Sugarcane disease resistance evaluation method based on metabonomics

Abstract

The invention discloses a metabonomics-based sugarcane disease resistance evaluation method which specifically comprises the following steps: S1, experimental design: selecting at least two types of sugarcane varieties including high disease resistance and high susceptibility, and after artificially inoculating target pathogenic bacteria, respectively collecting samples of sugarcane tissues at a plurality of preset time points; meanwhile, an uninoculated control group is set. The invention relates to the technical field of agricultural biology. According to the sugarcane disease resistance evaluation method based on metabonomics, through integration of liquid chromatography-mass spectrometry and gas chromatography-mass spectrometry multidimensional metabonomics technologies, comprehensive capture of volatile and non-volatile metabolites in sugarcane is realized, and the metabolite detection range is greatly expanded. The dynamic response rule of the disease-resistant related metabolites can be accurately captured in combination with time dynamics analysis, so that an accurate evaluation system covering a broad-spectrum chemical space is constructed, and the comprehensiveness and accuracy of disease resistance evaluation are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of agricultural biotechnology, specifically to a method for evaluating the disease resistance of sugarcane based on metabolomics. Background Technology

[0002] As one of the world's most important sugar and bioenergy crops, the stable development of the sugarcane industry is of great significance to national food security and energy strategy.

[0003] According to the patent document, entitled "A System and Method for Identifying Field Leaf Disease Resistance in Large-Scale Sugarcane Germplasm" (Patent Publication No.: CN119709938A, Patent Publication Date: 2025-03-28), the invention includes the following steps: classifying the susceptibility level of sugarcane samples to be tested according to preset sugarcane leaf disease categories, preset sugarcane leaf susceptibility grades, and symptom descriptions; obtaining the disease index of the sugarcane plant based on the susceptibility level of the sugarcane samples; and obtaining the sugarcane leaf disease resistance level and overall disease resistance level based on the disease index and a preset resistance evaluation system. This method can quickly complete the field identification of leaf disease resistance in large-scale sugarcane germplasm resources. Most importantly, this invention comprehensively evaluates sugarcane leaf disease resistance by considering multiple leaf diseases, reflecting the overall disease resistance of sugarcane after infection by various leaf diseases, thus better meeting the practical needs of breeders in sugarcane breeding.

[0004] Based on the above description, existing traditional methods rely on a single metabolomics platform and cannot simultaneously cover volatile and non-volatile metabolites, resulting in incomplete metabolite information. Furthermore, existing methods are mostly based on static or single-time-point metabolite analysis, which cannot reflect the dynamic changes of metabolites over time during disease resistance and make it difficult to achieve early warning. Therefore, this invention provides a metabolomics-based method for evaluating sugarcane disease resistance. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a metabolomics-based method for evaluating sugarcane disease resistance. This method solves the problems of existing traditional methods relying on a single metabolomics platform, which cannot simultaneously cover volatile and non-volatile metabolites, resulting in incomplete metabolite information. Furthermore, existing methods are mostly based on static or single-time-point metabolite analysis, which cannot reflect the temporal dynamic changes of metabolites during disease resistance, making it difficult to achieve early warning.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for evaluating sugarcane disease resistance based on metabolomics, specifically comprising the following steps: S1. Experimental Design: Select sugarcane varieties that include at least two types: highly resistant and highly susceptible. After artificial inoculation with the target pathogen, collect sugarcane tissue samples at multiple pre-set time points. At the same time, set up an uninoculated control group. S2. Sample preparation and multidimensional metabolite extraction: The sample is pretreated and non-volatile and volatile metabolites are extracted from the same sample in parallel using methods suitable for liquid chromatography-mass spectrometry and gas chromatography-mass spectrometry. S3. Multidimensional metabolomics data acquisition: The extracted non-volatile metabolites were qualitatively and quantitatively analyzed using a liquid chromatography-mass spectrometry platform, while the extracted volatile metabolites were qualitatively and quantitatively analyzed using a gas chromatography-mass spectrometry platform to obtain characteristic metabolite ion spectra covering a broad spectrum of chemical space. S4. Construction of time-kinetic data matrix: Integrate the quantitative data of all metabolites obtained by liquid chromatography-mass spectrometry and gas chromatography-mass spectrometry analysis at different varieties, different treatments and different time points, and construct a four-dimensional data matrix containing metabolite identification, variety type, treatment status and time point; S5. Screening of dynamic biomarkers: Based on the four-dimensional data matrix, multivariate statistical analysis and machine learning algorithms are used to screen out key differential metabolites that exhibit specific dynamic change patterns in response to pathogen infection in disease-resistant varieties, as a group of dynamic biomarkers for evaluating disease resistance. S6. Construction and Validation of Disease Resistance Evaluation Model: Using the abundance change data of the dynamic biomarker group over time, a discrimination model that can distinguish the disease resistance level of sugarcane is constructed, and the accuracy and robustness of the model are validated using an independent validation sample set. S7. Disease resistance evaluation application: For an unknown sugarcane variety to be evaluated, repeat steps S1 to S4 to obtain the data of its dynamic biomarker group in the time series, input it into the discriminant model established in step S6, and output its disease resistance evaluation results relative to known disease-resistant varieties.

