Tea tree variety processability discrimination method based on characteristic lipid of fresh tea leaves

Abstract

The invention relates to a tea tree variety manufacturability discrimination method based on characteristic lipids of fresh tea leaves. The method comprises the following steps: collecting a sample, screening a lipid marker combination for discriminating the manufacturability of the tea tree variety based on the sample, and establishing a multi-stage classification model based on the lipid marker combination; and carrying out tea tree variety manufacturability discrimination by using the trained model. According to the invention, a lipid marker combination is determined as a discrimination index for identifying the tea tree variety processability, and a new thought and technical means are provided for accurate identification of the tea tree variety processability (suitable for green tea preparation, suitable for black tea preparation, suitable for white tea preparation and suitable for oolong tea preparation); the multi-stage classification model can effectively distinguish tea tree varieties suitable for green tea, black tea, white tea and oolong tea, the area AUC under an ROC curve is close to or equal to 1.0, and experimental operation and data analysis are simple, convenient and high in accuracy and reliability; tea tree resource diversity cognition is deepened, a new tool is provided for tea variety traceability and quality control and evaluation, and accurate tea processing can be conveniently guided.

Description

Technical Field

[0001] The present invention relates to the technical field of testing or analyzing materials by means of measuring the chemical or physical properties of the materials, and in particular to a method for distinguishing the suitability of tea varieties based on characteristic lipids of fresh tea leaves. Background Art

[0002] Tea is made from fresh tea leaves through a series of processing steps. It is favored by many consumers for its excellent flavor and various health benefits. Selecting the right tea variety is the primary prerequisite for tea processing and a key factor influencing tea flavor and quality. Tea varieties are categorized as suitable for processing based on the most suitable tea variety: green tea, black tea, white tea, and oolong tea.

[0003] The suitability of tea varieties is closely related to their endogenous composition. Currently, the most commonly used biochemical indicator for determining tea variety suitability is the phenol-to-amine ratio, which is the ratio of tea polyphenols to amino acids. Generally, fresh tea leaves with a phenol-to-amine ratio of less than 8 are suitable for producing green tea; those between 8 and 15 are suitable for producing both black and green tea; and those greater than 15 are suitable for producing black tea. However, this method is relatively crude and limited to discriminating between trial-produced green and black teas. A reliable and effective method is urgently needed to comprehensively and accurately determine the suitability of tea varieties.

[0004] Chinese patent publication number CN110174472A discloses a method for determining the suitability of tea varieties based on the fatty acid composition of fresh leaves. When the ratio of unsaturated fatty acids to saturated fatty acids is greater than 2, and the ratio of linolenic acid to palmitic acid is greater than 1.2, the tea variety is suitable for fermented tea. When the ratio of unsaturated fatty acids to saturated fatty acids is greater than 2, and the ratio of linolenic acid to palmitic acid is less than 1.2, the tea variety is suitable for non-fermented tea. This method primarily extracts the ratio of unsaturated fatty acids to saturated fatty acids and the ratio of linolenic acid to palmitic acid, and distinguishes whether a tea variety is suitable for fermented tea. However, it is currently unable to further distinguish between specific varieties of fermented and non-fermented tea. Summary of the Invention

[0005] The present invention solves the problems existing in the prior art and provides a method for distinguishing the suitability of tea varieties based on characteristic lipids of fresh tea leaves, which is suitable for quickly and accurately distinguishing suitable green tea, suitable black tea, suitable white tea and suitable oolong tea varieties.

[0006] The technical principle of the present invention is that lipids are an important type of biological molecules in tea. Fat-soluble pigments such as chlorophyll and carotenoids significantly affect the color of tea. Lipids are also the precursors of volatile substances in tea and are crucial to the formation of tea aroma quality. In fact, there are significant differences in the lipid profiles of fresh leaves of tea varieties with different adaptability. Specific lipid molecules are closely related to the adaptability of tea varieties. Therefore, key specific lipid molecules are used as marker combinations to judge the adaptability of tea varieties.

