks;
[0136] The larger the absolute value, the stronger the dominance of that direction.
[0137] S9: Assess the health status of ruminants based on the RMNBI ecological balance index health threshold. This step aims to achieve a graded assessment of ruminant health status. A 95% confidence interval threshold is set based on a healthy herd, and risk levels are determined according to the threshold. This distinguishes between health and disease gradients, supporting disease early warning, disease assessment, and treatment monitoring.
[0138] In this embodiment, the RMNBI ecological balance index health threshold is calculated based on a sample of at least 50 clinically diagnosed healthy dairy cows, using their mean RMNBI value. and standard deviation Taking the 95% confidence interval as the health reference range, the health threshold formula for the ecological balance index is:
[0139] .
[0140] In this embodiment of the application, the health status assessment of ruminants includes:
[0141] like They were deemed healthy or at low risk.
[0142] like If the condition is deemed high-risk, it indicates the presence of subclinical or clinical health issues. The health reference range can be recalculated based on different farms, breeds, or dietary conditions; this invention allows for flexible adjustments.
[0143] Verification of experimental design:
[0144] To verify the reliability of the model, this invention collected rumen fluid through an oral rumen tube at a large-scale ranch in Henan Province, and collected samples from 32 dairy cows with ketosis based on clinical blood ketone levels. At the same time, based on somatic cell count results, the dairy cows with mastitis were divided into three levels: subclinical mastitis, clinical mastitis, and severe mastitis, and a total of 47 samples were collected.
[0145] Figure 2 The RMNBI trend in ketotic dairy cows is shown. The RMNBI of healthy cows was -0.15, falling into the negative range, indicating that the rumen microbiota was dominated by negative interactions (competition), maintaining a stable state of mutual balance. In contrast, the RMNBI of ketotic cows was positive, increasing from 0.18 to 0.85, indicating that positive interactions (cooperation / cross-feeding) dominated their rumen microbiota, exhibiting an over-cooperative metabolic oligopoly characteristic, consistent with the trend of rumen dysfunction in ketosis. As the RMNBI value increased, the severity of ketosis also showed an increasing trend, indicating that this index can not only clearly distinguish between healthy and ketotic states but also has potential value in assessing disease severity and dynamically monitoring treatment effectiveness.
[0146] Figure 3To test the robustness of RMNBI for mastitis, the RMNBI of healthy dairy cows was -0.3, falling into the negative range, indicating that their rumen microbiota was dominated by negative interactions (competition) and the community structure was stable. As the severity of mastitis increased, the RMNBI showed a gradual upward trend: in individuals with subclinical mastitis, one sample was -0.05 and another was 0.3, showing a transition from competition-dominated to cooperation-dominated states; in mastitis individuals, it rose to 0.6, and in individuals with severe mastitis, it reached as high as 0.9, all positive values. This gradient change indicates that the more severe the mastitis, the more dominant the positive interactions (cooperation / cross-feeding) of the rumen microbiota, and the more pronounced the over-cooperative "metabolic oligopoly" characteristics. RMNBI can not only effectively distinguish between healthy and mastitis states but also reflect the severity of the disease, showing potential as an indicator for assessing and monitoring mastitis.
[0147] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A method for evaluating the health status of ruminants based on the rumen microecological balance index, characterized in that, The steps are as follows: S1: Obtain multiple sets of rumen fluid samples from dairy cows, and perform metagenomic sequencing and rumen physicochemical index determination respectively. The first species abundance matrix is obtained from metagenomic sequencing. Multiple functional groups are defined based on rumen metabolic pathways, and the abundance matrix of the first functional group is calculated based on the abundance of metabolic pathways in each functional group. With the mapping relationship between the first species abundance matrix and the first functional group abundance matrix as the objective, a species-functional group mapping spectrum is constructed based on the non-negative least squares method. S2: Calculate the correlation coefficients between functional groups based on the abundance matrix of the first functional group, and construct positive and negative interaction matrices based on the correlation coefficients between functional groups; S3: Construct a functional redundancy matrix by calculating the similarity between functional groups in rumen metabolic pathways; S4: Constructing environmental correction coefficients based on rumen physicochemical index data; S5: 16S sequencing was performed on the rumen fluid samples from dairy cows to obtain the second species abundance matrix; S6: Map the second species abundance matrix using the species-functional group mapping spectrum and normalize it to obtain the second functional group abundance matrix. S7: Calculate the positive interaction strength of all functional group pairs based on the second functional group abundance matrix and the positive interaction matrix. Use the functional redundancy matrix as a correction factor and combine the second functional group abundance matrix and the negative interaction matrix to calculate the negative interaction strength of all functional group pairs. S8: Calculate the original ecological balance index based on the positive and negative interaction strengths, and correct the original ecological balance index using the environmental correction coefficient to obtain the final RMNBI ecological balance index. S9: Assess the health status of ruminants based on the health threshold of the RMNBI ecological balance index for healthy ruminants.