[0007] Preferably, in S1, the preset multiple time points at least cover the early response period, the intermediate defense establishment period, and the late result stabilization period of pathogen infection, specifically including one or more time points from 0 hours, 12 hours, 24 hours, 48 ​​hours, 72 hours, 7 days, and 14 days after inoculation.

[0008] Preferably, in step S2, the specific process of parallel extraction is as follows: the same milled sugarcane tissue sample is divided into two equal parts; one part is extracted using a methanol-water-chloroform system, and the extract is mainly used for liquid chromatography-mass spectrometry analysis; the other part is extracted using methanol, and after methoxyamine and silanization derivatization, it is used for gas chromatography-mass spectrometry analysis.

[0009] Preferably, in step S3, the liquid chromatography-mass spectrometry platform uses an electrospray ionization source to acquire data in both positive and negative ion modes; the gas chromatography-mass spectrometry platform uses an electron impact ionization source and is equipped with a non-polar or weakly polar chromatographic column to achieve high-resolution separation of metabolites.

[0010] Preferably, in S5, the specific dynamic change pattern includes: in disease-resistant varieties, the key differential metabolites are rapidly and significantly induced to accumulate in the early stage of infection, and the accumulation magnitude and speed are significantly higher than in susceptible varieties; the key differential metabolites are maintained at a high level in disease-resistant varieties, while in susceptible varieties they show no response, delayed response, or premature decay.

[0011] Preferably, in step S5, the screening process specifically includes: A1. At each individual time point, orthogonal partial least squares discriminant analysis was used to compare the differences in metabolites between disease-resistant and disease-susceptible varieties inoculated with pathogens, and metabolites with a variable projection importance index greater than 1.0 were selected as candidates. A2. Perform trajectory analysis on the quantitative data of the candidates over the entire time series, and identify metabolites with dynamic accumulation trajectories that are highly synchronized with the disease resistance phenotype using a clustering algorithm; A3. The metabolites selected through trajectory analysis are input into a random forest or support vector machine model to rank their feature importance. Finally, the top N metabolites in terms of importance are selected to form the dynamic biomarker group, where N is an integer between 10 and 30.

[0012] Preferably, in step S6, the discrimination model is a classification model built based on random forest, support vector machine or deep neural network. The input feature of the model is the standardized abundance value of the dynamic biomarker group at all preset time points or subsets of key time points, and the output of the model is the disease resistance level classification of sugarcane.

[0013] Preferably, after step S7, the method further includes: S8. Elucidation of disease resistance mechanism: Metabolic pathway enrichment analysis was performed on the dynamic biomarker group to determine the core biosynthetic pathway to which it belongs in the sugarcane metabolic network, thereby elucidating the disease resistance physiological mechanism of sugarcane from the perspective of dynamic metabolic regulation.

[0014] Beneficial effects

[0015] This invention provides a method for evaluating sugarcane disease resistance based on metabolomics. Compared with existing technologies, it has the following advantages: 1. This metabolomics-based method for evaluating sugarcane disease resistance integrates multidimensional metabolomics techniques using liquid chromatography-mass spectrometry (LC-MS) and gas chromatography-mass spectrometry (GC-MS), enabling comprehensive capture of volatile and non-volatile metabolites in sugarcane and significantly expanding the detection range of metabolites. Combined with time-kinetic analysis, it can accurately capture the dynamic response patterns of disease resistance-related metabolites, thereby constructing a precise evaluation system covering a broad spectrum of chemical properties and significantly improving the comprehensiveness and accuracy of disease resistance evaluation.