[0007] The technical solution adopted by the present invention is a method for distinguishing the suitability of tea varieties based on characteristic lipids of fresh tea leaves. The method collects samples, screens lipid marker combinations for distinguishing the suitability of tea varieties based on the samples, and establishes a multi-level classification model based on the lipid marker combinations; The trained model is used to identify the adaptability of tea varieties.

[0008] Preferably, the multi-level classification model includes: A first classification unit, used to classify and distinguish between suitable green tea varieties and other suitable varieties; A second classification unit, used to classify and distinguish between black tea varieties suitable for processing and other varieties suitable for processing; A third classification unit, used to classify and distinguish between varieties suitable for making white tea and varieties suitable for making oolong tea; The first classification unit, the second classification unit, and the third classification unit are arranged in sequence.

[0009] Preferably, a confidence module is provided in conjunction with the first classification unit, the second classification unit and the third classification unit.

[0010] Preferably, the lipid marker combination for the first taxonomic unit includes phosphatidylcholine (34:2), the first free fatty acid group, chlorophyll a and monogalactosyldiglyceride (36:3).

[0011] Preferably, the first free fatty acid group includes linolenic acid and palmitic acid.

[0012] Preferably, the lipid marker combination of the second classification unit includes phosphatidylinositol (36:1), phosphatidylserine group, 16:0-glucosyl-campesterol, phosphatidylcholine group, monogalactosyldiglyceride (36:2), digalactosylmonoacylglycerol (18:3) and oleic acid.

[0013] Preferably, the phosphatidylserine group includes phosphatidylserine (40:2) and phosphatidylserine (34:2); the phosphatidylcholine group includes phosphatidylcholine (32:3), phosphatidylcholine (35:3) and phosphatidylcholine (37:2).

[0014] Preferably, the lipid marker combination coordinated with the third classification unit includes the second free fatty acid group, lysophosphatidylcholine (24:0) and monogalactosyldiglyceride (34:4).

[0015] Preferably, the second free fatty acid group includes free fatty acid (20:3), free fatty acid (27:0) and linoleic acid.

[0016] Preferably, the process of using the trained model to identify the suitability of tea varieties includes the following steps: S1 uses liquid chromatography-mass spectrometry to determine the content of characteristic lipid compounds in fresh leaves of the tea plant to be tested; S2 uses a trained multi-level classification model based on the combination of characteristic lipid compounds to gradually identify the suitability of tea varieties.

[0017] The present invention relates to a method for distinguishing the adaptability of tea varieties based on characteristic lipids of fresh tea leaves. The method comprises the following steps: collecting samples, screening a lipid marker combination for distinguishing the adaptability of tea varieties based on the samples, establishing a multi-level classification model based on the lipid marker combination, and distinguishing the adaptability of tea varieties using the trained model.

[0018] The beneficial effects of the present invention are: (1) Clarify the combination of lipid markers as a discriminant indicator for identifying the suitability of tea varieties, and provide new ideas and technical means for the accurate identification of tea varieties suitable for making green tea, suitable for making black tea, suitable for making white tea, and suitable for making oolong tea; (2) The discrimination accuracy of the multi-class classification model can reach 100%, and it can effectively distinguish the tea varieties suitable for making green tea, black tea, white tea, and oolong tea. The area under the receiver operating characteristic (ROC) curve (AUC) is close to or equal to 1.0. The experimental operation and data analysis are simple and convenient, with high accuracy and reliability. (3) Deepen the understanding of tea tree resource diversity, provide new tools for tea variety traceability, quality control and evaluation, and facilitate guidance for precise tea processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 is a flow chart of the method of the present invention; Figure 2 In the figure, (a) is the PLS-DA score scatter plot of fresh leaves of four different tea varieties with different processing suitability (suitable for processing green tea, suitable for processing black tea, suitable for processing white tea, and suitable for processing oolong tea), and (b) is the corresponding hierarchical clustering dendrogram; Figure 3 Schematic diagram of the structure of the multi-level classification model of the present invention; Figure 4 This is a flow chart of the present invention for gradually distinguishing tea varieties with different adaptability using a multi-level classification model; Figure 5In the figure, (a) is the ROC curve obtained by SVM model analysis based on different variable combinations for green tea varieties and other varieties, and (b) is the corresponding characteristic variables; Figure 6 In the figure, (a) shows the ROC curves obtained by performing SVM model analysis on black tea and other varieties based on different variable combinations after removing green tea varieties, and (b) shows the corresponding characteristic variables; Figure 7 In the figure, (a) is the ROC curve obtained by performing SVM model analysis on white tea and oolong tea varieties based on different variable combinations after removing green tea and black tea varieties, and (b) is the corresponding characteristic variable. DETAILED DESCRIPTION

