CN-121978191-A - Golden fungus dyeing diagnosis method and system based on characteristic metabolite detection

CN121978191ACN 121978191 ACN121978191 ACN 121978191ACN-121978191-A

Abstract

The invention belongs to the technical field of agricultural intelligent detection, in particular to a tremella aurantialba bacteria-infection diagnosis method and system based on characteristic metabolite detection, according to the method, quantitative values of three characteristic metabolites of azelaic acid, glycerophosphorylcholine and L-arabitol are accurately obtained through optimized ternary solvent extraction and targeted mass spectrometry. And calculating a standardized diagnostic index of the sample relative to a stage health reference by combining the sample growth stage information so as to eliminate growth difference interference. And performing abnormal preliminary screening based on a support vector machine model, and generating first bacteria infection evaluation information containing specific bacteria infection type tendency through similarity matching of metabolic characteristics and historical bacteria infection clustering modes by adopting two-stage intelligent analysis. Meanwhile, second bacteria-staining evaluation information representing group metabolism consistency is generated by calculating the mahalanobis distance between the sample to be tested and the samples in the same batch. Finally, two kinds of evaluation information are fused, a multidimensional decision rule table is queried, and a grading diagnosis result is output.

Inventors

ZHANG YUPING
CHEN MING
ZHANG LILI
XIN LING
ZOU RUIFAN
ZHANG LEI

Assignees

安徽省农业科学院蚕桑研究所

Dates

Publication Date: 20260505
Application Date: 20251224

Claims (10)

