Three traceability methods of mutton based on mineral fingerprint characteristics

Abstract

The application discloses a sheep meat tracing method based on mineral fingerprint characteristics and belongs to the technical field of food tracing detection. The method takes the longissimus dorsi muscle of Yemule Baiyang, Kazakh sheep and Turpan black sheep in northern Xinjiang as samples, determines the content of 28 kinds of mineral elements by using an ICP-MS method, screens ten kinds of elements B, Na, Mg, V, Zn, As, Rb, Sr, Mo and Ag based on Wilks' Lambda criterion, and constructs a stepwise linear discriminant model, wherein the discriminant accuracy of model back substitution and cross validation reaches 100%. The method has strong pertinence, high precision and good stability, can quickly and accurately identify three kinds of characteristic sheep meat, and provides a reliable technical means for Xinjiang northern characteristic sheep meat origin tracing, variety identification, brand protection and market supervision.

Description

Technical Field

[0001] This invention belongs to the field of food traceability and testing technology, specifically relating to a method for tracing mutton based on mineral fingerprint characteristics. Background Technology

[0002] Lamb is an important source of protein, fat, and a large amount of absorbable minerals, and occupies an important position in food culture. my country has the world's largest sheep population and has become an important pillar industry of the agricultural and rural economy. As an important livestock base in my country, the northern Xinjiang region is one of the important sheep breeding areas in China. Yemule white sheep, Kazakh sheep, and Turpan black sheep are local characteristic germplasm resources (Li Tiantian, Zhang Jie, Li Yingying, et al. Origin, evolution and genetic diversity analysis of Yemule white sheep [J]. China Animal Husbandry Journal, 2024, 60(10): 198-207.). The unique geographical environment and natural resources have nurtured lamb products with regional characteristics, which are loved by consumers. The high added value of regionally distinctive mutton products relies on the quality characteristics bestowed by their unique geographical environment, such as climate, soil, and pasture. However, in today's market, the phenomenon of unclear origins and mixed quality of meat products is becoming increasingly serious. Consumers find it difficult to distinguish the true breed and origin of meat, disrupting market order. This not only affects consumers' rights but also poses a potential threat to food safety (Zhao Ruting. Research on the Origin Traceability of Sunite Mutton, a Geographical Indication Product [D]. Beijing: Chinese Academy of Agricultural Sciences, 2022). Therefore, developing a rapid and accurate method for mutton traceability and identification has become a key issue that urgently needs to be addressed in the current process of high-quality development of the mutton industry.

[0003] In the field of food traceability, mineral elements are ideal traceability markers due to their environmental dependence. Animals obtain their mineral elements primarily from external sources through feed and water. Their content and composition stably reflect the geographical characteristics of the animal's habitat and are not easily affected by processing or storage. Based on mineral fingerprint identification technology, precise analysis of the mineral content and composition in meat can effectively distinguish meat products from different sources, providing a reliable technical means for food traceability (HOU YR, WANG XR, YANG DS, et al. InvestigationTracing the Origin of Tan Sheep Visceral Tissues through Mineral Elements[J].Foods (Basel, Switzerland), 2023, 12(13): 2438). Currently, relevant studies have applied mineral elements to the field of mutton traceability: Wang Baihui et al. successfully established an Albas mutton origin traceability model using seven mineral elements, Ca, Fe, K, Mg, Se, Ti and V, as characteristic indicators (Wang Baihui, Liu Ting, Yang Chenhui, et al. Origin traceability of Albas mutton based on mineral element and stable isotope fingerprint analysis [J]. Journal of Food Safety and Quality Inspection, 2024, 15(16):278-283.). Zhao Laiyu et al. found that mineral elements have a good effect on the identification of mutton breeds in the differentiation between Pishan Red Sheep and Hu Sheep (Zhao Laiyu, Zhang Hongru, Wang Jing, et al. Identification of the authenticity of Xinjiang Pishan Red Sheep based on mineral element fingerprint spectrum [J]. Food Science, 2024, 45(4): 300-306.).