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0021] First, it is agreed that in the text and drawings of the present invention, GT corresponds to green tea, BT corresponds to black tea, WT corresponds to white tea, and OT corresponds to oolong tea.

[0022] like Figure 1 The present invention relates to a method for discriminating the suitability of tea varieties based on characteristic lipids of fresh tea leaves. The method collects samples, screens lipid marker combinations for discriminating the suitability of tea varieties based on the samples, and establishes a multi-level classification model based on the lipid marker combinations; The trained model is used to identify the adaptability of tea varieties.

[0023] In the present invention, the adaptation discrimination of tea varieties using the trained model includes the following steps: S1 uses liquid chromatography-mass spectrometry (LC-MS) technology to determine the content of characteristic lipid compounds in fresh leaves of the tea plant to be tested; Specifically, the following steps are included: S1.1 Collect fresh leaves from different tea varieties. In actual training, fresh leaves from different tea varieties with known suitability are collected. However, when using this method for field testing, fresh leaves from tea varieties with known or unknown suitability, or with the same or different suitability, may be collected, depending on the discrimination requirements. For each tea variety, fresh leaves are randomly collected from no less than five tea trees. All samples adopt the "one bud, one or two leaves" tenderness standard, and the collected fresh leaves are thoroughly mixed. S1.2 Extraction of fresh leaf lipids: Fresh leaf samples were freeze-dried and then ground into powder. An appropriate amount of tea powder (generally 20 mg) was accurately weighed and lipids were extracted using a methanol-methyl tert-butyl ether (MTBE)-water solvent system containing an internal standard. The extract was centrifuged at high speed and the supernatant was collected. S1.3 uses LC-MS technology to quantitatively analyze the characteristic lipid compounds in fresh leaves based on the internal standard method. Specifically, free fatty acids are quantified using the internal standard FFA16:0-d3, hemolytic phospholipids are quantified using LPC (19:0), glycerophospholipids, glyceroglycolipids, and chlorophylls are quantified using the internal standard PE (15:0 / 15:0), sphingolipids and sterol lipids are quantified using the internal standard Cer (d18:1 / 17:0), and acylglycerides are quantified using the internal standard TG (15:0 / 15:0 / 15:0). It should be noted that after evaluation, sphingolipids and acylglycerides were not included in the final combination of 20 characteristic lipid compounds.

[0024] In fact, this method is also used to extract data during the training phase.

[0025] S2 uses a trained multi-level classification model based on the combination of characteristic lipid compounds to gradually identify the suitability of tea varieties.

[0026] like Figure 3 As shown, the multi-level classification model includes: A first classification unit, used to classify and distinguish between suitable green tea varieties and other suitable varieties; A second classification unit, used to classify and distinguish between black tea varieties suitable for processing and other varieties suitable for processing; A third classification unit, used to classify and distinguish between varieties suitable for making white tea and varieties suitable for making oolong tea; The first classification unit, the second classification unit, and the third classification unit are arranged in sequence.

[0027] In the present invention, the backbone of the multi-level classification model is SVM. In practical applications, it is a superposition of three SVMs. By inputting the corresponding lipid marker combination into each SVM, binary classification is achieved, and finally the step-by-step binary classification of the adaptability of the four tea varieties is achieved.

