Dried orange peel variety identification method and system based on fingerprint spectrum and small molecule metabolite

Abstract

The invention provides a dried orange peel variety identification method and system based on fingerprints and small molecule metabolites. The method comprises the following steps: constructing a polysaccharide fingerprint of each dried orange peel variety; performing small molecule metabolite analysis on each dried orange peel variety, obtaining UPLC-Q-TOF-MS data according to an analysis result, and screening out a plurality of target small molecule metabolites from all the small molecule metabolites according to the UPLC-Q-TOF-MS data; respectively acquiring peak areas corresponding to each common peak and each target small molecule metabolite, and filling the corresponding peak areas according to the sample numbers and the feature numbers by taking the sample numbers as row names and the feature numbers as column names to obtain a fusion feature matrix; and constructing a dried orange peel variety distinguishing model according to the fusion feature matrix, and inputting the fusion feature matrix of the to-be-identified dried orange peel into the variety distinguishing model to obtain the variety of the to-be-identified dried orange peel. The method can be used for accurately and efficiently identifying the dried orange peel, especially Xinhui dried orange peel, clovershrub and other varieties with similar morphology.

Description

Technical Field

[0001] The present invention relates to the technical field of tangerine peel variety identification, and in particular to a tangerine peel variety identification method and system based on fingerprints and small molecule metabolites. Background Art

[0002] Tangerine peel is a plant of Rutaceae ( Citrus reticulata Blanco ) and its cultivated varieties of dried mature peel, pungent, bitter, warm in nature, enter the spleen and lung meridians, have the functions of promoting qi and strengthening the spleen, relieving adverse reactions and stopping vomiting, regulating the stomach and appetizing, drying dampness and resolving phlegm, and are widely used in daily health care and disease treatment. There are many varieties of tangerines at present, and the different sources and varieties on the market lead to differences in quality. However, the Chinese Pharmacopoeia (2025 edition) only uses 4 cultivated varieties, including Xinhui Tangerine Peel ( C. reticulata Blanco cv .'Chachi'), Dahongpao ( C.reticulata Blanco cv .'Dahongbao'), Wenzhou mandarin orange ( C. reticulata Blanco cv .'Unshiu') and Fukuju ( C.reticulata Blanco cv 'Tangerina'). However, there are other cultivars available in the market for medicinal purposes, such as Camphor Red ( C. reticulata Blanco cv .Zhangshuensi), Nanfeng mandarin orange ( C.reticulata Blanco cv .Kinokuni), ponkan ( C. reticulata Blanco cv Therefore, exploring the method of differentiating varieties of dried tangerine peel is of great significance for ensuring its quality, clinical efficacy and resource protection.

[0003] The current methods for distinguishing dried tangerine peel mainly include morphological analysis, microscopic differentiation, and index component content analysis. The above methods are limited to traditional morphology, single chemical component detection, and single technology, lacking integration and in-depth analysis, and are difficult to be used for complex situation analysis such as the morphologically similar Xinhui dried tangerine peel and Dahongpao and their extracts. Summary of the Invention

[0004] The purpose of the present invention is to provide a method and system for identifying tangerine peel varieties based on fingerprints and small molecule metabolites, aiming to solve the problem that traditional technology is difficult to use for complex situation analysis, such as the identification of varieties such as Xinhui tangerine peel and Dahongpao, which have similar morphologies, due to traditional morphology, single chemical component detection, single technology, lack of fusion and in-depth analysis.

[0005] In a first aspect, the present invention provides a method for identifying tangerine peel varieties based on fingerprints and small molecule metabolites, characterized in that the method comprises: Constructing polysaccharide fingerprints for each tangerine peel variety, the tangerine peel varieties including at least Dahongpao, Ponkan, Xinhui, and Zhangtouhong, and marking multiple common peaks of all tangerine peel varieties based on all polysaccharide fingerprints, each tangerine peel variety involving multiple samples, each sample corresponding to a polysaccharide fingerprint; Performing small molecule metabolite analysis on each tangerine peel variety, obtaining UPLC-Q-TOF-MS data based on the analysis results, and screening multiple target small molecule metabolites from all small molecule metabolites based on the UPLC-Q-TOF-MS data; Obtain the peak areas corresponding to each common peak and each target small molecule metabolite respectively, use the sample number as the row name and the feature number as the column name, and fill the corresponding peak areas according to the sample number and feature number to obtain a fusion feature matrix; A tangerine peel variety differentiation model is constructed according to the fusion feature matrix, and the fusion feature matrix of the tangerine peel to be identified is input into the variety differentiation model to obtain the variety of the tangerine peel to be identified.

[0006] In some embodiments, the step of constructing a polysaccharide fingerprint for each tangerine peel variety comprises: The tangerine peel sample is ground into powder, and the tangerine peel powder is mixed with ultrapure water, heated and magnetically stirred for extraction, and then filtered, the filtrate is mixed with ethanol for alcohol precipitation, the precipitate is collected, dissolved in water, heated to remove ethanol, and freeze-dried to obtain tangerine peel polysaccharide; After mixing tangerine peel polysaccharide with trifluoroacetic acid, hydrolyze at 80-100°C, cool to room temperature, dissolve with methanol, remove trifluoroacetic acid with nitrogen, and then add ultrapure water to make up to volume, thereby obtaining a polysaccharide hydrolyzate; Mix the polysaccharide hydrolyzate with NaOH solution and pre-column derivatization agent, react in a water bath at 60-80°C, cool, neutralize with HCL, then add chloroform for liquid-liquid extraction, repeat extraction 3 times, and collect the supernatant; The supernatant was tested by high performance liquid chromatography to obtain polysaccharide fingerprint data of the tangerine peel sample.

