Prediction model of whey protein content and prediction method thereof

Abstract

The invention relates to a prediction model of whey protein content and a prediction method thereof. The prediction model comprises a beta-lactoglobulin content prediction model and an alpha-lactalbumin content prediction model. The beta-lactoglobulin content prediction model is as follows: y1 = b1-k1 * W1; the alpha-lactalbumin content prediction model is as follows: y2 = b2-k2 * W2; wherein y1 represents the content of beta-lactoglobulin with the unit of mg / g protein, and y2 represents the content of alpha-lactalbumin with the unit of mg / g protein; w1 and W2 represent whiteness values detected by a spectrophotometer; b1 is a constant with the value ranging from 600 to 1400, and k1 is a constant with the value ranging from 8 to 14; b1 is a constant with a value of 200-350, b2 is a constant with a value of 200-350, and k2 is a constant with a value of 2.5-4. The prediction model provided by the invention not only reduces the cost of whey protein content detection, simplifies the detection process, avoids the use of toxic reagents, but also can ensure higher prediction accuracy, and can be widely applied to quality identification and application characteristic discrimination of milk and dairy products.

Description

Technical Field

[0001] The invention relates to the technical field of milk or dairy product analysis and detection, and in particular to a prediction model and a prediction method for whey protein content. Background Art

[0002] Whey protein is a protein extracted from milk. It is regarded as the "king of protein" for its high absorption rate and the most reasonable amino acid composition. It is not only easy to digest, but also has the characteristics of high biological value, high digestibility, high protein efficacy ratio and high utilization rate. It is recognized as one of the high-quality protein supplements for the human body. It is currently widely used in sports nutrition, weight management, elderly nutrition and other fields. Whey protein accounts for about 18-20% of milk protein and can be divided into heat-stable and heat-unstable whey proteins. When the whey liquid is boiled at pH = 4.6-4.7 for 20 minutes, a type of protein that precipitates is heat-unstable whey protein, mainly including α-lactalbumin and β-lactoglobulin.

[0003] α-Lactalbumin and β-lactoglobulin are important components of whey protein, with high nutritional value and functional properties. Studies have shown that α-lactalbumin is an excellent source of essential amino acids and branched-chain amino acids, and has the function of regulating protein balance and promoting the development of the immune system; α-lactalbumin combined with zinc ions can activate immune active factors and reduce diarrhea; in addition, tryptophan, which is rich in α-lactalbumin, is an important factor in neurodevelopment and can significantly improve the sleep quality of infants and young children. β-lactoglobulin can bind to fat-soluble vitamins to promote their absorption; β-lactoglobulin has functional properties such as anti-hypertension, anti-virus, and helps calcium and mineral absorption, and has been increasingly used in dairy products and infant formula foods. Because active α-lactalbumin and β-lactoglobulin have thermal gel properties, they also have the function of strengthening food texture. The content of active α-lactalbumin and β-lactoglobulin is positively correlated with the degree of texture strengthening.

[0004] Therefore, the content of α-lactalbumin and β-lactoglobulin is an important determinant of the quality of whey protein powder raw materials and their application characteristics. However, the detection method for active α-lactalbumin and β-lactoglobulin in China has not yet been officially released, and the agricultural industry standard "High Performance Liquid Chromatography for the Determination of α-lactalbumin and β-lactoglobulin in Milk and Dairy Products" is still in the compilation stage. This method requires the use of large-scale precision instruments and toxic and harmful reagents. Although it has high accuracy and precision, it is cumbersome, time-consuming and costly to operate. Summary of the invention

[0005] In order to solve the above technical problems, the present invention provides a prediction model and prediction method for whey protein content. The prediction model for the content of α-lactalbumin and β-lactoglobulin based on the whiteness value of the sample provided by the present invention not only reduces the cost of whey protein content detection, simplifies the detection process, avoids the use of toxic reagents, and improves the detection efficiency, but also ensures a high prediction accuracy, and can be widely used in the quality identification and application characteristic discrimination of milk and dairy products.