[0004] However, existing technologies still have significant shortcomings: on the one hand, existing traceability methods are mostly aimed at single breeds or specific regions of mutton, and have not yet formed a dedicated traceability scheme for the multiple breeds of distinctive sheep in northern Xinjiang (Yemule white sheep, Kazakh sheep, and Turpan black sheep), making it difficult to adapt to the development needs of the mutton industry in this region; on the other hand, existing studies have selected a limited number of mineral elements, resulting in limited traceability accuracy and applicability, which cannot fully meet the actual needs of identifying the authenticity of distinctive mutton, protecting brands, and regulating the industry.

[0005] Therefore, to address the traceability needs of specialty mutton from northern Xinjiang, developing a highly accurate and applicable mutton traceability method based on mineral element fingerprints is of significant practical importance and application value for protecting local specialty germplasm resources, regulating the mutton market, ensuring food safety, and enhancing industrial competitiveness. Summary of the Invention

[0006] To address the problems of existing technologies, this invention compares the performance of three multivariate statistical models—PCA, LDA, and PLS-DA—in mutton traceability. PCA is an unsupervised dimensionality reduction method that does not rely on predefined groups. Through PCA, we found a certain clustering trend in the mineral elemental composition of the three mutton breeds, proving that breed differentiation based on mineral fingerprints is feasible. However, due to the limited classification effect of PCA and the significant overlap of samples in two-dimensional space, possibly because the principal components failed to fully capture breed-specific elemental information, we further employed supervised PLS-DA to screen key feature elements (VIP>1). Finally, we used LDA to construct a traceability model with a 100% discrimination accuracy.

[0007] On the one hand, this invention provides a method for constructing a stepwise linear discriminant model for mutton traceability, comprising the following steps:

[0008] (1) The longissimus dorsi muscles of Yemule white sheep, Kazakh sheep and Turpan black sheep in northern Xinjiang were collected as mutton samples.

[0009] (2) The contents of 28 mineral elements in the sample were determined by ICP-MS.

[0010] (3) Screen for mineral elements with significant differences between groups and perform principal component analysis and partial least squares discriminant analysis;

[0011] (4) Select feature elements to construct a stepwise linear discriminant model;

[0012] The stepwise linear discriminant model is as follows:

[0013] (1) Y1=-173.107+9.065X1+0.148X2+0.712X3+15.192X4+0.663X5+22.387X6+12.385X7-6.96X8-73.183X9+31.986X 10 ;

[0014] (2) Y2=-165.077-2.189X1+0.095X2+0.94X3-158.04X4+0.773X5+38.831X6+2.415X7+33.509X8+292.068X9-33.372X 10 ;

[0015] (3) Y3=-134.546+2.357X1+0.095X2+0.816X3+62.503X4+0.411X5-545.71X6-3.467X7+48.268X8+384.474X9-31.441X 10 ;

[0016] In the stepwise linear discriminant model, X1 to X 10 For feature elements ， The contents of B, Na, Mg, V, Zn, As, Rb, Sr, Mo, and Ag are listed in that order.

[0017] Specifically, the mutton sample is the longissimus dorsi muscle of a one-year-old sheep that is free-range and not artificially fed. Preferably, the sample is flash-frozen in liquid nitrogen and then stored at -20°C.

[0018] Specifically, the 28 minerals are B, Na, Mg, Al, P, K, V, Mn, Cu, Zn, As, Rb, Sr, Mo, Ag, Sb, Cs, Ba, La, Ce, Hf, Pt, Ca, Fe, Co, Sn, Nd, and Ir.

[0019] Specifically, the mineral elements that show significant differences in step (3) are B, Na, Mg, Al, P, K, V, Mn, Cu, Zn, As, Rb, Sr, Mo, Ag, Sb, Cs, Ba, La, Ce, Hf, and Pt.

[0020] Specifically, the partial least squares discriminant analysis screened out high contribution elements with VIP>1 as Ag, Na, B, Al, Zn, and As.