[0007] In some embodiments, the liquid phase conditions of the high performance liquid chromatograph are as follows: Chromatographic column: Shim-pack GIS; column temperature: 30°C; mobile phase: liquid A is 0.1 mol / L potassium dihydrogen phosphate, and the pH is adjusted to 6.85 with sodium hydroxide, and liquid B is acetonitrile, ratio: 0-50 min: 18% B, 50-75 min: 18%-25%; flow rate: 1.0 mL / min; injection volume: 10 μL; detector wavelength: 245 nm.

[0008] In some embodiments, the steps of performing small molecule metabolite analysis on each tangerine peel variety and obtaining UPLC-Q-TOF-MS data based on the analysis results include: Tangerine peel powder and methanol were mixed in a volume ratio of 1:20-40, ultrasonically extracted, and then filtered. The filtrate was taken, added with an internal standard, and filtered through a 0.22 μm organic nylon membrane. The filtrate was detected by a high-resolution mass spectrometer to obtain UPLC-Q-TOF-MS data of the tangerine peel sample.

[0009] In some embodiments, the step of screening out a plurality of target small molecule metabolites from all small molecule metabolites based on the UPLC-Q-TOF-MS data comprises: According to the UPLC-Q-TOF-MS data, the relative contents of all small molecule metabolites were arranged in ascending order, the median was calculated after sorting, and twice the median was used as the relative content threshold; All small molecule metabolites with relative contents greater than the relative content threshold were screened out.

[0010] In some embodiments, the step of constructing a tangerine peel variety differentiation model based on the fused feature matrix and inputting the fused feature matrix of the tangerine peel to be identified into the variety differentiation model to obtain the tangerine peel variety to be identified comprises: The fusion feature matrix was normalized by Z-Score and then imported into SMICA14.1 software to construct the tangerine peel variety differentiation model, and the VIP value corresponding to each feature number was obtained through fitting; All VIP values ​​were sorted, and target feature numbers greater than the VIP threshold were screened out according to the sorting results. The target markers corresponding to the target feature numbers were obtained, and the relative content range of each target marker under each tangerine peel variety was recorded. The VIP value of each target marker of the tangerine peel to be identified is obtained, so as to determine the variety of the tangerine peel to be identified according to the VIP value of each target marker of the tangerine peel to be identified and the relative content range of each target marker under each tangerine peel variety.

[0011] In a second aspect, the present invention provides a tangerine peel variety identification system based on fingerprints and small molecule metabolites, the system comprising: a fingerprint construction module for constructing polysaccharide fingerprints for various tangerine peel varieties, including at least Dahongpao, Ponkan, Xinhui, and Zhangtouhong, and marking multiple common peaks of all tangerine peel varieties based on all polysaccharide fingerprints, wherein each tangerine peel variety involves multiple samples, and each sample corresponds to a polysaccharide fingerprint; An analysis module is used to perform small molecule metabolite analysis on each tangerine peel variety, obtain UPLC-Q-TOF-MS data based on the analysis results, and screen multiple target small molecule metabolites from all small molecule metabolites based on the UPLC-Q-TOF-MS data; The feature fusion module is used to obtain the peak areas corresponding to each common peak and each target small molecule metabolite, and fill the corresponding peak areas according to the sample number and feature number to obtain a fusion feature matrix. The differentiation model construction module is used to construct a tangerine peel variety differentiation model based on the fusion feature matrix, and input the fusion feature matrix of the tangerine peel to be identified into the variety differentiation model to obtain the variety of the tangerine peel to be identified.

[0012] In a third aspect, the present invention provides a storage medium storing one or more programs, which, when executed by a processor, implement the above-mentioned method for identifying tangerine peel varieties based on fingerprints and small molecule metabolites.

[0013] In a fourth aspect, the present invention provides an electronic device, comprising a memory and a processor, wherein: The memory is used to store computer programs; When the processor is used to execute the computer program stored in the memory, the above-mentioned method for identifying tangerine peel varieties based on fingerprints and small molecule metabolites is implemented.