[0006] To achieve this object, the present invention adopts the following technical solutions:

[0007] In a first aspect, the present invention provides a prediction model for whey protein content, the prediction model comprising a prediction model for β-lactoglobulin content and a prediction model for α-lactalbumin content;

[0008] The β-lactoglobulin content prediction model is:

[0009] y1=b1-k1×W1;

[0010] The α-lactalbumin content prediction model is:

[0011] y2=b2-k2×W2;

[0012] Wherein, y1 represents the β-lactoglobulin content, and the unit is mg / g protein; y2 represents the α-lactalbumin content, and the unit is mg / g protein; b1 is a constant, and the value is 600-1400 (for example, it can be 600, 700, 800, 900, 1000, 1100, 1200, 1300, 1400, etc.), k1 is a constant, and the value is 8-14 (for example, it can be 8, 9, 10, 11, 12, 13, 14, etc.); b2 is a constant, and the value is 200-350 (for example, it can be 200, 220, 240, 260, 280, 300, 320, 340, 350, etc.), and k2 is a constant, and the value is 2.5-4 (for example, it can be 2.5, 2.8, 3, 3.2, 3.5, 3.8, 4, etc.).

[0013] α-lactalbumin and β-lactoglobulin are important components of whey protein, and their content is an important determinant of the quality of whey protein powder raw materials and their application characteristics. However, the detection method for active α-lactalbumin and β-lactoglobulin in China has not yet been officially released, and the agricultural industry standard "High Performance Liquid Chromatography for the Determination of α-lactalbumin and β-lactoglobulin in Milk and Dairy Products" is still in the compilation stage. This method requires the use of large-scale precision instruments and toxic and harmful reagents. Although it has high accuracy and precision, it is cumbersome, time-consuming and costly to operate.

[0014] Based on this, the present invention creatively associates the whiteness value of the sample with the content of α-lactalbumin and β-lactoglobulin contained therein, and converts the content value obtained by liquid chromatography detection, which requires complex operations and high costs, into the whiteness value obtained by spectrophotometer detection, which is simple, easy to obtain and time-saving, so as to establish a prediction model for obtaining the content of α-lactalbumin and β-lactoglobulin based on the whiteness value of the sample. The construction of the above model not only reduces the cost of whey protein content detection, simplifies the detection process, avoids the use of toxic reagents, and improves the detection efficiency, but also ensures a high prediction accuracy, and can be widely used in the quality identification and application characteristics of milk and dairy products.

[0015] The whiteness standards disclosed in the prior art and the formulas defined therein vary and are mostly used in the fields of papermaking, textiles, building materials, etc. Moreover, each whiteness formula has certain limitations and cannot meet the actual measurement needs of various industries and types of samples.

[0016] The present invention creatively associates the whiteness characteristics of the sample with the whey protein content, and introduces a specific whiteness standard through correlation analysis to address the gap in whiteness standards in the dairy industry, thereby achieving a low-cost, high-efficiency, high-accuracy, and good repeatability whey protein content prediction effect, and has high practical application value.

[0017] Preferably, when the detection method of W1 is based on the Hunter whiteness formula, b1 takes a value of 1000-1300 (for example, it can be 1000, 1050, 1100, 1150, 1200, 1250, 1300, etc.), and k1 takes a value of 11-14 (for example, it can be 11, 11.5, 12, 12.5, 13, 13.5, 14, etc.).

[0018] Preferably, when the detection method of W1 is based on the blue light whiteness formula, b1 takes a value of 880-920 (for example, it can be 880, 890, 900, 910, 920, etc.), and k1 takes a value of 12-13 (for example, it can be 12, 12.2, 12.4, 12.6, 12.8, 13, etc.).

[0019] Preferably, when the detection method of W1 is based on GBT5950-96 standard, b1 takes a value of 600-800 (for example, 600, 650, 700, 750, 800, etc.), and k1 takes a value of 9-12 (for example, 9, 10, 11, 12, etc.).