[0028] Specifically, the lipid marker combination corresponding to the first taxonomic unit includes phosphatidylcholine (34:2), the first free fatty acid group, chlorophyll a and monogalactosyldiglyceride (36:3).

[0029] The first free fatty acid group includes linolenic acid and palmitic acid.

[0030] The Chinese and English comparisons of the lipid marker combinations in the first taxonomic unit are shown in Table 1 ; Table 1 Chinese-English comparison of lipid marker combinations Lipid compound name Chinese name of compound PC (34:2) Phosphatidylcholine (34:2) FFA (18:3) Linolenic acid FFA (16:0) Palmitic acid chlorophyll a Chlorophyll a MGDG (36:3) Monogalactosyldiglyceride (36:3) Among the markers in Table 1, chlorophyll can give tea leaves a bright green appearance, which is beneficial to the "green and moist" color quality of green tea. It can be inferred that tea varieties rich in chlorophyll are more suitable for making green tea, but chlorophyll a is not the only marker suitable for making green tea.

[0031] The lipid marker combination for the second classification unit includes phosphatidylinositol (36:1), phosphatidylserine group, 16:0-glucosyl-campesterol, phosphatidylcholine group, monogalactosyldiglycerol (36:2), digalactosylmonoacylglycerol (18:3) and oleic acid.

[0032] The phosphatidylserine group includes phosphatidylserine (40:2) and phosphatidylserine (34:2).

[0033] The phosphatidylcholine group includes phosphatidylcholine (32:3), phosphatidylcholine (35:3) and phosphatidylcholine (37:2).

[0034] The Chinese and English comparisons of lipid marker combinations in the second taxonomic unit are shown in Table 2 ; Table 2 Chinese-English comparison of lipid marker combinations Lipid compound name Chinese name of compound PI (36:1) Phosphatidylinositol (36:1) PS(40:2) Phosphatidylserine (40:2) 16:0-glc-campesterol 16:0-glucosyl-campesterol PC(32:3) Phosphatidylcholine (32:3) PC(35:3) Phosphatidylcholine (35:3) PC(37:2) Phosphatidylcholine (37:2) PS(34:2) Phosphatidylserine (34:2) MGDG (36:2) Monogalactosyldiglyceride (36:2) DGMG (18:3) Digalactosylmonoacylglycerol (18:3) FFA (18:1) Oleic acid The lipid marker combination that matches the third taxonomic unit includes the second free fatty acid group, lysophosphatidylcholine (24:0) and monogalactosyldiglyceride (34:4).

[0035] The second free fatty acid group includes free fatty acid (20:3), free fatty acid (27:0) and linoleic acid.

[0036] The Chinese and English comparisons of lipid marker combinations in the third taxonomic unit are shown in Table 3 ; Table 3 Chinese-English comparison of lipid marker combinations Lipid compound name Chinese name of compound FFA (20:3) Free fatty acids (20:3) LPC(24:0) Lysophosphatidylcholine (24:0) FFA (18:2) Linoleic acid FFA (27:0) Free fatty acids (27:0) MGDG (34:4) Monogalactosyldiglyceride (34:4) Among the markers in Table 3, the very long-chain fatty acids FFA (20:3) and FFA (27:0) can effectively identify tea varieties suitable for making white tea, but FFA (20:3) and FFA (27:0) here are not the only markers suitable for making white tea; and MGDG (34:4), as a typical unsaturated glycolipid, can effectively identify tea varieties suitable for making oolong tea, but MGDG (34:4) here is not the only marker suitable for making oolong tea.

[0037] Furthermore, a confidence module is provided in conjunction with the first classification unit, the second classification unit, and the third classification unit.