[0014] Compared with the prior art, the present invention has the following advantages: The present invention comprehensively covers the characteristic information of dried tangerine peel at the polysaccharide and small molecule metabolite levels through comprehensive analysis of polysaccharide fingerprints and target small molecule metabolites, making feature extraction more comprehensive. The construction of a fusion feature matrix achieves effective integration and organization of data, providing a high-quality data foundation for model construction. The dried tangerine peel variety differentiation model constructed based on this can accurately and efficiently identify dried tangerine peel. Specifically, polysaccharide fingerprints for multiple dried tangerine peel varieties such as Dahongpao, Ponkan, Xinhui, and Zhangtouhong are constructed and common peaks are marked. At the same time, small molecule metabolite analysis is performed on each variety to obtain UPLC-Q-TOF-MS data to screen target small molecule metabolites. By obtaining the peak areas of common peaks and target small molecule metabolites, a fusion feature matrix is ​​constructed using sample numbers and feature numbers, and then a dried tangerine peel variety differentiation model is constructed, thereby completing the variety identification of the dried tangerine peel to be identified. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a flow chart of a method for identifying tangerine peel varieties based on fingerprints and small molecule metabolites proposed in one embodiment of the present invention; Figure 2Schematic diagram of polysaccharide fingerprints of four varieties of dried tangerine peel; Figure 3 This is a comparison chart of the effects of obtaining polysaccharide fingerprints using the traditional method and the present invention; Figure 4 This is a scatter plot of PCA analysis of four tangerine peel varieties based on tangerine peel polysaccharides; Figure 5 This is a scatter plot of PLS-DA analysis of four tangerine peel varieties based on tangerine peel polysaccharides; Figure 6 This is a scatter plot of PCA analysis of tangerine peel polysaccharides and small molecule metabolites of four tangerine peel varieties; Figure 7 This is a scatter plot of PLS-DA analysis of tangerine peel polysaccharides and small molecule metabolites of four tangerine peel varieties; Figure 8 Schematic diagram of 200 cross-validation based on PLS-DA analysis of tangerine peel polysaccharides and small molecule metabolites of four tangerine peel varieties; Figure 9 Schematic diagram of VIP value ranking based on PLS-DA analysis of tangerine peel polysaccharides and small molecule metabolites of four tangerine peel varieties; Figure 10 The figure is a schematic diagram of the relative contents of 18 differential markers in four varieties of dried tangerine peel; Figure 11 The PCA analysis scatter plot for the verification of the tangerine peel to be identified; Figure 12 This is a schematic diagram of the structure of a tangerine peel variety identification system based on fingerprints and small molecule metabolites proposed in one embodiment of the present invention.

[0016] The following specific embodiments will further illustrate the present invention in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. Unless otherwise defined, the technical terms or scientific terms used herein should be the common meanings understood by people with ordinary skills in the field to which the invention belongs. The words "including" and similar words used in this article mean that the elements or objects appearing before the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects.

[0018] like Figure 1As shown, an embodiment of the present invention provides a method for identifying tangerine peel varieties based on fingerprints and small molecule metabolites, the method comprising steps S101 to S104, wherein: Step S101: constructing polysaccharide fingerprints for various tangerine peel varieties, including at least Dahongpao, Ponkan, Xinhui, and Zhangtouhong, and marking multiple common peaks of all tangerine peel varieties based on all polysaccharide fingerprints, wherein each tangerine peel variety involves multiple samples, and each sample corresponds to a polysaccharide fingerprint; It should be noted that, in some embodiments, the construction of the polysaccharide fingerprint requires the collection of a large number of tangerine peel samples, including four known varieties: Dahongpao (DHP), Ponkan (PG), Xinhui (XH), and Zhangtouhong (ZTH). Information on some batches is shown in Table 1 below: Table 1 Detailed information of some batches of samples of different varieties

[0019]

[0020] After obtaining different varieties of dried tangerine peel from each batch, all dried tangerine peel samples were ground into powder, and then the dried tangerine peel powder was mixed with ultrapure water, heated and magnetically stirred for extraction, and then filtered. The filtrate was mixed with ethanol for alcohol precipitation, and the precipitate was collected, dissolved with water, heated to remove the ethanol, and freeze-dried to obtain dried tangerine peel polysaccharide; 3 mg of dried tangerine peel polysaccharide was mixed with 1-4 mol / L trifluoroacetic acid, and hydrolyzed at 80-100 °C for 5-8 h. After cooling to room temperature, the trifluoroacetic acid was removed by methanol dissolution and nitrogen blowing, and then ultrapure water was added to make the volume 1 mL to obtain a polysaccharide hydrolyzate; 250 μL of the polysaccharide hydrolyzate was mixed with NaOH solution and a pre-column derivatizing agent, and then reacted in a water bath at 60-80 °C for 50-80 min. After cooling, HCL was added for neutralization, and chloroform was added for liquid-liquid extraction. The extraction was repeated 3 times, and the supernatant was collected and detected by high performance liquid chromatography to obtain the polysaccharide fingerprint data of the dried tangerine peel sample.

[0021] In some embodiments, examples of obtaining polysaccharide fingerprint data are as follows: (1) Sample preparation Tangerine peel powder and ultrapure water were mixed in a 1:33 volume ratio, extracted with magnetic stirring at 83°C for 60 minutes, and then filtered. The filtrate was precipitated with 50% ethanol. The resulting precipitate was dissolved, the alcohol removed, and freeze-dried to obtain tangerine peel polysaccharide. Tangerine peel polysaccharide (3 mg) was mixed with 2 mL of 3 mol / L trifluoroacetic acid and hydrolyzed at 100°C in the dark for 6 hours. After cooling to room temperature, 1 mL of methanol solution was added and evaporated to dryness under nitrogen purge. This process was repeated three times to remove the trifluoroacetic acid. Finally, purified water was added to the volume to 1 mL to obtain the polysaccharide hydrolyzate. 250 μL of the polysaccharide hydrolyzate was placed in a 5 mL stoppered test tube. 500 μL of 0.3 mol / L NaOH solution and 500 μL of 0.5 mol / L 1-phenyl-3-methyl-5-pyrazolone (PMP) methanol solution were added sequentially. The mixture was vortexed and incubated in a 70°C water bath for 60 min. The mixture was removed, allowed to stand, and cooled to room temperature. 500 μL of 0.3 mol / L HCl was added for neutralization. 1 mL of chloroform was added, vortexed, and allowed to stand. The chloroform layer was discarded. The extraction was repeated three times, and the supernatant was collected. The supernatant was filtered through a 0.22 μm filter membrane and stored. High-performance liquid chromatography was used to analyze the polysaccharide fingerprint data of the dried tangerine peel sample.