[0016] 2. This method for evaluating sugarcane disease resistance based on metabolomics constructs a four-dimensional data matrix and employs machine learning algorithms to screen key metabolic biomarkers highly synchronized with disease resistance phenotypes from massive dynamic data. The resulting discriminant model not only accurately distinguishes sugarcane disease resistance levels but also possesses early predictive capabilities, providing a powerful prospective screening tool for disease-resistant breeding.

[0017] 3. This method for evaluating sugarcane disease resistance based on metabolomics elucidates the physiological mechanisms of sugarcane's disease resistance through metabolic pathway enrichment analysis. The discovered dynamic biomarker group is directly linked to its biosynthetic pathways, revealing the intrinsic causes of disease resistance from a metabolic regulation perspective, and providing a solid theoretical basis for achieving precise molecular breeding and developing targeted cultivation strategies. Attached Figure Description

[0018] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a flowchart of the screening process in S5 of the present invention. Detailed Implementation

[0019] 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.

[0020] Please see Figures 1-2 This invention provides two technical solutions: A method for evaluating sugarcane disease resistance based on metabolomics includes the following steps: S1. Experimental Design: Select sugarcane varieties that include at least two types: highly resistant and highly susceptible. After artificial inoculation with the target pathogen, collect sugarcane tissue samples at multiple pre-set time points. At the same time, set up an uninoculated control group. S2. Sample preparation and multidimensional metabolite extraction: The sample is pretreated and non-volatile and volatile metabolites are extracted from the same sample in parallel using methods suitable for liquid chromatography-mass spectrometry and gas chromatography-mass spectrometry. S3. Multidimensional metabolomics data acquisition: The extracted non-volatile metabolites were qualitatively and quantitatively analyzed using a liquid chromatography-mass spectrometry platform, while the extracted volatile metabolites were qualitatively and quantitatively analyzed using a gas chromatography-mass spectrometry platform to obtain characteristic metabolite ion spectra covering a broad spectrum of chemical space. S4. Construction of time-kinetic data matrix: Integrate the quantitative data of all metabolites obtained by liquid chromatography-mass spectrometry and gas chromatography-mass spectrometry analysis at different varieties, different treatments and different time points, and construct a four-dimensional data matrix containing metabolite identification, variety type, treatment status and time point; S5. Screening of dynamic biomarkers: Based on the four-dimensional data matrix, multivariate statistical analysis and machine learning algorithms are used to screen out key differential metabolites that exhibit specific dynamic change patterns in response to pathogen infection in disease-resistant varieties, as a group of dynamic biomarkers for evaluating disease resistance. S6. Construction and Validation of Disease Resistance Evaluation Model: Using the abundance change data of the dynamic biomarker group over time, a discrimination model that can distinguish the disease resistance level of sugarcane is constructed, and the accuracy and robustness of the model are validated using an independent validation sample set. S7. Disease resistance evaluation application: For an unknown sugarcane variety to be evaluated, repeat steps S1 to S4 to obtain the data of its dynamic biomarker group in the time series, input it into the discriminant model established in step S6, and output its disease resistance evaluation results relative to known disease-resistant varieties.

[0021] In this embodiment, in S1, the preset multiple time points at least cover the early response period, the intermediate defense establishment period, and the late result stabilization period of pathogen infection, specifically including one or more time points from 0 hours, 12 hours, 24 hours, 48 ​​hours, 72 hours, 7 days, and 14 days after inoculation.

[0022] By setting multiple time points covering the early response period, the mid-term defense establishment period, and the late-term result stabilization period of pathogen infection, the dynamic changes of sugarcane metabolites during pathogen infection can be comprehensively captured. This allows for a more accurate reflection of the physiological response differences between resistant and susceptible varieties at different stages, providing a reliable data foundation for subsequent time-dynamic analysis and enhancing the temporality and comprehensiveness of disease resistance evaluation.

