Method for identifying Luzhou-flavor liquor

Abstract

The invention relates to the technical field of food detection, and particularly discloses a method for identifying Luzhou-flavor liquor, which comprises the following steps: acquiring a plurality of different types of Luzhou-flavor liquor samples as a training set; carrying out non-targeted metabonomics analysis on the training set samples by adopting ultra-high performance liquid chromatography-tandem mass spectrometry, detecting non-volatile compounds in the training set samples, and obtaining composition and relative content data of the training set samples; based on the training set sample information and the non-volatile compound detection result, establishing a qualitative identification model of the Luzhou-flavor liquor by using an orthogonal partial least square discriminant analysis method; and performing non-targeted metabonomics analysis on a to-be-detected Luzhou-flavor liquor sample, and inputting obtained data into the qualitative identification model of the Luzhou-flavor liquor to obtain category information of the Luzhou-flavor liquor. According to the method provided by the invention, the established qualitative identification model of the Luzhou-flavor liquor based on different alcoholic strengths, producing areas and brands is reliable and good in predictability, and a new method can be provided for the qualitative identification of the Luzhou-flavor liquor.

Description

Technical Field

[0001] This invention relates to the field of food testing technology, specifically to a method for identifying strong-aroma baijiu (Chinese liquor). Background Technology

[0002] Strong-aroma baijiu is the largest-volume, most widely consumed, and most influential type of baijiu in China, characterized by its unique style of "rich cellar aroma, mellow and sweet taste, harmonious flavor, and clean, lingering finish." In recent years, driven by the upgrading of baijiu consumption, the development of industrial clusters, and technological advancements, strong-aroma baijiu has continuously integrated modern technology while maintaining traditional craftsmanship, achieving quality optimization and brand enhancement. However, this rapid industrial development has also led to insufficient differentiation in raw materials, processes, and aroma expression among strong-aroma baijiu from different regions, resulting in product style convergence, blurred regional characteristics, and difficulty in meeting consumers' demands for diverse and personalized flavors. Especially within the same production area, where environmental factors are largely consistent, the differences in flavor characteristics of strong-aroma baijiu become even more difficult to distinguish, thus posing higher requirements and challenges for the qualitative identification of strong-aroma baijiu from the same production area. Therefore, there is an urgent need to establish a scientific and accurate qualitative identification method for strong-aroma baijiu from the same production area. Summary of the Invention

[0003] To develop a qualitative identification method for strong-aroma baijiu from the same production area, this invention provides a method for identifying strong-aroma baijiu. The method provided by this invention uses ultra-high performance liquid chromatography-tandem mass spectrometry (UHPLC-MS / MS) to detect and analyze different types of strong-aroma baijiu, establishing a reliable and predictive qualitative identification model for strong-aroma baijiu.

[0004] This invention provides a method for identifying strong-aroma baijiu, comprising the following steps: Untargeted metabolomics analysis was performed on the modeling samples using ultra-high performance liquid chromatography-tandem mass spectrometry to detect non-volatile compounds and obtain their composition and relative content data. The modeling samples were different types of strong-aroma baijiu. The composition and relative content data of non-volatile compounds were used as feature variables. An OPLS-DA classification model was established based on the feature variables and sample attributes. After evaluating the model using 7-fold cross-validation, the variable importance projection value (VIP) and relative content ratio (FC) of each compound were calculated. Finally, based on the criteria of VIP>1 and |log2(FC)|>1, key differential metabolites were screened to establish the identification model. Non-targeted metabolomics analysis was performed on the strong-aroma baijiu samples to be tested, and the obtained data was input into the qualitative identification model of the strong-aroma baijiu to obtain its category information.

[0005] This invention provides a method for identifying different types of strong-aroma baijiu based on non-volatile compounds. The method uses non-targeted metabolomics LC-MS / MS to detect non-volatile compounds, and combines orthogonal partial least squares discriminant analysis (OPLS-DA) to statistically analyze the data. A qualitative identification model for strong-aroma baijiu with different alcohol contents, origins, and brands is established, which can provide a new technical means for the qualitative identification of strong-aroma baijiu.

[0006] Furthermore, the sample attributes include different alcohol contents, different origins, and different brands.