[0022] (2) Liquid phase conditions Chromatographic column: Shim-pack GIS (4.6×250 mm, 5 μm); column temperature: 30°C; mobile phase: solution A (0.1 mol / L potassium dihydrogen phosphate (pH adjusted to 6.85 with sodium hydroxide) and solution B (acetonitrile), ratio: 0-50 min: 18% B, 50-75 min: 18%-25%; flow rate: 1.0 mL / min; injection volume: 10 μL; detector wavelength: 245 nm.

[0023] (3) Data processing and result analysis The data were imported into the “Chinese Herbal Medicine Chromatographic Fingerprint Similarity Evaluation System” (2012 version) software, S1 was set as the reference chromatogram, the time window width was selected as 0.2 min, the average method was used to generate the fingerprint, and the multi-point correction method was used to match the chromatographic peaks to obtain the fingerprint matching map, see Figure 2 , and similarity was calculated. The results showed that the similarity between all batches of samples and their control patterns was >0.961, indicating that it is difficult to distinguish different varieties of dried tangerine peel using polysaccharide fingerprint analysis alone. In addition, the fingerprint spectrum marked a total of 10 common peaks, of which 7 were assigned to D-mannose (peak 3), L-rhamnose (peak 5), D-galacturonic acid (peak 6), D-glucose (peak 7), D-galactose (peak 8), D-xylose (peak 9), and L-arabinose (peak 10).

[0024] To demonstrate the advantages of the present invention for obtaining polysaccharide fingerprints, the traditional method is as follows: tangerine peel powder and ultrapure water are mixed in a volume ratio of 1:30, extracted at 50°C for 60 minutes, and then filtered. The filtrate is then precipitated with 50% ethanol. The resulting precipitate is dissolved, the alcohol removed, and freeze-dried to obtain tangerine peel polysaccharide. 3 mg of tangerine peel polysaccharide is mixed with 2 mL of 3 mol / L trifluoroacetic acid and hydrolyzed at 70°C in the dark for 6 hours. After cooling to room temperature, 1 mL of methanol solution is added and evaporated to dryness under nitrogen. This process is repeated three times to remove the trifluoroacetic acid. Finally, purified water is added to the volume to 1 mL to obtain the polysaccharide hydrolyzate. 250 μL of the polysaccharide hydrolyzate was transferred to a 5 mL stoppered test tube. 500 μL of 0.3 mol / L NaOH solution and 500 μL of 0.5 mol / L 1-phenyl-3-methyl-5-pyrazolone (PMP) methanol solution were added sequentially. The mixture was vortexed and incubated in a 70°C water bath for 60 min. The mixture was removed, allowed to stand, and cooled to room temperature. 500 μL of 0.3 mol / L HCl was added for neutralization. 1 mL of chloroform was then added, vortexed, and allowed to stand. The chloroform layer was discarded. The extraction was repeated three times, and the supernatant was collected. The supernatant was filtered through a 0.22 μm filter membrane and stored for analysis by high-performance liquid chromatography.

[0025] Results: As shown in Table 2 and Figure 3 The polysaccharide extraction yield of this method was 24.78%, a 25.3% increase compared to the 19.78% yield of the traditional method. The number of peaks obtained by HPLC with this method was 25, a significant 67% increase compared to the 15 peaks obtained by the traditional method. The total peak area obtained by this method was 10,210, an 84.5% increase compared to the 5,533 obtained by the traditional method. This method has a higher extraction rate of tangerine peel polysaccharides, and the number and total peak area of ​​HPLC peaks are greater, which is conducive to the subsequent differentiation of tangerine peel varieties.

[0026] Table 2 Comparison between traditional method and this method

[0027] Step S102: performing small molecule metabolite analysis on each tangerine peel variety, obtaining UPLC-Q-TOF-MS data based on the analysis results, and screening a plurality of target small molecule metabolites from all small molecule metabolites based on the UPLC-Q-TOF-MS data; In this step, the small molecule metabolite analysis process is as follows: tangerine peel powder and methanol are mixed in a volume ratio of 1:20-40, ultrasonically extracted for 20-40 minutes, and then filtered. 1 mL of the filtrate is taken, 50 μL of internal standard is added, and the volume is adjusted to 5 mL. The filtrate is filtered through a 0.22 μm organic nylon membrane, and the filtrate is detected by a high-resolution mass spectrometer to obtain UPLC-Q-TOF-MS data of the tangerine peel sample; the mass spectrometry data are analyzed using MS-DIAL and MS-FINDER software in sequence to obtain the qualitative information of the tangerine peel small molecule metabolites and their peak areas; the obtained UPLC-Q-TOF-MS data are subjected to double median screening to obtain characteristic small molecule metabolites with higher content.

[0028] In some embodiments, examples of small molecule metabolite analysis are as follows: (1) Sample preparation Tangerine peel powder and methanol were mixed in a 1:25 volume ratio, extracted ultrasonically for 30 minutes, and filtered. 1 mL of the filtrate was added with 50 μL of 3-chlorophenylalanine as an internal standard at a concentration of 10 μg / mL. The solution was then diluted to 5 mL with water to obtain the test solution. After filtering through a 0.22 μm organic nylon membrane, the solution was analyzed using an AB Sciex X500R QTOF high-resolution mass spectrometer to obtain UPLC-Q-TOF-MS data for the tangerine peel sample. 10 μL of each test sample was aspirated and mixed thoroughly for use as a quality control sample.