[0020] Preferably, when the detection method of W1 is based on the QBT1503-92 standard, b1 takes a value of 800-900 (for example, it can be 800, 820, 840, 860, 880, 900, etc.), and k1 takes a value of 12-14 (for example, it can be 12, 12.2, 12.5, 12.8, 13, 13.2, 13.5, 13.8, 14, etc.).

[0021] Preferably, when the detection method of W2 is based on the Hunter whiteness formula, b2 takes a value of 300-350 (for example, it can be 300, 310, 320, 330, 340, 350, etc.), and k2 takes a value of 3.2-3.8 (for example, it can be 3.2, 3.3, 3.4, 3.5, 3.6, 3.7, 3.8, etc.).

[0022] Preferably, when the detection method of W2 is based on the blue light whiteness formula, b2 takes a value of 240-280 (for example, it can be 240, 250, 260, 270, 280, etc.), and k2 takes a value of 3-4 (for example, it can be 3, 3.2, 3.4, 3.6, 3.8, 4, etc.).

[0023] Preferably, when the detection method of W2 is based on GBT5950-96 standard, b2 takes a value of 220-260 (for example, it can be 220, 230, 240, 250, 260, etc.), and k2 takes a value of 2.8-3.4 (for example, it can be 2.8, 2.9, 3, 3.1, 3.2, 3.3, 3.4, etc.).

[0024] Preferably, when the detection method of W2 is based on the QBT1503-92 standard, b2 takes a value of 200-290 (for example, it can be 200, 220, 240, 260, 280, 290, etc.), and k2 takes a value of 2.5-4 (for example, it can be 2.5, 2.8, 3, 3.2, 3.5, 3.8, 4, etc.).

[0025] In a second aspect, the present invention provides a method for constructing a prediction model for whey protein content according to the first aspect, the construction method comprising the following steps:

[0026] (1) Collection of whey protein powder raw materials and information collation

[0027] Collect whey protein raw materials of different manufacturers and models from multiple raw material suppliers and organize the information;

[0028] (2) Testing of physical and chemical indicators of whey protein raw materials

[0029] ① Protein content: Determined in accordance with the "National Food Safety Standard - Determination of Protein in Food" (First Method: Kjeldahl Nitrogen Determination);

[0030] ②α-lactalbumin and β-lactoglobulin content: detected by high performance liquid chromatography-tandem mass spectrometry, using a Phenomenex Jupiter® C18 300A column, formic acid acetonitrile-water as the mobile phase, a flow rate of 0.3 mL / min, and a column temperature of 35°C; mass spectrometry detection method: multiple ion reaction monitoring; electrospray ion source positive ion scanning mode; curtain gas pressure of 35 psi (241.29 kPa), spray voltage of 5500 V, and desolvation temperature of 550°C;

[0031] (3) Sample solution preparation

[0032] a. Prepare a 3% protein solution, stir at 45-55℃ for 30-60 min, and hydrate for 1-2 h;

[0033] b. Homogenization: primary pressure 150-200 bar, secondary pressure 25-50 bar, stand for 20-40 min;

[0034] (4) Whiteness detection

[0035] The whiteness of the samples was recorded using a KONICA MINOLTA CM-3600d spectrophotometer:

[0036] a. Install the target mask. You can choose CM-A107 / 106 / 105, which are used for measuring aperture SAVφ4 mm / MANφ8 mm / LAVφ25.4 mm and lighting aperture φ7 mm / φ11 mm / φ30 mm respectively;

[0037] b. Install the zero calibration box and perform zero calibration;

[0038] c. Place the white calibration plate for calibration;

[0039] d. Stir the sample evenly, use a 15 mL syringe to take 10-14 mL of the sample and transfer it into a 40×40×10 mm quartz cuvette, and cover the cuvette with a lid;

[0040] e. Use the sample holder to fix the cuvette containing the sample;

[0041] f. Use D65 light source and 10° observer angle to measure the reflectance color mode, and record the whiteness of the sample according to the whiteness formula in Table 1;

[0042] Table 1

[0043]

[0044] (5) Data statistical analysis and modeling

[0045] Use the Excel plug-in XLSTAT 2019 to perform statistical analysis on the data, including correlation analysis and regression analysis;

[0046] (6) Model validation

[0047] The whiteness data of the non-modeled whey protein samples were brought into the model to predict the active whey protein in the samples, and a correlation analysis was performed between the measured content and the predicted content. If the correlation coefficient was ≥0.90 within the 95% confidence interval, it indicated that the model was accurate and reliable.