2. The method for evaluating the health status of ruminants based on the rumen microecological balance index according to claim 1, characterized in that, Multiple functional groups are defined based on rumen metabolic pathways, including: Based on the key role of metabolic pathways in rumen energy metabolism, the detectability of metabolic pathways in metagenomic data, and the known associations between metabolic pathways and dairy cow production performance, functional groups are defined using the MetaCyc standardized metabolic pathway database as a reference. At least the following functional groups are defined: fiber degradation, starch degradation, methanogenesis, propionic acid production, lactic acid metabolism, protein degradation, and sulfate reduction.
3. The method for evaluating the health status of ruminants based on the rumen microecological balance index according to claim 2, characterized in that, The first functional group abundance matrix is calculated based on the abundance of metabolic pathways in each functional group, including: for each sample, aggregating the abundance of all metabolic pathways in each functional group to obtain an initial functional group abundance matrix; and normalizing the abundance vector of each functional group in the initial functional group abundance matrix to obtain the first functional group abundance matrix. N is the number of samples, and K is the number of functional groups.
4. The method for evaluating the health status of ruminants based on the rumen microecological balance index according to claim 3, characterized in that, Aiming at the mapping relationship between the first species abundance matrix and the first functional group abundance matrix, a species-functional group mapping spectrum is constructed based on the non-negative least squares method, including: For each functional group, establish a mapping relationship between species abundance and functional group abundance: , This represents the abundance vector of the N samples corresponding to the k-th functional group. Let be the contribution weight vector of species s to functional group k. This is random error; For each functional group, predict the abundance of the functional group. With actual functional group abundance The optimization objective is to minimize the sum of squared errors, while applying non-negative weights. The constraints are solved using the non-negative least squares algorithm to obtain the contribution weight vector of each species to each functional group; The contribution weight vector of each species to each functional group is normalized so that the sum of the contributions of each species to all functional groups is 1, thus obtaining the species-functional group mapping spectrum. S represents the number of microbial species.
5. The method for evaluating the health status of ruminants based on the rumen microecological balance index according to claim 4, characterized in that, When calculating the inter-functional group correlation coefficient based on the first functional group abundance matrix, the Spearman rank correlation coefficient is used to characterize the inter-functional group correlation coefficient: ,in, , Let a and b represent the abundance vectors of the N samples corresponding to the a-th and b-th functional groups, respectively. This represents the function for calculating the Spearman rank correlation coefficient. Construct positive and negative interaction matrices based on the correlation coefficients between functional groups, including: Construct a positive interaction matrix: ; Construct a negative interaction matrix: .
6. The method for evaluating the health status of ruminants based on the rumen microecological balance index according to claim 5, characterized in that, A functional redundancy matrix is constructed by calculating the similarity between functional groups in rumen metabolic pathways, including: The similarity between functional groups a and b in rumen metabolic pathways was calculated using Jaccard similarity. The redundancy matrix is composed of the similarity between multiple functional group pairs. The formula is: ； in, The similarity between functional groups a and b in rumen metabolic pathways. , These represent the function group numbers of function group a and function group b, respectively.
7. The method for evaluating the health status of ruminants based on the rumen microecological balance index according to claim 6, characterized in that, The method for constructing environmental correction coefficients based on rumen physicochemical index data is as follows: ； in, Let m be the measured value of the m-th environmental factor. For the m-th environmental factor, This is a sensitivity coefficient used to control the rate of decrease of the correction coefficient when deviating from the optimal value. For the weighting coefficients, satisfying , This is the product symbol.
8. The method for evaluating the health status of ruminants based on the rumen microecological balance index according to any one of claims 4-7, characterized in that, The formula for mapping the second species abundance matrix using the species-functional group mapping spectrum is as follows: ； in, The second species abundance matrix is the result of prediction. This is the second species abundance matrix obtained from the samples to be tested.
9. The method for evaluating the health status of ruminants based on the rumen microecological balance index according to claim 8, characterized in that, The method for calculating the positive interaction strength of all functional group pairs based on the second functional group abundance matrix and the positive interaction matrix is as follows: ； in, Let be the strength of the positive interaction between functional group a and functional group b. , These represent the functional group abundance of the k-th functional group of the i-th test sample in the second species abundance matrix; The method for calculating the negative interaction strength of all functional group pairs, using the functional redundancy matrix as a correction factor and combining the second functional group abundance matrix and the negative interaction matrix, is as follows: ； in, The negative interaction strength between functional group a and functional group b.
10. The method for evaluating the health status of ruminants based on the rumen microecological balance index according to claim 9, characterized in that, The method for calculating the original ecological balance index based on positive and negative interaction strengths is as follows: ； in, Let be the original ecological balance index of the i-th sample to be tested. It is a very small positive number; The method for correcting the original ecological balance index using environmental correction coefficients is as follows: ； in, Let i be the ecological balance index of the i-th sample to be tested. is the environmental correction coefficient for the i-th sample to be tested.