[0029] (2) Chromatographic and mass spectrometry analysis conditions Chromatographic conditions: Chromatographic column: Waters ACQUITY UPLC® BEH-C18 (2.1 × 50 mm, 1.7 μm). Mobile phase: A: 0.01% formic acid in water, B: acetonitrile, gradient elution. Mobile phase A ratio: 0–10 min, 15%–30% B, 10–22 min, 30%–80% B, 22–23 min, 80%–95% B, 23–26 min, 95% B, 26.1–30 min, 15% B. Flow rate: 0.3 mL / min, column temperature: 30°C, injection volume: 3 μL.

[0030] Mass spectrometry conditions: An electrospray ionization (ESI) source was used with both positive and negative ion scan modes. The primary mass spectrometry (MS / MS) scan range was 100–1000 Da, and the secondary mass spectrometry (MS / MS) scan range was 50–1000 Da. The following parameters were used in both positive and negative ion modes: nebulizer gas (GS1) and auxiliary gas (GS2) pressures were both 55 psi; curtain gas pressure (CUR) was 35 psi; and the ion source temperature (TEM) was 550°C. In positive ion mode, the ion spray voltage (ISVF) was 5500 V; the declustering potential (DP) was 80 V, the parent ion collision energy (CE) was 10 V, the product ion collision energy (CE) was 35 V, and the product ion collision energy difference (CES) was 15 V. In negative ion mode, the ion spray voltage (ISVF) was -4500 V; the declustering potential (DP) was -80 V, the parent ion collision energy (CE) was -10 V, the product ion collision energy (CE) was -35 V, and the product ion collision energy difference (CES) was 15 V.

[0031] (3) Identification and screening of small molecule metabolites Untargeted metabolomics data from different varieties of dried tangerine peel collected by UPLC-Q-TOF-MS were imported into MS-DIAL and MS-FINDER software for peak extraction, alignment, and identification. First, raw data were peak extracted and aligned using MS-DIAL v5.2. Data collection parameters included an MS1 ​​delta of 0.01 Da and an MS2 delta of 0.025 Da; a minimum peak height amplitude of 1000, a mass slice width of 0.1 Da, and a retention time delta of 0.1 min. The resulting .mgf file was then exported to MS-FINDER v3.61 for compound identification. The chemical formula of each peak was determined based on the accurate mass of the assigned protonated molecular ion or adduct ion using the built-in Predictor formula and its corresponding isobaric molecular ion. Elemental composition was limited to C, H, O, and N. MS1 and MS2 tolerances were set to 5 ppm and 10 ppm, respectively. Compounds were identified by matching against a self-built database of 977 compounds from plant sources.

[0032] Next, the relative content data of all small molecule metabolites were arranged in ascending order, and the median of these data was calculated. Using twice the median as the relative content threshold, the content of each small molecule metabolite was compared with the relative content threshold. Small molecule metabolites with relative contents higher than twice the median were screened out. These screened metabolites are high-abundance small molecule metabolites, i.e., target small molecule metabolites. For example, 107 high-abundance small molecule metabolites were screened out (as shown in Table 3 below), including 7 organic acids, 21 flavonoids, 14 dihydroflavonoids, 42 polymethoxyflavonoids, 6 coumarins, 5 limonoids, and 12 other classes.

[0033] Table 3 Summary of high-abundance small molecule metabolites in tangerine peel

[0034]

[0035]

[0036] Step S103: Obtain the peak areas corresponding to each common peak and each target small molecule metabolite respectively, use the sample number as the row name and the feature number as the column name, and fill the corresponding peak areas according to the sample number and feature number to obtain a fusion feature matrix; For example, the substances corresponding to the 10 common peaks obtained in the above steps and the 107 target small molecule metabolites are summarized and numbered to obtain characteristic numbers.

[0037] Step S104: constructing a tangerine peel variety differentiation model based on the fused feature matrix, and inputting the fused feature matrix of the tangerine peel to be identified into the variety differentiation model to obtain the tangerine peel variety to be identified.

[0038] It should be noted that the tangerine peel variety differentiation model refers to a PCA or PLS-DA model.

[0039] In addition, if Figure 4 and Figure 5 As shown in the figure, if the polysaccharide fingerprint data of tangerine peel is directly imported into SMICA14.1 software, PCA and PLS-DA models are established. The resulting PCA and PLS-DA scatter plots show that there is a clear classification trend between the two varieties of Ponkan and Zhangtouhong, and there is a certain degree of overlap between Dahongpao and Xinhui, indicating that the polysaccharide fingerprint alone cannot accurately distinguish different varieties of tangerine peel. The discrimination accuracy correlation parameter R obtained by the PCA discrimination model is 2 X=0.632, Q 2 =0.404; the discrimination accuracy correlation parameter R obtained by the PLS-DA discrimination model 2 Y=0.451,Q2 =0.408. The results are shown in Table 4.