[0021] Specifically, the stepwise linear discriminant model is constructed based on Wilks' Lambda criterion, P < 0.05.

[0022] Specifically, it also includes the steps of back-substitution validation and cross-validation of the stepwise linear discriminant model.

[0023] On the one hand, the present invention provides a method for tracing the origin of mutton based on mineral fingerprint characteristics, which takes the content of characteristic elements in the sheep sample to be tested and puts it into a stepwise linear discriminant model constructed using the construction method described in any one of claims 1-6, so as to realize the identification of mutton breed and place of origin.

[0024] Specifically, the characteristic elements are B, Na, Mg, V, Zn, As, Rb, Sr, Mo, and Ag, and preferably, the content is determined by ICP-MS.

[0025] Specifically, by comparing the three values ​​obtained from the stepwise linear discriminant model, the breed corresponding to the maximum value is the classification of the sample to be tested: if Y1 is the largest, it belongs to Yemule white sheep (Emin County); if Y2 is the largest, it belongs to Kazakh sheep (Tekes County); if Y3 is the largest, it belongs to Turpan black sheep (Toksun County), thus realizing the identification of sheep breed and place of origin.

[0026] On the other hand, the present invention relates to the application of the traceability method described above in the identification and origin traceability of distinctive mutton breeds in northern Xinjiang.

[0027] Compared with the prior art, the present invention has the following beneficial effects: (1) The present invention is the first to establish a special mineral fingerprint traceability scheme for Yemule white sheep, Kazakh sheep and Turpan black sheep in northern Xinjiang, filling the gap in traceability technology for multi-variety characteristic sheep meat in the region; (2) Multi-element joint analysis is adopted to screen high contribution characteristic elements, the traceability indicators are more comprehensive, and the discrimination accuracy and applicability are significantly improved; (3) The stepwise linear discrimination model constructed has a 100% accuracy rate in back substitution and cross-validation, the identification results are stable and reliable, and can achieve rapid and accurate traceability; (4) The detection method complies with national standards, the operation is standardized, it is not easily affected by processing and storage, and can effectively support market supervision and brand protection. Attached Figure Description

[0028] Figure 1 This is a radar chart of the eigenvectors of the first three principal components.

[0029] Figure 2 The scatter plot is obtained by standardizing the first two principal components of the three sheep species, where 1 is: Yemule White Sheep; 2 is: Kazakh Sheep; and 3 is: Turpan Black Sheep.

[0030] Figure 3 This is the PLS-DA score chart, where 1 represents: Yemule white sheep; 2 represents: Kazakh sheep; and 3 represents: Turpan black sheep.

[0031] Figure 4 This is the VIP score chart of the PLS-DA model for three types of mutton.

[0032] Figure 5 The following is a stepwise discrimination model score chart for three types of mutton: 1 is Yemule white sheep; 2 is Kazakh sheep; and 3 is Turpan black sheep. Detailed Implementation

[0033] The present invention will be further described below with reference to specific embodiments, and the advantages and features of the present invention will become clearer as a result of the description. However, these embodiments are merely illustrative and do not constitute any limitation on the scope of protection defined by the claims of the present invention.

[0034] It should be understood that the terminology used in this invention is merely for describing particular embodiments and is not intended to limit the invention. Furthermore, with respect to numerical ranges in this invention, it should be understood that the upper and lower limits of the range and each intermediate value between them are specifically disclosed. Any stated value or intermediate value within a stated range, as well as each smaller range between any other stated value or intermediate value within said range, are also included in this invention. The upper and lower limits of these smaller ranges may be independently included or excluded from the range.

[0035] Unless otherwise stated, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. While only preferred methods and materials have been described herein, any methods and materials similar or equivalent to those described herein may be used in the implementation or testing of this invention. All references to this specification are incorporated by way of citation to disclose and describe methods and / or materials associated with those references. In the event of any conflict with any incorporated reference, the content of this specification shall prevail.