1. The tremella aurantialba infection diagnosis method based on characteristic metabolite detection is characterized by comprising the following steps of: Acquiring mass spectrum detection data of a golden fungus sample to be detected, and determining a quantitative response value of a characteristic metabolite according to the mass spectrum detection data, wherein the characteristic metabolite comprises azelaic acid, glycerophosphorylcholine and L-arabitol; according to the quantitative response value of the characteristic metabolite, combining the growth stage information of the golden fungus sample to be detected, and determining a standardized diagnostic index relative to a health reference of a corresponding growth stage; Performing anomaly detection and pattern matching analysis on the standardized diagnostic index to obtain first bacteria-infection evaluation information; Obtaining the metabolic consistency information of the same batch of groups of cultivation units to which the golden fungus sample to be detected belongs, and determining second bacteria-staining evaluation information according to the metabolic consistency information; and determining a final diagnosis result of the tremella aurantialba infection state according to the first infection evaluation information and the second infection evaluation information.
2. The method for diagnosing tremella aurantialba infection based on the detection of characteristic metabolites according to claim 1, wherein the determination of the quantitative response value of the characteristic metabolites according to the mass spectrum detection data comprises: Performing quality control verification on the mass spectrum detection data, wherein the mass spectrum detection data is obtained by acquiring a golden fungus sample to be detected through ultra-high performance liquid chromatography-tandem mass spectrometry; extracting signal intensity of a corresponding target characteristic peak from mass spectrum data passing verification based on preset characteristic metabolite identification information; And correcting the signal intensity of the extracted target characteristic peak by using the signal intensity of the internal standard corresponding to each characteristic metabolite to obtain the quantitative response value of the target metabolite.
3. The method for diagnosing tremella aurantialba infection based on characteristic metabolite detection according to claim 2, wherein the extracting of signal intensity based on the preset characteristic metabolite identification information comprises: locating a corresponding chromatographic peak from the mass spectrum chromatogram according to the retention time window and the accurate mass-to-charge ratio information associated with each characteristic metabolite; Integrating the positioned chromatographic peaks, and calculating the peak areas or peak heights of the chromatographic peaks; And taking the integrated peak area or peak height value as the signal intensity of the target characteristic peak.
4. The method for diagnosing tremella aurantialba infection based on the characteristic metabolite detection according to claim 1, wherein the determining of the standardized diagnostic index with respect to the corresponding growth phase health standard by combining the growth phase information of the tremella aurantialba sample to be tested comprises: obtaining azelaic acid health mean value corresponding to the stage from a pre-constructed staged health metabolite reference library according to the growth stage information And standard deviation of Healthy mean value of glycerophosphorylcholine And standard deviation of Mean value of L-arabitol health And standard deviation of ; The following normalized diagnostic index was calculated: First index of , ; Wherein, the And Respectively corresponding stage health samples Mean and standard deviation of values; Second index , Wherein, the quantitative response value is L-arabitol.
5. The method according to claim 4, wherein the performing anomaly detection and pattern matching analysis on the standardized diagnostic index to obtain first infectious microbe assessment information comprises: Inputting the first index Zr and the second index Zc into a pre-trained pattern recognition model to judge abnormal states; the pattern recognition model is a support vector machine model trained based on healthy tremella aurantialba sample data; if the pattern recognition model judges that the first index Zr and the second index Zc are in the healthy data distribution range, generating first bacteria-infection evaluation information, and representing that the metabolism state of the sample is normal; If the pattern recognition model judges that the first index Zr and the second index Zc deviate from the distribution range of the health data, performing similarity matching on vectors [ Zr, zc ] formed by the first index Zr and the second index Zc and a plurality of predefined typical bacterial infection pattern prototype vectors; And determining the state label associated with the prototype vector of the typical staining pattern with the highest similarity as the first staining evaluation information.
6. The method according to claim 5, wherein the similarity matching of vectors [ Zr, zc ] formed by the first index Zr and the second index Zc with a predefined plurality of prototype vectors of typical staining patterns comprises: And calculating cosine similarity Sk of the vector [ Zr, zc ] and a kth typical infection model prototype vector Pk, wherein the cosine similarity Sk is as follows: ; Wherein, the Representing a vector dot product of the vector, Representing the euclidean norm of the vector; the matching result is to obtain a prototype vector Pk corresponding to the maximum cosine similarity max (Sk) and an associated state label thereof; Wherein, the prototype vector Pk of the typical staining pattern is predefined by: Collecting M historical diagnostic bacteria samples, wherein each sample i corresponds to a characteristic vector ; { For the M eigenvectors Performing density peak-based cluster analysis; for the kth cluster obtained by cluster analysis, the prototype vector Pk is calculated according to the following formula: ; Wherein, the For the number of samples belonging to the kth cluster, A set of all samples in the kth cluster; And according to the main pathogen type or the bacterial contamination stage corresponding to the sample in the kth cluster, a corresponding state label is assigned to the prototype vector Pk.
7. The method of claim 6, wherein the status tag comprises at least one of "defense reaction dominant", "membrane damage dominant", "osmotic stress responsive" or "mixed metabolic disorder".
8. The method for diagnosing tremella aurantialba infection based on characteristic metabolite detection according to claim 6, wherein obtaining metabolic uniformity information of the same batch of groups of cultivation units to which the tremella aurantialba sample to be detected belongs and determining second estimation information of tremella aurantialba infection according to the metabolic uniformity information comprises: Acquiring mass spectrum detection data of at least N other tremella aurantialba samples from the same cultivation unit and the same harvesting period, wherein N is more than or equal to 3; Determining the respective first index Zr and second index Zc according to mass spectrum detection data of the other tremella aurantialba samples; calculating a covariance matrix of a data set formed by [ Zr, zc ] vectors of the other tremella aurantialba samples, and calculating the mahalanobis distance of the [ Zr, zc ] vectors of the tremella aurantialba samples to be detected relative to the data set based on the covariance matrix; And generating second bacteria-staining evaluation information representing the consistency degree of the metabolic modes of the sample to be tested and the group in the same batch based on the Markov distance.
9. The method according to claim 8, wherein determining the final diagnosis result of the tremella aurantialba infection state based on the first and second estimation information comprises: mapping the evaluation state contained in the first bacteria infection evaluation information into a first confidence score, and mapping the consistency level contained in the second bacteria infection evaluation information into a second confidence score; Weighting and summing the first confidence coefficient and the second confidence coefficient according to a preset weight coefficient to obtain a comprehensive decision index; Inquiring a pre-constructed multidimensional decision rule table according to the type of the state label contained in the first bacteria-infection evaluation information, the numerical interval of the comprehensive decision index and the consistency grade of the second bacteria-infection evaluation information; and outputting a corresponding final diagnosis result according to the query result of the multi-dimensional decision rule table.
10. A tremella aurantialba infection diagnosis system based on characteristic metabolite detection for the steps of the tremella aurantialba infection rapid diagnosis method as claimed in any one of claims 1 to 9, characterized in that the system comprises: The mass spectrum data acquisition unit is used for acquiring mass spectrum detection data of the golden fungus sample to be detected; A data preprocessing unit for determining quantitative response values of characteristic metabolites according to the mass spectrometry detection data, wherein the characteristic metabolites comprise azelaic acid, glycerophosphorylcholine and L-arabitol; the index calculation unit is used for determining a standardized diagnostic index relative to a health standard of a corresponding growth stage according to the quantitative response value of the characteristic metabolite and the growth stage information of the golden fungus sample to be detected; The first evaluation unit is used for performing anomaly detection and pattern matching analysis on the standardized diagnostic index to obtain first bacteria-infection evaluation information; the second evaluation unit is used for acquiring the metabolic consistency information of the same batch of groups of the cultivation units to which the golden fungus sample to be detected belongs and determining second bacteria-dyeing evaluation information according to the metabolic consistency information; And the decision fusion unit is used for determining a final diagnosis result of the tremella aurantialba infection state according to the first infection evaluation information and the second infection evaluation information.