[0040] Table 4 Chemometric analysis parameters based on tangerine peel macromolecular polysaccharides and fusion characteristic variables

[0041] In this step, the liquid phase peak area data of the polysaccharide fingerprint and the mass spectrum peak area data of high-abundance small molecule metabolites are merged into a fusion feature matrix, and the Z-Score method is used to normalize the data to obtain the fusion feature variables. The fusion feature variables of all samples are imported into the SMICA14.1 software to construct the PCA or PLS-DA discrimination model. Figures 6 to 8 As shown in the figure, the PCA and PLS-DA scatter plots show that there is an obvious classification trend among the four varieties of Dahongpao, Ponkan, Xinhui, and Zhangtouhong. The discrimination accuracy correlation parameter R obtained by the PCA discrimination model is 2 X and Q 2 were 0.747 and 0.605, respectively, which were significantly higher than the PCA model constructed based on polysaccharide fingerprint data (R 2 X=0.632, Q 2 =0.404); the discrimination accuracy related parameter R obtained by the PLS-DA discrimination model 2 Y and Q 2 were 0.974 and 0.951, respectively, which were significantly higher than the PCA model constructed based on polysaccharide fingerprint data (R 2 Y=0.451,Q 2 =0.408), the results are shown in Table 3. The above research results show that the prediction ability of the PCA model and PLS-DA model based on the fusion feature variables is significantly better than that of the model built using only polysaccharide data. In addition, the PLS-DA model based on the fusion feature variables was cross-validated 200 times and the R 2 is 0.18 (<0.4), Q 2 The value is -0.428 (<0.05), which indicates that the model is not overfitting and has excellent discrimination ability.

[0042] Specifically, the fusion feature matrix is ​​standardized by Z-Score and then imported into SMICA14.1 software to construct the tangerine peel variety differentiation model, and the VIP value corresponding to each feature number is obtained through fitting; then all the VIP values ​​are sorted, and the target feature number greater than the VIP threshold is screened out according to the sorting result, and the target marker corresponding to the target feature number is obtained, and the relative content range of each target marker under each tangerine peel variety is recorded; then the VIP value of each target marker of the tangerine peel to be identified is obtained, so as to determine the variety of the tangerine peel to be identified according to the VIP value of each target marker of the tangerine peel to be identified and the relative content range of each target marker under each tangerine peel variety.

[0043] It should be noted that since each variety has a large number of samples, each shared peak or target small molecule metabolite has multiple peak areas. The peak area populated in the fusion feature matrix is ​​the average of all corresponding peak areas, and the upper and lower limits of each shared peak or target small molecule metabolite are also recorded. Once the target marker is determined from all shared peaks and target small molecule metabolites, the upper and lower limits are the relative content range of the target marker. For example, the relative content range (peak area range) of each target marker is shown in Table 5 below.

[0044] Table 5 Relative content range of target markers

[0045] The VIP values ​​of 117 characteristic variables are sorted as follows: Figure 9 As shown in Table 6, 18 differential markers with VIP values ​​greater than 1.2 were screened out, including apigenin-7-O-(2G-rhamnose) gentioside, 5-hydroxy-3',4',7-trimethoxyflavone, 3-hydroxy-5,7,8-trimethoxyflavone, kaempferol-3-O-sophoroside, 5,7-dimethoxycoumarin, 5,6,7,3',4'-pentamethoxyflavanone, 5-hydroxy-3,6,7,8,3',4'-pentamethoxyflavanone, '-Hexamethoxyflavone, synephrine, 7-hydroxy-3,5,6,8,3',4'-hexamethoxyflavone, 5,6,7,4'-tetramethoxyflavanone, dihydroactin, rutin, 4'-hydroxy-5,6,7-trimethoxyflavone, corymbiferin, luteolin-6,8-di-C-glucoside, trans-isoferulic acid, 5-hydroxy-7,8,3',4'-tetramethoxyflavone, citronellol-3-rhamnoside.

[0046] Table 6 Characteristic variables with VIP values ​​greater than 1.2 in PLS-DA analysis of four varieties of dried tangerine peel

[0047] In addition, if Figure 10The relative contents of 5-hydroxy-3',4',7-trimethoxyflavone, 3-hydroxy-5,7,8-trimethoxyflavone, kaempferol 3-O-robinoside, corymbose, luteolin-6,8-di-C-glucoside and 5-hydroxy-7,8,3',4'-tetramethoxyflavone in Dahongpao were the highest and significantly higher than those in the other three varieties; the relative contents of 5-hydroxy-3,6,7,8,3',4'-hexamethoxyflavone, synephrine, 7-hydroxy-3,5,6,8,3',4'-hexamethoxyflavone, rutin and 4'-hydroxy-5,6,7,-trimethoxyflavone in Ponkan were the highest and significantly higher than those in the other three varieties; the relative contents of 5,6,7,3',4'-pentamethoxyflavanone, 5,6, The relative content of 7,4'-tetramethoxyflavanone was the highest in Xinhui and significantly higher than in the other three varieties. The relative contents of apigenin-7-O-(2G-rhamnosyl)gentioside, 7-hydroxy-3,5,6,8,3',4'-hexamethoxyflavone, and 5,6,7,4'-tetramethoxyflavanone were the lowest in Dahongpao and significantly lower than in the other three varieties. The relative contents of dihydroactin and 5,7-dimethoxycoumarin were the lowest in Zhangtouhong and significantly lower than in the other three varieties. The relative contents of hydroxy-5,7,8-trimethoxyflavone, trans-isoferulic acid, 5-hydroxy-7,8,3',4'-tetramethoxyflavone, and limonin 3-rhamnoside were the lowest in Xinhui and significantly lower than in the other three varieties. Furthermore, the contents of these 18 components in the four varieties of dried tangerine peel showed significant differences. Therefore, relative content analysis further confirmed that the differential markers screened out by the differentiation model can reflect the differences between different varieties of dried tangerine peel.