[0036] Example 1

[0037] 1. Materials and Methods

[0038] Experimental animals: Samples of grazing local sheep were collected from pastoral areas in northern Xinjiang: 30 Yemule white sheep from Emin County, 30 Kazakh sheep from Tekes County, and 20 Turpan black sheep from Toksun County. 50g of the longissimus dorsi muscle was collected from each sheep, flash-frozen in liquid nitrogen, sealed in sterile self-sealing bags, and stored at -20℃ for later use. All samples were selected from healthy one-year-old individuals, and the rearing environment was free of artificial feeding interventions to ensure the natural expression of genetic background and production performance.

[0039] Reagents and equipment: Petroleum ether, Sinopharm Chemical Reagent Co., Ltd.; Concentrated nitric acid, Fuzhou Shenhui Chemical Instrument Equipment Co., Ltd.; Internal standard (National Nonferrous Metals and Electronic Materials Analysis and Testing Center; National Standard Material Center); 7800 Inductively Coupled Plasma Mass Spectrometer, Agilent Technologies, Inc.; SYG-2 Digital Display Thermostatic Water Bath, Changzhou Langyue Instrument Manufacturing Co., Ltd.; AL204-IC Electronic Balance, Mettler Toledo Instruments (Shanghai) Co., Ltd.

[0040] Mineral content determination: Testing was conducted by a qualified professional institution. Mineral content was determined according to GB5009.268—2016 "Determination of Multiple Elements in Food". 0.20g of mutton sample was accurately weighed into a polytetrafluoroethylene digestion tube, and 6mL of analytical grade nitric acid was added. The sample was allowed to stand at room temperature for 2 hours for pre-digestion. Then, 2mL of hydrogen peroxide was added, and the sample was allowed to stand for 30 minutes. The digestion tube was then capped and tightened, and placed in a microwave digester for digestion. After digestion, the acid was removed, the volume was adjusted, and the content of each mineral element was determined using inductively coupled plasma mass spectrometry (ICP-MS).

[0041] Statistical analysis: SPSS (19.0) and R (4.4.1) software were used to perform analysis of variance, principal component analysis and discriminant analysis on the mineral content of mutton from three types of sheep.

[0042] 2. Analysis of differences in mineral element mass fraction among different sheep breeds

[0043] The longissimus dorsi muscle samples from Yemule white sheep in Emin County, Kazakh sheep in Tekes County, and Turpan black sheep in Toksun County, all located in the pastoral areas of northern Xinjiang, were analyzed for the content of 28 mineral elements. Combined with nutritional element analysis and safety assessment, the results showed that the As content in the three types of mutton ranged from 0.004 to 0.012 mg / kg, which was significantly lower than the limit of 0.15 mg / kg for inorganic arsenic in meat products stipulated in the National Food Safety Standard for Heavy Metals in Food (GB 2762—2017). Other potential risk elements, Pb, Cd, and Hg, were not detected in this study (Liu Xiaorong, Chen Yuan. Research progress on heavy metal pollution in mutton and its detection methods [J]. China Animal Health, 2024, 26(2): 99-100.). In terms of mineral element content characteristics, there were significant differences in mineral element content among Yemule white sheep, Kazakh sheep, and Turpan black sheep, which may be related to their growth environment, feed source, and genetic background. As shown in Table 1, the minerals with significant differences among the three sheep breeds were B, Na, Mg, Al, P, K, V, Mn, Cu, Zn, As, Rb, Sr, Mo, Ag, Sb, Cs, Ba, La, Ce, Hf, and Pt (P<0.05). No statistically significant differences were observed in Ca, Fe, Co, Sn, Nd, and Ir among the groups (P>0.05). Yemule white sheep had higher levels of B, Na, and Al, Kazakh sheep showed superior levels of Mg, P, and Cu, while Turpan black sheep had an advantage in Sr, V, and Sb. The large standard deviation of mineral elements among the different sheep breeds further confirms the influence of geographical environment, such as soil and pasture, on the accumulation of elements in the body. Principal component analysis and linear discriminant analysis can be further used to identify mutton, providing characteristic basis for the construction of subsequent discriminant models.