[0023] In this embodiment, the specific process of parallel extraction in S2 is as follows: the same milled sugarcane tissue sample is divided into two parts. One part is extracted using a methanol-water-chloroform system, and the extract is mainly used for liquid chromatography-mass spectrometry analysis. The other part is extracted with methanol and derivatized by methoxyamine and silanization, and then used for gas chromatography-mass spectrometry analysis.

[0024] By dividing the same sample into two equal parts and extracting them separately using a methanol-water-chloroform system and methanol, parallel extraction of non-volatile and volatile metabolites was achieved. This ensured the integrity and representativeness of metabolite information, avoided the omission of metabolites caused by a single extraction method, provided high-quality sample preparation for multidimensional metabolomics data acquisition, and improved the accuracy and coverage of subsequent analyses.

[0025] In this embodiment, in step S3, the liquid chromatography-mass spectrometry platform uses an electrospray ionization source to collect data in both positive and negative ion modes; the gas chromatography-mass spectrometry platform uses an electron impact ionization source and is equipped with a non-polar or weakly polar chromatographic column to achieve high-resolution separation of metabolites.

[0026] By using a liquid chromatography-mass spectrometry (LC-MS) platform (employing an electrospray ionization source in both positive and negative ion modes) and a gas chromatography-mass spectrometry (GC-MS) platform (employing an electron impact ionization source equipped with a non-polar or weakly polar column), high-resolution separation and sensitive detection of metabolites were achieved. Characteristic ion spectra covering a broad chemical spectrum were obtained, thereby improving the reliability and repeatability of qualitative and quantitative analysis of metabolites and laying the foundation for constructing an accurate data matrix.

[0027] In this embodiment, the specific dynamic change pattern in S5 includes: in disease-resistant varieties, the key differential metabolites are rapidly and significantly induced to accumulate in the early stage of infection, and the accumulation magnitude and speed are significantly higher than those in susceptible varieties; the key differential metabolites are maintained at a high level in disease-resistant varieties, while in susceptible varieties they show no response, delayed response, or premature decay.

[0028] By defining a dynamic pattern in which key differential metabolites accumulate rapidly and remain at high levels in disease-resistant varieties in the early stages, while showing no or delayed response in susceptible varieties, we can accurately identify metabolites that are highly correlated with the disease resistance phenotype. These metabolites, as a dynamic group of biomarkers, not only reflect the disease resistance of sugarcane, but also provide early warning indicators, enhancing the sensitivity and specificity of disease resistance evaluation.

[0029] In this embodiment, the screening process in step S5 specifically includes: A1. At each individual time point, orthogonal partial least squares discriminant analysis was used to compare the differences in metabolites between disease-resistant and disease-susceptible varieties inoculated with pathogens, and metabolites with a variable projection importance index greater than 1.0 were selected as candidates. A2. Perform trajectory analysis on the quantitative data of the candidates over the entire time series, and identify metabolites with dynamic accumulation trajectories that are highly synchronized with the disease resistance phenotype using a clustering algorithm; A3. The metabolites selected through trajectory analysis are input into a random forest or support vector machine model to rank their feature importance. Finally, the top N metabolites in terms of importance are selected to form the dynamic biomarker group, where N is an integer between 10 and 30.

[0030] By combining orthogonal partial least squares discriminant analysis, trajectory clustering, and machine learning (such as random forest or support vector machine) in a multi-step screening method, the top N dynamic biomarkers of importance are efficiently screened from massive metabolite data. This ensures that the selected metabolites are highly synchronized with disease resistance in time series, thereby improving the reliability of biomarkers and the predictive performance of the discriminant model, and providing core features for disease resistance evaluation.

[0031] In this embodiment, in step S6, the discrimination model is a classification model built based on random forest, support vector machine or deep neural network. The input feature of the model is the standardized abundance value of the dynamic biomarker group at all preset time points or subsets of key time points. The output of the model is the disease resistance level classification of sugarcane.

[0032] By constructing a classification model based on random forest, support vector machine, or deep neural network, and using the standardized abundance values ​​of dynamic biomarker groups at all time points as input, the disease resistance level of sugarcane can be accurately distinguished. The model has been tested on an independent validation set and has good accuracy and robustness, realizing a rapid and objective evaluation of the disease resistance of unknown varieties, and providing a practical tool for breeding practice.