[0048] like Figure 11 As shown in the figure, the scatter plot results show that there is an obvious classification trend among the four varieties of Dahongpao, Ponkan, Xinhui, and Zhangtouhong. The discrimination accuracy correlation parameter R obtained by the PCA discrimination model is 2 X and Q 2 0.884 and 0.684, respectively, which are significantly higher than the PCA model constructed based on all data (R 2 X=0.747, Q 2 =0.605), further verifying that the screened differential markers can reflect the differences between different varieties of dried tangerine peel.

[0049] In summary, the present invention comprehensively covers the characteristic information of dried tangerine peel at the polysaccharide and small molecule metabolite levels through comprehensive analysis of polysaccharide fingerprints and target small molecule metabolites, making feature extraction more comprehensive; the construction of a fusion feature matrix realizes the effective integration and organization of data, providing a high-quality data foundation for model construction; the dried tangerine peel variety differentiation model constructed based on this can accurately and efficiently identify dried tangerine peel. Specifically, polysaccharide fingerprints for multiple dried tangerine peel varieties such as Dahongpao, Ponkan, Xinhui, and Zhangtouhong are constructed and common peaks are marked. At the same time, small molecule metabolite analysis is performed on each variety to obtain UPLC-Q-TOF-MS data to screen target small molecule metabolites. By obtaining the peak area of ​​the common peak and the target small molecule metabolite, a fusion feature matrix is ​​constructed with the sample number and feature number, and then a dried tangerine peel variety differentiation model is constructed, thereby completing the variety identification of the dried tangerine peel to be identified.

[0050] like Figure 12 As shown, one embodiment of the present invention provides a tangerine peel variety identification system based on polysaccharide fingerprints and small molecule metabolites, the system comprising: a fingerprint construction module 10 for constructing polysaccharide fingerprints for various tangerine peel varieties, including at least Dahongpao, Ponkan, Xinhui, and Zhangtouhong, and marking multiple common peaks of all tangerine peel varieties based on all polysaccharide fingerprints, wherein each tangerine peel variety involves multiple samples, and each sample corresponds to a polysaccharide fingerprint; An analysis module 20 is used to perform small molecule metabolite analysis on each tangerine peel variety, obtain UPLC-Q-TOF-MS data based on the analysis results, and screen multiple target small molecule metabolites from all small molecule metabolites based on the UPLC-Q-TOF-MS data; The feature fusion module 30 is used to obtain the peak areas corresponding to each common peak and each target small molecule metabolite, and fill the corresponding peak areas according to the sample number and feature number using the sample number as the row name and the feature number as the column name to obtain a fusion feature matrix; The differentiation model construction module 40 is used to construct a tangerine peel variety differentiation model based on the fusion feature matrix, and input the fusion feature matrix of the tangerine peel to be identified into the variety differentiation model to obtain the variety of the tangerine peel to be identified.

[0051] On the other hand, the present invention also provides a storage medium having one or more programs stored thereon, which, when executed by a processor, implements the above-mentioned method for identifying tangerine peel varieties based on fingerprints and small molecule metabolites.

[0052] On the other hand, the present invention also proposes an electronic device, including a memory and a processor, wherein the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory to implement the above-mentioned method for identifying tangerine peel varieties based on fingerprints and small molecule metabolites.

[0053] Those skilled in the art will appreciate that the logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device), or in conjunction with such instruction execution system, apparatus, or device. For purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transmit a program for use by an instruction execution system, apparatus, or device, or in conjunction with such instruction execution system, apparatus, or device.

[0054] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting, or processing it in another suitable manner as necessary, and then storing it in a computer memory.

[0055] It should be understood that various components of the present invention may be implemented using hardware, software, firmware, or a combination thereof. In the aforementioned embodiments, multiple steps or methods may be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one or a combination of the following technologies known in the art may be used: a discrete logic circuit having logic gate circuits for implementing logic functions on data signals, an application-specific integrated circuit having suitable combinational logic gate circuits, a programmable gate array (PGA), a field-programmable gate array (FPGA), etc.

[0056] While the embodiments of the present invention have been described in detail above, it will be apparent to those skilled in the art that various modifications and variations of these embodiments are possible. However, it should be understood that such modifications and variations are within the scope and spirit of the present invention as set forth in the claims. Furthermore, the invention described herein is susceptible to other embodiments and may be practiced or implemented in a variety of ways.

Claims

1. A method for identifying tangerine peel varieties based on fingerprints and small molecule metabolites, characterized in that: The method comprises: Constructing polysaccharide fingerprints for each tangerine peel variety, the tangerine peel varieties including at least Dahongpao, Ponkan, Xinhui, and Zhangtouhong, and marking multiple common peaks of all tangerine peel varieties based on all polysaccharide fingerprints, each tangerine peel variety involving multiple samples, each sample corresponding to a polysaccharide fingerprint; Performing small molecule metabolite analysis on each tangerine peel variety, obtaining UPLC-Q-TOF-MS data based on the analysis results, and screening multiple target small molecule metabolites from all small molecule metabolites based on the UPLC-Q-TOF-MS data; Obtain the peak areas corresponding to each common peak and each target small molecule metabolite respectively, use the sample number as the row name and the feature number as the column name, and fill the corresponding peak areas according to the sample number and feature number to obtain a fusion feature matrix; A tangerine peel variety differentiation model is constructed according to the fusion feature matrix, and the fusion feature matrix of the tangerine peel to be identified is input into the variety differentiation model to obtain the variety of the tangerine peel to be identified.