[0044] Table 1. Mass fraction of mineral elements (mg / kg) in three types of mutton

[0045] ;

[0046] Note: Data from the same row containing the same letter or having no letter in the superscript indicates no significant difference (P>0.05), while data from different rows containing different letters indicates a significant difference (P<0.05).

[0047] 3. Principal component analysis (PCA) of mineral elements in different sheep breeds

[0048] Principal component analysis (PCA) was performed on the contents of 22 mineral elements that showed significant differences among varieties. The eigenvectors and cumulative variance contribution rates of the principal components are shown in Table 2. The variance contribution rate of the first principal component was 25.192%, and that of the second principal component was 22.549%. The cumulative variance contribution rate of the first five principal components reached 70.561%. The 25.192% and 22.549% variance contribution rates of the first and second principal components indicate that the first two principal components are the "most important" information carriers. The cumulative contribution rate of 47.741% indicates that nearly half of the information was retained using only two dimensions. The 70.561% cumulative variance contribution rate of the first five principal components indicates that more than 70% of the information can be retained using five dimensions (good data quality and stable differences). However, since the first two dimensions account for less than 50%, the samples on the two-dimensional graph will inevitably interweave, so LDA is needed for further precise discrimination. The eigenvector diagrams of the first three principal components for the three varieties are shown below. Figure 1 The dominant elements of the first principal component (contribution rate 25.192%) are B, Sr, and Mg; the dominant elements of the second principal component (contribution rate 22.549%) are Na, Rb, and Ag; the dominant elements of the third principal component (contribution rate 10.369%) are P, K, and Mg; the dominant elements of the fourth principal component (contribution rate 7.072%) are Mn, Cs, and Cu; and the dominant elements of the fifth principal component (contribution rate 5.38%) are Sb, Mn, and Cs. A scatter plot was created using the standardized scores of the first two principal components. Figure 2 It can be seen that although different varieties can be clustered together, there is an overlap of samples from different varieties, making them difficult to distinguish, and further discriminant analysis is needed.

[0049] Table 2. Eigenvectors and cumulative variance contribution rates of principal components

[0050] ;

[0051] 4. Partial Least Squares Discriminant Analysis (PLS-DA)

[0052] To further improve the discrimination effect, PLS-DA discriminant analysis was used to select variables that were effective in variety discrimination from 22 elements with significant differences (B, Na, Mg, Al, P, K, V, Mn, Cu, Zn, As, Rb, Sr, Mo, Ag, Sb, Cs, Ba, La, Ce, Hf, Pt). Figure 3 The PLS-DA model shows better discrimination performance compared to the PCA model. In the PLS-DA model, both component 1 and component 2 explain 24% of the total variance. "Component 1 explains 24%" means that the first latent variable explains 24% of the total variance of the model, and "Component 2 explains 24%" means that the second latent variable explains 24% of the total variance of the model. Together, they explain 48%. Figure 4The graph shows the score of the variable important inprojection (VIP) of different mineral elements in the mutton sample. The larger the VIP value, the higher the contribution. If the VIP value of a mineral element is greater than 1, it indicates that the element's contribution to the overall discrimination model is higher than the average level. A VIP value greater than 1.5 indicates that the parameter is a highly significant contributing variable (Gao Xiaoqin. Research on the Origin Tracing of Fuji Apples Based on Mineral Element Content Characteristics [D]. Beijing: Chinese Academy of Agricultural Sciences, 2022). It can be seen that six mineral elements, Ag, Na, B, Al, Zn, and As, have a high contribution to the model, and the model accuracy reaches over 95% (Table 3). The determination of characteristic elements provides important theoretical support for the establishment of mineral element fingerprinting.