[0007] Furthermore, the steps for performing non-targeted metabolomics analysis on the training set samples include sample pretreatment, instrument detection, and qualitative and quantitative analysis.

[0008] Furthermore, the sample pretreatment is as follows: after centrifuging a strong-aroma baijiu sample, the supernatant is taken and an internal standard is added and mixed.

[0009] Furthermore, the centrifugation conditions are: 8000 r / min~10000 r / min, 20℃~22℃, centrifugation for 20min~25min; the internal standard is a 2-chloro-L-phenylalanine solution.

[0010] Furthermore, the non-targeted metabolomics UPLC-MS / MS detection conditions are as follows: chromatographic column: 100 mm × 2.1 mm, 1.7 μm ACQUITY UPLC HSS T3 column; column temperature: 45℃; flow rate: 0.4 mL / min; mobile phase A: 0.1% formic acid; mobile phase B: methanol containing 0.1% formic acid; gradient elution program: 1% mobile phase B for 12 min to 100%, retain for 12 min; injection volume: 1 μL; mass spectrometry scan range: 67 amu to 1000 amu; electrospray ionization source: spray voltage: 3.2 kV and 2.8 kV; ion source temperature: 320℃; s-lens voltage: 50 V; MS / MS secondary scan: data-dependent scan.

[0011] Furthermore, the qualitative and quantitative analysis includes: identifying compounds using the HMDB database and calculating the relative content of non-volatile compounds using the internal standard method.

[0012] Furthermore, the R obtained from the 7-fold cross-validation... 2 X and Q 2 These are used to evaluate the model's explanatory power for categorical variables and the model's predictive performance, respectively.

[0013] This invention also provides a method for constructing an identification model for strong-aroma baijiu, comprising the following steps: Obtain samples of different categories of strong-aroma baijiu as the training set; The training set samples were analyzed using ultra-high performance liquid chromatography-tandem mass spectrometry (UHPLC-MS / MS) for non-targeted metabolomics analysis to detect non-volatile compounds and obtain their composition and relative content data. Based on the sample information in the training set and the detection results of non-volatile compounds, a qualitative identification model for strong-aroma baijiu was established using orthogonal partial least squares discriminant analysis.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention provides a method for identifying different types of strong-aroma baijiu based on non-volatile compounds. The method uses non-targeted metabolomics LC-MS / MS to detect non-volatile compounds, and combines orthogonal partial least squares discriminant analysis (OPLS-DA) to statistically analyze the data. A qualitative identification model for strong-aroma baijiu with different alcohol contents, origins, and brands is established, which can provide a new technical means for the qualitative identification of strong-aroma baijiu.

[0015] This invention employs non-targeted metabolomics technology to comprehensively explore the composition of non-volatile compounds in strong-aroma baijiu. The sample pretreatment is simple and reproducible, enabling differential analysis of non-volatile compounds. It demonstrates a high accuracy in identifying homogeneous strong-aroma baijiu products, and has high promotional value and broad application prospects. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a composition diagram of non-volatile compounds in an embodiment of the present invention; Figure 2 This is an OPLS-DA model for strong-aroma baijiu with different alcohol contents in the embodiments of the present invention; Figure 3 This is a heatmap of key differentiating compounds of strong-aroma baijiu with different alcohol contents in the embodiments of the present invention; Figure 4 This is an OPLS-DA model for strong-aroma baijiu from different production areas in this invention embodiment; Figure 5 This is a heatmap of key differentiating compounds of strong-aroma baijiu from different production areas in the embodiments of the present invention; Figure 6 The examples in this invention are OPLS-DA models of different brands of strong-aroma baijiu. Figure 7This is a heatmap showing the key differentiating compounds of different brands of strong-aroma baijiu in the embodiments of the present invention. Detailed Implementation

[0018] The specific embodiments of the present invention are described in detail below, but it should be understood that the scope of protection of the present invention is not limited to the specific 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. Unless otherwise specified, the experimental methods described in the embodiments of the present invention are conventional methods, and the materials and reagents used in the following embodiments are commercially available unless otherwise specified.