2. The method for identifying tangerine peel varieties based on fingerprints and small molecule metabolites according to claim 1, wherein The steps of constructing the polysaccharide fingerprints of various tangerine peel varieties include: The tangerine peel sample is ground into powder, and the tangerine peel powder is mixed with ultrapure water, heated and magnetically stirred for extraction, and then filtered, the filtrate is mixed with ethanol for alcohol precipitation, the precipitate is collected, dissolved in water, heated to remove ethanol, and freeze-dried to obtain tangerine peel polysaccharide; After mixing tangerine peel polysaccharide with trifluoroacetic acid, hydrolyze at 80-100°C, cool to room temperature, dissolve with methanol, remove trifluoroacetic acid with nitrogen, and then add ultrapure water to make up to volume, thereby obtaining a polysaccharide hydrolyzate; Mix the polysaccharide hydrolyzate with NaOH solution and pre-column derivatization agent, react in a water bath at 60-80°C, cool, neutralize with HCL, then add chloroform for liquid-liquid extraction, repeat extraction 3 times, and collect the supernatant; The supernatant was tested by high performance liquid chromatography to obtain polysaccharide fingerprint data of the tangerine peel sample.

3. The method for identifying tangerine peel varieties based on fingerprint and small molecule metabolites according to claim 2, wherein: The liquid phase conditions of the high performance liquid chromatograph are as follows: Chromatographic column: Shim-packGIS; Column temperature: 30°C; mobile phase: liquid A is 0.1 mol / L potassium dihydrogen phosphate, and the pH is adjusted to 6.85 with sodium hydroxide, and liquid B is acetonitrile, ratio: 0-50 min: 18% B, 50-75 min: 18%-25%; flow rate: 1.0 mL / min; injection volume: 10 μL; detector wavelength: 245 nm.

4. The method for identifying tangerine peel varieties based on fingerprints and small molecule metabolites according to claim 3, wherein The steps of performing small molecule metabolite analysis on each tangerine peel variety and obtaining UPLC-Q-TOF-MS data based on the analysis results include: Tangerine peel powder and methanol were mixed in a volume ratio of 1:20-40, ultrasonically extracted, and then filtered. The filtrate was taken, added with an internal standard, and filtered through a 0.22 μm organic nylon membrane. The filtrate was detected by a high-resolution mass spectrometer to obtain UPLC-Q-TOF-MS data of the tangerine peel sample.

5. The method for identifying tangerine peel varieties based on fingerprints and small molecule metabolites according to claim 4, wherein: The step of screening out a plurality of target small molecule metabolites from all small molecule metabolites according to the UPLC-Q-TOF-MS data comprises: According to the UPLC-Q-TOF-MS data, the relative contents of all small molecule metabolites were arranged in ascending order, the median was calculated after sorting, and twice the median was used as the relative content threshold; All small molecule metabolites with relative contents greater than the relative content threshold were screened out.

6. The method for identifying tangerine peel varieties based on fingerprints and small molecule metabolites according to claim 4, wherein: The step of constructing a tangerine peel variety differentiation model based on the fusion feature matrix and inputting the fusion feature matrix of the tangerine peel to be identified into the variety differentiation model to obtain the variety of the tangerine peel to be identified comprises: The fusion feature matrix was normalized by Z-Score and then imported into SMICA14.1 software to construct the tangerine peel variety differentiation model, and the VIP value corresponding to each feature number was obtained through fitting; All VIP values ​​were sorted, and target feature numbers greater than the VIP threshold were screened out according to the sorting results. The target markers corresponding to the target feature numbers were obtained, and the relative content range of each target marker under each tangerine peel variety was recorded. The VIP value of each target marker of the tangerine peel to be identified is obtained, so as to determine the variety of the tangerine peel to be identified according to the VIP value of each target marker of the tangerine peel to be identified and the relative content range of each target marker under each tangerine peel variety.

7. A tangerine peel variety identification system based on fingerprints and small molecule metabolites, characterized in that: The system comprises: a fingerprint construction module for constructing polysaccharide fingerprints for various tangerine peel varieties, including at least Dahongpao, Ponkan, Xinhui, and Zhangtouhong, and marking multiple common peaks of all tangerine peel varieties based on all polysaccharide fingerprints, wherein each tangerine peel variety involves multiple samples, and each sample corresponds to a polysaccharide fingerprint; An analysis module is used to perform small molecule metabolite analysis on each tangerine peel variety, obtain UPLC-Q-TOF-MS data based on the analysis results, and screen multiple target small molecule metabolites from all small molecule metabolites based on the UPLC-Q-TOF-MS data; The feature fusion module is used to obtain the peak areas corresponding to each common peak and each target small molecule metabolite, and fill the corresponding peak areas according to the sample number and feature number to obtain a fusion feature matrix. The differentiation model construction module is used to construct a tangerine peel variety differentiation model based on the fusion feature matrix, and input the fusion feature matrix of the tangerine peel to be identified into the variety differentiation model to obtain the variety of the tangerine peel to be identified.

8. A storage medium, characterized in that: The storage medium stores one or more programs, which, when executed by the processor, implement the method for identifying tangerine peel varieties based on fingerprints and small molecule metabolites as described in any one of claims 1 to 7.

9. An electronic device, characterized in that: The electronic device comprises a memory and a processor, wherein: The memory is used to store computer programs; When the processor is used to execute the computer program stored in the memory, it implements the method for identifying tangerine peel varieties based on fingerprints and small molecule metabolites as described in any one of claims 1 to 7.

Images