[0038] Taking into account that in the process of judging the suitability of tea tree varieties by a multi-level classification model, there are indeed some fresh tea leaves that are suitable for two varieties at the same time. For this situation, generally speaking, the classification of the multi-level classification model will be based on the suitable variety that is triggered first, and thus miss the possibility of some more suitable varieties to be classified next. Therefore, the output results of the first classification unit, the second classification unit, and the third classification unit are respectively judged by the confidence module. The original data of the result with a confidence lower than the preset value is returned to the multi-level classification model and the parameters of the corresponding classification unit are modified until the confidence of the classification result is higher than the preset value; for example, the result obtained by the classification of the second classification unit should be a variety suitable for making black tea, but it is actually also suitable for making oolong tea, and the confidence judgment result of the confidence module at this time is lower than the preset value, which is generally set to 0.8. Therefore, the data at this time is returned to the multi-level classification model, and the parameters of the first classification unit and the second classification unit are adjusted; In the specific implementation process, the confidence determination can be based on expert experience or by setting labels for the training data set. After completing the parameter fine-tuning of the first classification unit, the second classification unit, and the third classification unit, the confidence module is used for sampling detection in the actual test process and no longer participates in the process of judging the suitability of tea varieties. Figure 3 The dashed arrow in the image no longer works.

[0039] A specific embodiment of the present invention is given below.

[0040] (1) Collection and processing of fresh tea leaf samples Fresh leaves from 22 tea varieties were collected, including four suitable green tea varieties (Longjing 43, Longjing Longye, Zhongcha 108, and Zhongcha 302), eight suitable black tea varieties (Fuyun 6, Gaoyaqi, Jianbohuang 13, Ningzhou 2, Wannong 95, Yinghong 9, Xiuhong, and Quyeqi 12), five suitable oolong tea varieties (Maoxie, Tieguanyin, Jinxuan, Jinguanyin, and Dahongpao), and four suitable white tea varieties (Zhenghe Dabai, Fuding Dabai, Fuding Dahao, and Fu'an Dabai). For each tea variety, five individual plants were randomly selected for fresh leaf collection. All samples adhered to the "one bud, one to two leaves" tenderness standard. The collected leaves were immediately freeze-dried and then ground into a powder.

[0041] (2) Lipid component extraction Weigh 20 mg of tea powder, add 300 μL of a methanol solution containing an internal standard, vortex for 30 seconds, then add 1 mL of MTBE solution. Shake for 40 minutes, add 300 μL of pure water, vortex for 30 seconds, and centrifuge at 10,000 rpm for 10 minutes. The internal standards include FFA16:0-d3, LPC (19:0), PE (15:0 / 15:0), Cer (d18:1 / 17:0), and TG (15:0 / 15:0 / 15:0), with concentrations of 1.67 μg / mL, 1.67 μg / mL, 1.00 μg / mL, 1.00 μg / mL, and 1.67 μg / mL, respectively.

[0042] (III) LC-MS analysis Liquid chromatography conditions: ACQUITY UPLC HSS T3 column (2.1×100 mm, 1.8 μm, Waters); mobile phase A: acetonitrile / water = 6:4 (v / v, containing 10 mM ammonium acetate); mobile phase B: isopropanol / water = 9:1 (v / v, containing 10 mM ammonium acetate); gradient: 0–2 min, 32% B; 2–4 min, 60% B; 4–13 min, 97% B; 13–17 min, 97% B; 17–17.1 min, 32% B; 17.1–20 min, 32% B; column temperature: 45°C, injection volume: 1 μL, flow rate: 0.26 mL / min.

[0043] Mass spectrometry operating conditions: high-resolution full-scan mode was used in ESI(+) and ESI(-) modes, with the mass-to-charge ratio (m / z) range of 200-1500 and 180-1500, respectively; electrospray voltage was 3.5 kV, capillary temperature was 300°C, sheath gas was 35 arb, and auxiliary gas was 10 arb.