[0019] This invention provides a method for identifying strong-aroma baijiu, comprising the following steps: Obtain samples of different categories of strong-aroma baijiu as the training set; The training set samples were analyzed using ultra-high performance liquid chromatography-tandem mass spectrometry (UHPLC-MS / MS) for non-targeted metabolomics analysis to detect non-volatile compounds and obtain their composition and relative content data. Based on the sample information in the training set and the detection results of non-volatile compounds, a qualitative identification model for strong-aroma baijiu was established using orthogonal partial least squares discriminant analysis. Non-targeted metabolomics analysis was performed on the strong-aroma baijiu samples to be tested, and the obtained data was input into the qualitative identification model of the strong-aroma baijiu to obtain its category information.

[0020] The OPLS-DA model obtained by this invention is shown in the figure below. Figure 2 , Figure 4 and Figure 6 As shown, through R 2 X, Q 2 These two values ​​are used to determine the reliability of the model, where R... 2 X, Q 2 These represent the model's fit rate and prediction rate, respectively. The closer these values ​​are to 1, the more reliable the model. VIP > 1 and |log2(FC)| > 1 are used as the screening criteria for key differential compounds. The R-value of the identification model for strong-aroma baijiu with different alcohol contents in this embodiment is... 2 X, Q 2 The values ​​were 0.721 and 0.955 respectively, as shown in the following figures. Figure 2 As shown, samples with different alcohol contents can be clearly clustered, and a total of 56 non-volatile substances were screened out, which may be the key differentiators for distinguishing baijiu with different alcohol contents. Figure 3 Clustering heatmaps of key differential compounds; R-squared values ​​for identification models of strong-aroma baijiu from different production areas. 2 X, Q 2 The values ​​were 0.834 and 0.935 respectively, as shown in the following figures. Figure 4As shown, 137 non-volatile substances were screened out. These key non-volatile compounds are the material basis for identifying different origins. Figure 5 Clustering heatmaps of key differential compounds; R-squared values ​​for identification models of different brands of strong-aroma baijiu. 2 X, Q 2 The values ​​were 0.883 and 0.946 respectively, as shown in the results. Figure 6 As shown, a total of 66 non-volatile substances were found, which can clearly classify Yanghe, Shuangou, and other brands. The clustering heatmap of key differentiating compounds is shown below. Figure 7 As shown, based on different category settings, the key differential metabolites exhibit differences in both quantity and type, demonstrating that differences in the quantity and concentration of non-volatile substances in the samples are the material basis for qualitative identification of different wine samples. The model was further validated through permutation verification, and the results show that different identification models are reliable and have good predictive ability.

[0021] Example 1: A method for identifying strong-aroma baijiu.

[0022] 1. Steps for identifying strong-aroma baijiu (1) Samples of strong-aroma baijiu with different alcohol contents, origins, and brands were collected from Jiangsu production areas as research objects. The sample information is shown in Table 1. The samples were divided into three categories according to the origin: Yanghe (YH), ​​Sihong (SH), and Lianshui (LS); according to the brand representativeness: Yanghe (YH), ​​Shuangou (SG), and others (QT); and according to the alcohol content: samples with an alcohol content of 52% vol (H Alcohol) and samples with an alcohol content of 42% vol (L Alcohol).

[0023] Table 1 Sample Information Table (2) The non-volatile compounds in the above-mentioned strong-aroma baijiu samples were detected using non-targeted metabolomics LC-MS / MS to obtain detection data. The specific method is as follows: 0. Sample pretreatment: Take 5 mL of strong-aroma baijiu sample, centrifuge at 10000 rpm and 20℃ for 20 min, take 1.5 mL of supernatant and transfer it to the sample bottle, add 7.5 μL of 100 μg / mL 2-chloro-L-phenylalanine solution as internal standard to be tested; 1. Instrumental analysis: The sample was analyzed using UPLC-MS / MS to obtain material information; Chromatographic conditions included: a 100 mm × 2.1 mm, 1.7 μm ACQUITY UPLC HSS T3 column; column temperature 45 ℃; flow rate 0.4 mL / min; mobile phase A was 0.1% formic acid; mobile phase B was methanol containing 0.1% formic acid; gradient elution program was: 1% mobile phase B for 12 min to 100%, retain for 12 min; injection volume was 1 μL.