[0033] In this embodiment, step S7 is followed by: S8. Elucidation of disease resistance mechanism: Metabolic pathway enrichment analysis was performed on the dynamic biomarker group to determine the core biosynthetic pathway to which it belongs in the sugarcane metabolic network, thereby elucidating the disease resistance physiological mechanism of sugarcane from the perspective of dynamic metabolic regulation.

[0034] By enriching metabolic pathways, dynamic biomarker groups were mapped to core biosynthetic pathways in the sugarcane metabolic network, providing an in-depth explanation of the disease resistance physiological mechanism from the perspective of dynamic metabolic regulation. This not only supplements the theoretical basis for disease resistance evaluation but also provides a scientific basis for the development of molecular breeding and targeted cultivation strategies, achieving a leap from phenomenon description to mechanism analysis.

[0035] In summary, sugarcane varieties with at least two types—those with high disease resistance and those with high disease susceptibility—were selected. After artificial inoculation with the target pathogen, sugarcane tissue samples were collected at multiple predetermined time points. A control group without inoculation was also included. The samples were pretreated, and non-volatile and volatile metabolites were extracted from the same sample using methods suitable for liquid chromatography-mass spectrometry (LC-MS) and gas chromatography-mass spectrometry (GC-MS). The extracted non-volatile metabolites were qualitatively and quantitatively analyzed using LC-MS, while the extracted volatile metabolites were qualitatively and quantitatively analyzed using GC-MS, obtaining characteristic metabolite ion spectra covering a broad chemical spectrum. The quantitative data of all metabolites obtained from LC-MS and GC-MS analyses at different varieties, treatments, and time points were integrated to construct a four-dimensional data matrix including metabolite identification, variety type, treatment status, and time point. Based on the four-dimensional data matrix, multivariate statistical analysis and machine learning algorithms are used to screen key differential metabolites that exhibit specific dynamic change patterns in response to pathogen infection in disease-resistant varieties, serving as a dynamic biomarker group for evaluating disease resistance. Using the abundance change data of the dynamic biomarker group over time, a discriminant model capable of distinguishing sugarcane disease resistance levels is constructed, and the accuracy and robustness of the model are verified using an independent validation sample set. For unknown sugarcane varieties to be evaluated, steps S1 to S4 are repeated to obtain the time-series data of the dynamic biomarker group, which is then input into the discriminant model established in step S6, outputting its disease resistance evaluation results relative to known disease-resistant varieties. Metabolic pathway enrichment analysis is performed on the dynamic biomarker group to determine its core biosynthetic pathway within the sugarcane metabolic network, thereby elucidating the disease resistance physiological mechanism of sugarcane from the perspective of dynamic metabolic regulation.

[0036] Furthermore, any content not described in detail in this specification is existing technology known to those skilled in the art.

[0037] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0038] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for evaluating sugarcane disease resistance based on metabolomics, characterized in that: Specifically, the following steps are included: S1. Experimental Design: Select sugarcane varieties that include at least two types: highly resistant and highly susceptible. After artificial inoculation with the target pathogen, collect sugarcane tissue samples at multiple pre-set time points. At the same time, set up an uninoculated control group. S2. Sample preparation and multidimensional metabolite extraction: The sample is pretreated and non-volatile and volatile metabolites are extracted from the same sample in parallel using methods suitable for liquid chromatography-mass spectrometry and gas chromatography-mass spectrometry. S3. Multidimensional metabolomics data acquisition: The extracted non-volatile metabolites were qualitatively and quantitatively analyzed using a liquid chromatography-mass spectrometry platform, while the extracted volatile metabolites were qualitatively and quantitatively analyzed using a gas chromatography-mass spectrometry platform to obtain characteristic metabolite ion spectra covering a broad spectrum of chemical space. S4. Construction of time-kinetic data matrix: Integrate the quantitative data of all metabolites obtained by liquid chromatography-mass spectrometry and gas chromatography-mass spectrometry analysis at different varieties, different treatments and different time points, and construct a four-dimensional data matrix containing metabolite identification, variety type, treatment status and time point; S5. Screening of dynamic biomarkers: Based on the four-dimensional data matrix, multivariate statistical analysis and machine learning algorithms are used to screen out key differential metabolites that exhibit specific dynamic change patterns in response to pathogen infection in disease-resistant varieties, as a group of dynamic biomarkers for evaluating disease resistance. S6. Construction and Validation of Disease Resistance Evaluation Model: Using the abundance change data of the dynamic biomarker group over time, a discrimination model that can distinguish the disease resistance level of sugarcane is constructed, and the accuracy and robustness of the model are validated using an independent validation sample set. S7. Disease resistance evaluation application: For an unknown sugarcane variety to be evaluated, repeat steps S1 to S4 to obtain the data of its dynamic biomarker group in the time series, input it into the discriminant model established in step S6, and output its disease resistance evaluation results relative to known disease-resistant varieties.