[0044] (IV) Lipid data analysis Lipid quantification: Free fatty acids were quantified using the internal standard FFA16:0-d3, lysophospholipids were quantified using LPC (19:0), glycerophospholipids, glyceroglycolipids (MGDG, DGMG), and chlorophyll were quantified using PE (15:0 / 15:0), sphingolipids and sterol lipids were quantified using Cer (d18:1 / 17:0), and acylglycerolipids (DG, TG) were quantified using TG (15:0 / 15:0 / 15:0).

[0045] (V) Differences in lipid composition of fresh leaves of tea varieties with different adaptability A total of 430 lipid molecules were detected in fresh leaves of 22 tea varieties using LC-MS technology, of which 385 lipids had good analytical precision and a coefficient of variation (RSD) of less than 20% and were used for subsequent analysis. Figure 2 As shown in the figure, PLS-DA analysis revealed that fresh leaf samples from tea cultivars of varying suitability exhibited a clear clustering trend. Overall, the GT cultivar exhibited the most significant differences in lipid profiles compared to other cultivars, with both located in the left half of the score plot. In contrast, the BT cultivar was distributed in the lower right quadrant of the score plot. Meanwhile, the WT and OT cultivars partially overlapped, located in the upper right quadrant. Similar results were also observed in the hierarchical cluster dendrogram. This result indicates significant differences in the lipid composition of fresh leaves from different tea cultivars.

[0046] (VI) Step-by-step discrimination of tea varieties with different adaptability based on a multi-level classification model Based on the established PLS-DA model, a total of 84 key differential lipids were initially screened using the dual screening criteria of variable weight (VIP) > 1.2 and significant difference (P < 0.05). These 84 key differential lipids were then assigned to the first, second, and third classification units of a multi-class classification model. A support vector machine (SVM) classifier was used for classifying each level, and receiver operating characteristic (ROC) curves were used to evaluate model performance. The ROC curve describes classifier performance by plotting the true positive rate (sensitivity) against the false positive rate (1-specificity). The area under the ROC curve (AUC) is an indicator of model performance. Typically, AUC values ​​range from 0.5 to 1.0 and are divided into five levels: unqualified (0.5-0.6), poor (0.6-0.7), fair (0.7-0.8), good (0.8-0.9), and excellent (0.9-1.0). AUC values ​​closer to 1 indicate better model performance. During the specific implementation process, 84 key differential lipid combinations were used as different variable combinations, and the ROC was used to evaluate the effect of the model. Then, the optimal model and the characteristic variables under the optimal model were selected, and finally 3 groups of 20 lipid marker combinations were obtained.

[0047] According to the clustering results of tea varieties, a multi-level classification model was established, and characteristic lipid combinations were used for step-by-step discrimination, such as Figure 4 As shown, the first classification unit is used to distinguish GT varieties from other varieties (BT, WT and OT), the second classification unit is used to distinguish BT varieties from WT and OT varieties, and the third classification unit is used to distinguish WT varieties from OT varieties.

[0048] In the first step of discrimination, when 5 characteristic variables are selected, the ROC curve AUC value of the model reaches the highest (1), such as Figure 5 (a) According to Figure 5In (b), the five characteristic lipid compounds with the highest frequency of model selection include PC (34:2), FFA (18:3), FFA (16:0), chlorophyll a, and MGDG (36:3). That is, the classification model established using these five markers has perfect discriminative ability (AUC = 1.0, accuracy 100%), and can effectively distinguish GT varieties from other varieties (BT, WT, OT).

[0049] In the second step of discrimination, when 10 characteristic variables are selected, the ROC curve AUC value of the model reaches the highest value (1), such as Figure 6 (a) According to Figure 6 In (b), the 10 most frequently selected characteristic lipid compounds are composed of PI (36:1), PS (40:2), 16:0-glc-campesterol, PC (32:3), PC (35:3), PC (37:2), PS (34:2), MGDG (36:2), DGMG (18:3) and FFA (18:1). That is, the discriminant model constructed based on this group of markers has an AUC value of 1.0 and a prediction accuracy of 100%, which can effectively distinguish BT varieties from WT and OT varieties.