[0024] Mass spectrometry conditions included: a scan range of 67 to 1000 amu; an electrospray ionization (ESI) source with spray voltages of 3.2 kV (positive mode) and 2.8 kV (negative mode); an ion source temperature of 320 °C; an s-lens voltage of 50 V; and data-dependent MS / MS secondary scans.

[0025] 2. Qualitative and Quantitative Analysis: All substances identified by positive and negative ions were combined, and duplicates were removed. The first round of screening was conducted based on a molecular weight error (Mw) of less than 3, isotope distribution >75%, and the acquisition of a secondary mass spectrum. Substances detected in three parallel samples were then screened. The compounds were qualitatively identified using HMDB library search analysis. A total of 648 flavor compounds were identified from the strong-aroma baijiu samples, including alkaloids and their derivatives (9), amino acids, peptides and their analogues (39), benzene ring compounds (23), carbohydrates and their compounds (20), fatty acids and their compounds (94), homogeneous nonmetallic compounds (4), hydrocarbons (13), lipids and lipid-like molecules (106), organic acids and their derivatives (96), organic nitrogen compounds (8), organic oxygen compounds (53), organic heterocyclic compounds (81), organic sulfur compounds (21), phenylpropane compounds and polyketides (20), and other compounds (11). The composition of non-volatile compounds is as follows: Figure 1 As shown. 4-Chlorophenylalanine and cinnamic acid acetate at a concentration of 0.5 μg / mL were selected as a mixed internal standard, and the relative content of each substance was calculated using the internal standard method.

[0026] (3) The analysis results of non-volatile compounds were imported into SIMCA 14.1 software, and UV scaling was used to preprocess the data. Then, according to different classification purposes, three orthogonal partial least squares discriminant analysis (OPLS-DA) models were established respectively: a. Different alcohol content identification models: Grouping criteria: The samples were divided into a high alcohol content group (H Alcohol, 52% vol) and a low alcohol content group (L Alcohol, around 42% vol) based on alcohol content.

[0027] Model validation: The explanatory power parameter R of this model 2 X = 0.721 and the predictive ability parameter Q2 = 0.955, indicating that the model is stable and reliable. Further verification through 200 permutation tests confirms that the model is not overfitting (Q2 = -0.724), indicating that the permutation validation results support the model's good predictive ability.

[0028] Differential compound screening: Based on this model, using variable importance projection (VIP) > 1 and |log2(fold change, FC)| > 1 as screening criteria, a total of 56 key differential nonvolatile compounds were identified. Figure 2 The differences in the clustering heatmaps of these substances can serve as a key material basis for the clear distinction between high and low alcohol content samples. Figure 3 ).

[0029] b. Different origin identification model: Grouping criteria: The samples were divided into three groups based on their place of origin: Yanghe (YH), ​​Sihong (SH), and Lianshui (LS).

[0030] Model validation: The R-value of this model 2 X = 0.842 and Q2 = 0.964. Through 200 permutation tests, the model was further confirmed not to be overfitting (Q2 = -0.794), indicating that the permutation validation results support the model's good predictive ability.

[0031] Differentiated compound screening: Based on this model, using variable importance projection (VIP) > 1 and |log2(fold change, FC)| > 1 as screening criteria, 137 key differential nonvolatile compounds were screened out. Figure 4 Its clustering heatmap shows that key differentiating compounds are the crucial basis for distinguishing samples from different origins. Figure 5 ).

[0032] c. Different brand identification models: Grouping criteria: The samples were divided into three groups based on brand: Yanghe (YH), ​​Shuangou (SG), and others (QT).

[0033] Model validation: The R-value of this model 2 X = 0.745 and Q2 = 0.916. Through 200 permutation tests, the model was further confirmed not to be overfitting (Q2 = -0.640), indicating that the permutation validation results support the model's good predictive ability.

[0034] Differential screening: Based on this model, using variable importance projection (VIP) > 1 and |log2(fold change, FC)| > 1 as screening criteria, a total of 66 key differential nonvolatile compounds were identified. Figure 6 Its clustering heatmap shows that samples from the three brands can be clearly classified based on key differential compounds. Figure 7).