2. The method for evaluating sugarcane disease resistance based on metabolomics according to claim 1, characterized in that: In S1, the preset multiple time points at least cover the early response period, the intermediate defense establishment period, and the late result stabilization period of pathogen infection, specifically including one or more time points from 0 hours, 12 hours, 24 hours, 48 ​​hours, 72 hours, 7 days, and 14 days after inoculation.

3. The method for evaluating sugarcane disease resistance based on metabolomics according to claim 1, characterized in that: In S2, the specific process of parallel extraction is as follows: the same milled sugarcane tissue sample is divided into two parts. One part is extracted using a methanol-water-chloroform system, and the extract is mainly used for liquid chromatography-mass spectrometry analysis. The other part is extracted with methanol and derivatized by methoxyamine and silanization, and then used for gas chromatography-mass spectrometry analysis.

4. The method for evaluating sugarcane disease resistance based on metabolomics according to claim 1, characterized in that: In S3, the liquid chromatography-mass spectrometry platform uses an electrospray ionization source to collect data in both positive and negative ion modes; the gas chromatography-mass spectrometry platform uses an electron impact ionization source and is equipped with a non-polar or weakly polar chromatographic column to achieve high-resolution separation of metabolites.

5. The method for evaluating sugarcane disease resistance based on metabolomics according to claim 1, characterized in that: In S5, the specific dynamic change pattern includes: in disease-resistant varieties, the key differential metabolites are rapidly and significantly induced to accumulate in the early stage of infection, and the accumulation magnitude and speed are significantly higher than those in susceptible varieties; the key differential metabolites are maintained at a high level in disease-resistant varieties, while in susceptible varieties they show no response, delayed response, or premature decay.

6. The method for evaluating sugarcane disease resistance based on metabolomics according to claim 1, characterized in that: In step S5, the screening process specifically includes: A1. At each individual time point, orthogonal partial least squares discriminant analysis was used to compare the differences in metabolites between disease-resistant and disease-susceptible varieties inoculated with pathogens, and metabolites with a variable projection importance index greater than 1.0 were selected as candidates. A2. Perform trajectory analysis on the quantitative data of the candidates over the entire time series, and identify metabolites with dynamic accumulation trajectories that are highly synchronized with the disease resistance phenotype using a clustering algorithm; A3. The metabolites selected through trajectory analysis are input into a random forest or support vector machine model to rank their feature importance. Finally, the top N metabolites in terms of importance are selected to form the dynamic biomarker group, where N is an integer between 10 and 30.

7. The method for evaluating sugarcane disease resistance based on metabolomics according to claim 1, characterized in that: In step S6, the discrimination model is a classification model built based on random forest, support vector machine or deep neural network. The input feature of the model is the standardized abundance value of the dynamic biomarker group at all preset time points or subsets of key time points. The output of the model is the disease resistance level classification of sugarcane.

8. The method for evaluating sugarcane disease resistance based on metabolomics according to claim 1, characterized in that: Step S7 is followed by: S8. Elucidation of disease resistance mechanism: Metabolic pathway enrichment analysis was performed on the dynamic biomarker group to determine the core biosynthetic pathway to which it belongs in the sugarcane metabolic network, thereby elucidating the disease resistance physiological mechanism of sugarcane from the perspective of dynamic metabolic regulation.

Images