[0050] In the third step, when 5 characteristic variables are selected, the AUC value of the ROC curve of the model reaches the highest (1), as shown in Figure 7 (a) According to Figure 7 In (b), the five most frequently selected characteristic lipid compounds are FFA (20:3), LPC (24:0), FFA (18:2), FFA (27:0) and MGDG (34:4). That is, the discriminant model constructed based on this group of markers can accurately and reliably distinguish WT varieties from OT varieties, and the AUC value can reach 1.0, and the corresponding prediction accuracy can reach 100%.

[0051] It will be understood by those skilled in the art that the multi-class classification model and its application mentioned in the embodiments of the present invention can be provided as a method, system, or computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0052] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0053] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0054] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0055] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0056] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A method for determining the suitability of tea varieties based on characteristic lipids in fresh tea leaves, characterized by: The method collects samples, screens lipid marker combinations for tea variety adaptability discrimination based on the samples, and establishes a multi-level classification model based on the lipid marker combinations; The trained model is used to identify the adaptability of tea varieties.

2. The method for determining the suitability of tea varieties based on characteristic lipids of fresh tea leaves according to claim 1, wherein: The multi-class classification model includes: A first classification unit, used to classify and distinguish between suitable green tea varieties and other suitable varieties; A second classification unit, used to classify and distinguish between black tea varieties suitable for processing and other varieties suitable for processing; A third classification unit, used to classify and distinguish between varieties suitable for making white tea and varieties suitable for making oolong tea; The first classification unit, the second classification unit, and the third classification unit are arranged in sequence.

3. The method for determining the suitability of tea varieties based on characteristic lipids of fresh tea leaves according to claim 2, wherein: A confidence module is provided in conjunction with the first classification unit, the second classification unit and the third classification unit.

4. The method for determining the suitability of tea varieties based on characteristic lipids of fresh tea leaves according to claim 2, wherein: The lipid marker combination corresponding to the first taxon includes phosphatidylcholine (34:2), the first free fatty acid group, chlorophyll a and monogalactosyldiglyceride (36:3).

5. The method for determining the suitability of tea varieties based on characteristic lipids of fresh tea leaves according to claim 4, characterized in that: The first free fatty acid group includes linolenic acid and palmitic acid.

6. The method for determining the suitability of tea varieties based on characteristic lipids of fresh tea leaves according to claim 2, wherein: The lipid marker combination for the second classification unit includes phosphatidylinositol (36:1), phosphatidylserine group, 16:0-glucosyl-campesterol, phosphatidylcholine group, monogalactosyldiglycerol (36:2), digalactosylmonoacylglycerol (18:3) and oleic acid.

7. The method for determining the suitability of tea varieties based on characteristic lipids in fresh tea leaves according to claim 6, characterized in that: The phosphatidylserine group includes phosphatidylserine (40:2) and phosphatidylserine (34:2); the phosphatidylcholine group includes phosphatidylcholine (32:3), phosphatidylcholine (35:3) and phosphatidylcholine (37:2).

8. The method for determining the suitability of tea varieties based on characteristic lipids in fresh tea leaves according to claim 2, wherein: The lipid marker combination that matches the third taxonomic unit includes the second free fatty acid group, lysophosphatidylcholine (24:0) and monogalactosyldiglyceride (34:4).

9. The method for determining the suitability of tea varieties based on characteristic lipids in fresh tea leaves according to claim 8, characterized in that: The second free fatty acid group includes free fatty acid (20:3), free fatty acid (27:0) and linoleic acid.

10. The method for determining the suitability of tea varieties based on characteristic lipids in fresh tea leaves according to claim 1, characterized in that: The following steps are involved in determining the suitability of tea varieties using the trained model: S1 uses liquid chromatography-mass spectrometry to determine the content of characteristic lipid compounds in fresh leaves of the tea plant to be tested; S2 uses a trained multi-level classification model based on the combination of characteristic lipid compounds to gradually identify the suitability of tea varieties.

Images