[0053] Table 3. Accuracy of PLS-DA discriminant model for three types of mutton

[0054] ;

[0055] 5. Linear Discriminant Analysis (LDA)

[0056] PLS-DA analysis of mineral elements in three types of sheep muscle revealed that mineral elements have a good effect on breed identification and can be used as a key indicator for identifying the authenticity of mutton. Therefore, a stepwise linear discriminant model was established based on differential elements, and the effectiveness of the model was verified. Differential elements were obtained through the following steps: (1) Determination: The content of 28 mineral elements in three types of mutton samples was determined by ICP-MS; (2) Initial screening: One-way ANOVA was used to screen out 22 elements with significant differences between groups (P<0.05); (3) Fine screening: Based on this, stepwise linear discriminant analysis was used, and Wilks' Lambda criterion was used (P<0.05) to automatically screen out the 10 key elements that contributed the most to breed identification, namely: B, Na, Mg, V, Zn, As, Rb, Sr, Mo, and Ag. The selection of these 10 elements is based on objective statistical screening. First, all 10 elements showed significant inter-variety differences in the analysis of variance (P<0.05), providing a statistical basis for distinguishing the three types of mutton. Second, in stepwise linear discriminant analysis, Wilks' Lambda criterion was used to progressively evaluate the contribution of each element to the discriminant model. At each step, the element that best reduced within-group variance and increased between-group variance was introduced, while redundant elements were eliminated. These 10 elements were the optimal feature subset that was ultimately retained. Finally, the LDA model constructed based on these 10 elements achieved a 100% accuracy rate in both back-substitution validation and cross-validation, fully demonstrating the reliability, stability, and generalization ability of this feature element combination.

[0057] Key elements from B, Na, Mg, V, Zn, As, Rb, Sr, Mo, and Ag10 were successively introduced into the discrimination model, and the following discrimination model was established:

[0058] Y1=-173.107+9.065B+0.148Na+0.712Mg+15.192V+0.663Zn+22.387As+12.385Rb-6.96Sr-73.183Mo+31.986Ag

[0059] Y2=-165.077-2.189B+0.095Na+0.94Mg-158.04V+0.773Zn+38.831As+2.415Rb+33.509Sr+292.068Mo-33.372Ag

[0060] Y3=-134.546+2.357B+0.095Na+0.816Mg+62.503V+0.411Zn-545.71As-3.467Rb+48.268Sr+384.474Mo-31.441Ag

[0061] The results were determined by using ICP-MS to analyze the content of 10 elements (B, Na, Mg, V, Zn, As, Rb, Sr, Mo, and Ag) in the mutton samples. Next, the measured contents were substituted into three discriminant functions to calculate three values: Y1, Y2, and Y3. Finally, the three values ​​were compared, and the breed corresponding to the maximum value was determined: if Y1 was the largest, the sample belonged to the Yemule White Sheep (Emin County); if Y2 was the largest, it belonged to the Kazakh Sheep (Tekes County); and if Y3 was the largest, it belonged to the Turpan Black Sheep (Toksun County).

[0062] The discrimination results are shown in Table 4. The overall back-substitution accuracy and cross-validation of the three sheep models are presented. (This model uses both back-substitution validation and leave-one-out cross-validation. Back-substitution validation re-substitutes all 80 modeling samples into the model for discrimination, achieving a 100% accuracy rate. Cross-validation removes one sample at a time, then uses the remaining samples to model and discriminate the removed sample, repeating this process 80 times, also achieving a 100% accuracy rate.) Meanwhile, the discrimination score graph (… Figure 5 The results show that the samples of the three varieties are completely separated in the discriminant space, with no overlap. These validation results fully demonstrate that the discriminant model constructed based on 10 characteristic elements has excellent fitting ability, stability, and generalization ability. Samples from the same origin cluster near their respective center points, while samples from different origins show no overlap and are widely spaced. The varieties are clearly distinguishable, indicating that the 10 selected mineral element indicators effectively establish the discriminant model.

[0063] Table 4. Correct discrimination rate of mineral element indicators for different breeds of mutton

[0064] ;

[0065] LDA analysis, based on Wilks' Lambda criterion (P<0.05), progressively screened ten elements—B, Na, Mg, V, Zn, As, Rb, Sr, Mo, and Ag—to establish discriminant functions. The model achieved 100% overall accuracy in distinguishing between the three sheep breeds, demonstrating its stability and reliability. This indicates that, based on the mineral element fingerprint characteristics of the longissimus dorsi muscle, combined with multivariate statistical analysis, it is possible to accurately distinguish between Yemule white sheep, Kazakh sheep, and Turpan black sheep from different geographical origins, providing effective technical support for the construction of a traceability system for Xinjiang's distinctive livestock products.