Description

Golden fungus dyeing diagnosis method and system based on characteristic metabolite detection Technical Field The invention belongs to the technical field of agricultural intelligent detection, and particularly relates to a tremella aurantialba bacteria-infection diagnosis method and system based on characteristic metabolite detection. Background At present, the detection of edible fungus infection such as tremella aurantialba (Tremella aurantialba) mainly depends on traditional sensory morphological observation, pathogen separation culture and microscopic examination, and molecular biological technologies such as PCR/qPCR. The sensory recognition is carried out by observing the changes of color, smell and the like of fruiting bodies, the method is visual but high in subjectivity, the separation culture method takes days to carry out flat culture and morphological identification of pathogenic bacteria, the accuracy is high but the efficiency is low, the molecular detection technology is used for amplifying pathogenic bacteria DNA based on specific primers, the sensitivity is high but the method can only aim at known pathogenic bacteria, and the physiological state change of a host under the stress of bacteria infection cannot be reflected. In recent years, metabonomics technology, particularly non-targeted metabonomics, has begun to be applied to biotic stress response studies, and the ability to detect the overall change of small molecule metabolites in organisms without bias provides the possibility of discovering biomarkers in disease or stress states. However, in the prior art system, the detection window of the sensory recognition and culture method is seriously lagged, early warning of bacteria infection cannot be realized, the molecular detection technology is limited by known pathogenic primers and lacks recognition capability for mixed infection or unknown infectious bacteria, the existing metabonomics application stays in the laboratory research stage, and the generated massive complex data are required to depend on professional personnel to carry out multi-element statistical analysis (such as PCA and OPLS-DA) with time consumption of hours to days, so that the complex data are difficult to be converted into a standardized diagnosis scheme suitable for production sites. Disclosure of Invention The invention aims to provide a tremella aurantialba infection diagnosis method and a tremella aurantialba infection diagnosis system based on characteristic metabolite detection, so as to establish a tremella aurantialba infection early detection technology which can be directly applied to a production line, is standardized in operation flow and can realize rapid interpretation on the premise of not identifying specific pathogenic bacteria in advance. The invention realizes the above purpose through the following technical scheme: in a first aspect, the present invention provides a method for diagnosing tremella aurantialba infection based on detection of characteristic metabolites, the method comprising: Acquiring mass spectrum detection data of a golden fungus sample to be detected, and determining a quantitative response value of a characteristic metabolite according to the mass spectrum detection data, wherein the characteristic metabolite comprises azelaic acid, glycerophosphorylcholine and L-arabitol; according to the quantitative response value of the characteristic metabolite, combining the growth stage information of the golden fungus sample to be detected, and determining a standardized diagnostic index relative to a health reference of a corresponding growth stage; Performing anomaly detection and pattern matching analysis on the standardized diagnostic index to obtain first bacteria-infection evaluation information; Obtaining the metabolic consistency information of the same batch of groups of cultivation units to which the golden fungus sample to be detected belongs, and determining second bacteria-staining evaluation information according to the metabolic consistency information; and determining a final diagnosis result of the tremella aurantialba infection state according to the first infection evaluation information and the second infection evaluation information. Further, the determining the quantitative response value of the characteristic metabolite according to the mass spectrum detection data comprises the following steps: Performing quality control verification on the mass spectrum detection data, wherein the mass spectrum detection data is obtained by acquiring a golden fungus sample to be detected through ultra-high performance liquid chromatography-tandem mass spectrometry; extracting signal intensity of a corresponding target characteristic peak from mass spectrum data passing verification based on preset characteristic metabolite identification information; And correcting the signal intensity of the extracted target characteristic peak by using the signal intensity of the interna