[0035] The information of the strong-aroma baijiu samples collected in this embodiment is shown in Table 1. A total of 12 samples of high and low alcohol content were collected from Jiangsu production area, including different series of 3 production areas and 3 brand categories. Analysis showed that alcohol content was the feature with the greatest difference between the samples. The key difference in non-volatile compounds is shown in Table 2. All samples were stored in a room temperature, light-proof and dry environment for later use.

[0036] Table 2. Key Differential Compound Information for High and Low Alcohol Content Strong-Aroma Baijiu Samples This embodiment uses strong-aroma baijiu from Jiangsu province as the sample, which reduces the differences in flavor characteristics caused by geographical conditions to some extent. It also means that the samples may be more homogeneous, increasing the difficulty of sample identification. Based on the differences in the types and contents of non-volatile compounds, this invention uses an identification model established by OPLS-DA for different alcohol contents, different origins, and different brands, achieving reliable predictive performance. This demonstrates that this invention provides a new method for identifying strong-aroma baijiu, and with expanded sample size, it can be extended to the identification of alcoholic beverages with different flavor attributes.

[0037] Although preferred embodiments of the invention have been described, those skilled in the art, once they have learned the basic inventive concept, can make other changes and modifications to these embodiments.

[0038] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for identifying strong-aroma baijiu, characterized in that, Includes the following steps: Untargeted metabolomics analysis was performed on the modeling samples using ultra-high performance liquid chromatography-tandem mass spectrometry to detect non-volatile compounds and obtain their composition and relative content data. The modeling samples were different types of strong-aroma baijiu. The composition and relative content data of non-volatile compounds were used as feature variables. An OPLS-DA classification model was established based on the feature variables and sample attributes. After evaluating the model using 7-fold cross-validation, the variable importance projection value (VIP) and relative content ratio (FC) of each compound were calculated. Finally, based on the criteria of VIP>1 and |log2(FC)|>1, key differential metabolites were screened to establish the identification model. Non-targeted metabolomics analysis was performed on the strong-aroma baijiu samples to be tested, and the obtained data was input into the qualitative identification model of the strong-aroma baijiu to obtain its category information.

2. The method for identifying strong-aroma baijiu according to claim 1, characterized in that, The sample attributes are different alcohol content, different place of origin, and different brand.

3. The method for identifying strong-aroma baijiu according to claim 1, characterized in that, The steps for performing non-targeted metabolomics analysis on the training set samples include sample pretreatment, instrument detection, and qualitative and quantitative analysis.

4. The method for identifying strong-aroma baijiu according to claim 3, characterized in that, The sample pretreatment is as follows: take a strong-aroma baijiu sample, centrifuge it, take the supernatant, add the internal standard and mix well.

5. The method for identifying strong-aroma baijiu according to claim 4, characterized in that, The centrifugation conditions were: 8000 r / min~10000 r / min, 20℃~22℃, centrifugation for 20 min~25 min; the internal standard was a 2-chloro-L-phenylalanine solution.

6. The method for identifying strong-aroma baijiu according to claim 1, characterized in that, The non-targeted metabolomics UPLC-MS / MS detection conditions were as follows: 100 mm × 2.1 mm, 1.7 μm ACQUITY UPLC HSS T3 column; column temperature 45℃; flow rate 0.4 mL / min; mobile phase A was 0.1% formic acid; mobile phase B was methanol containing 0.1% formic acid; gradient elution program was: 1% mobile phase B for 12 min to 100%, retain for 12 min; injection volume was 1 μL; mass spectrometry scan range was selected as 67 amu to 1000 amu; electrospray ionization source, spray voltage was 3.2 kV and 2.8 kV; ion source temperature was 320℃; s-lens voltage was 50 V; MS / MS secondary scan was data-dependent.

7. The method for identifying strong-aroma baijiu according to claim 3, characterized in that, The qualitative and quantitative analysis includes: identifying compounds using the HMDB database and calculating the relative content of non-volatile compounds using the internal standard method.

8. The method for identifying strong-aroma baijiu according to claim 1, characterized in that, The R obtained from the 7-fold cross-validation 2 Y and Q 2 These are used to evaluate the model's explanatory power for categorical variables and the model's predictive performance, respectively.

Images