[0066] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A method for constructing a stepwise linear discriminant model for mutton traceability, characterized in that, (1) The longissimus dorsi muscles of Yemule white sheep, Kazakh sheep and Turpan black sheep in northern Xinjiang were collected as mutton samples. (2) The contents of 28 mineral elements in the sample were determined by ICP-MS. (3) Screen for mineral elements with significant differences between groups and perform principal component analysis and partial least squares discriminant analysis; (4) Select feature elements to construct a stepwise linear discriminant model.

2. The construction method according to claim 1, wherein the mutton sample is the longissimus dorsi muscle of a one-year-old sheep that is pasture-raised and not artificially fed, and preferably, the sample is flash-frozen in liquid nitrogen and then stored at -20°C.

3. The tracing method according to claim 1, characterized in that, The 28 minerals are B, Na, Mg, Al, P, K, V, Mn, Cu, Zn, As, Rb, Sr, Mo, Ag, Sb, Cs, Ba, La, Ce, Hf, Pt, Ca, Fe, Co, Sn, Nd, and Ir.

4. The tracing method according to claim 1, characterized in that, The significantly different mineral elements in step (3) are B, Na, Mg, Al, P, K, V, Mn, Cu, Zn, As, Rb, Sr, Mo, Ag, Sb, Cs, Ba, La, Ce, Hf, and Pt. Preferably, the high contribution elements with VIP>1 selected by the partial least squares discriminant analysis are Ag, Na, B, Al, Zn, and As.

5. The tracing method according to claim 1, characterized in that, The stepwise linear discriminant model is constructed based on Wilks' Lambda criterion, P < 0.

05.

6. The construction method according to claim 1, characterized in that, It also includes steps for back-substitution validation and cross-validation of the stepwise linear discriminant model.

7. A method for tracing the origin of mutton based on mineral fingerprint characteristics, characterized in that, The content of characteristic elements in the sheep sample to be tested is substituted into the stepwise linear discriminant model obtained by the construction method described in any one of claims 1-6 to achieve the identification of mutton breed and place of origin; Specifically, the stepwise linear discriminant model is as follows: (1)Y1=-173.107+9.065X1+0.148X2+0.712X3+15.192X4+0.663X5+22.387X6+12.385X7-6.96X8-73.183X9+31.986X 10 ; (2)Y2=-165.077-2.189X1+0.095X2+0.94X3-158.04X4+0.773X5+38.831X6+2.415X7+33.509X8+292.068X9-33.372X 10 ; (3) Y3 = -134.546 + 2.357X1 + 0.095X2 + 0.816X3 + 62.503X4 + 0.411X5 - 545.71X6 - 3.467X7 + 48.268X8 + 384.474X9 - 31.441X 10 ; In the stepwise linear discriminant model, X1 to X 10 For feature elements ， The contents of B, Na, Mg, V, Zn, As, Rb, Sr, Mo, and Ag are listed in that order.

8. The mutton traceability method according to claim 7, characterized in that, The content of the characteristic elements was determined by ICP-MS.

9. The mutton traceability method according to claim 8, characterized in that, By comparing the three values ​​obtained from the stepwise linear discriminant model, the breed corresponding to the maximum value is the classification of the sample to be tested: if Y1 is the largest, it belongs to Yemule white sheep (Emin County); if Y2 is the largest, it belongs to Kazakh sheep (Tekes County); if Y3 is the largest, it belongs to Turpan black sheep (Toksun County), thus realizing the identification of sheep breed and place of origin.

10. The application of the traceability method according to any one of claims 7-9 in the identification and origin traceability of distinctive mutton breeds in northern Xinjiang.

Images