[0048] In a third aspect, the present invention provides a method for predicting whey protein content, the prediction method comprising:

[0049] The whiteness value of the sample is detected by a spectrophotometer and substituted into the prediction model described in the first aspect to obtain the α-lactalbumin content and / or β-lactoglobulin content of the sample.

[0050] Preferably, the whiteness value detection process comprises: preparing a sample solution, placing the solution in a cuvette, and detecting the solution using a spectrophotometer in a reflective color mode.

[0051] Preferably, the sample solution is homogenized and allowed to stand before detection.

[0052] Preferably, the primary pressure of the homogenization is 150-200 bar (for example, it can be 150 bar, 160 bar, 170 bar, 180 bar, 190 bar, 200 bar, etc.), and the secondary pressure of the homogenization is 25-50 bar (for example, it can be 25 bar, 30 bar, 35 bar, 40 bar, 45 bar, 50 bar, etc.).

[0053] Preferably, the standing time is 20-40 min (for example, 20 min, 25 min, 30 min, 35 min, 40 min, etc.).

[0054] Preferably, the detection is performed under D65 illuminant and 10° observer angle.

[0055] Compared with the prior art, the present invention has at least the following beneficial effects:

[0056] The present invention creatively associates the whiteness value of milk and dairy products with the content of α-lactalbumin and β-lactoglobulin therein, and converts the content value obtained by liquid chromatography detection, which requires complex operation and high cost, into the whiteness value obtained by spectrophotometer detection, which is simple, easy to obtain and time-saving, so as to establish a prediction model for obtaining the content of α-lactalbumin and β-lactoglobulin based on the whiteness value of the sample. The prediction of whey protein content provided by the present invention not only reduces the cost of whey protein content detection, simplifies the detection process, avoids the use of toxic reagents, and improves the detection efficiency, but also ensures high prediction accuracy and good repeatability, and can be widely used in milk and dairy product quality identification and application characteristic discrimination. DETAILED DESCRIPTION

[0057] For the convenience of understanding the present invention, the present invention lists the following embodiments. It should be understood by those skilled in the art that the embodiments are only to help understand the present invention and should not be regarded as specific limitations of the present invention.

[0058] Example

[0059] This embodiment provides a prediction model for the content of α-lactalbumin and β-lactoglobulin in whey protein and a method for constructing the same, comprising:

[0060] (1) Collection of whey protein powder raw materials and information collation

[0061] We collected 18 whey protein raw materials from multiple raw material suppliers with different manufacturers, models and functional claims, and sorted out the information. The summary is shown in Table 2.

[0062] Table 2

[0063]

[0064] (2) Testing of physical and chemical indicators of whey protein raw materials

[0065] ① Protein content: Determined in accordance with the "National Food Safety Standard - Determination of Protein in Food" (First Method: Kjeldahl Nitrogen Determination);

[0066] ②α-lactalbumin and β-lactoglobulin content: detected by high performance liquid chromatography-tandem mass spectrometry, using a Phenomenex Jupiter® C18 300A column, formic acid acetonitrile-water as the mobile phase, a flow rate of 0.3 mL / min, and a column temperature of 35°C; mass spectrometry detection method: multiple ion reaction monitoring; electrospray ion source positive ion scanning mode; curtain gas pressure of 35 psi (241.29 kPa), spray voltage of 5500 V, and desolvation temperature of 550°C.

[0067] The physical and chemical indicators of whey protein raw materials are summarized in Table 3.

[0068] Table 3

[0069]

[0070] (3) Sample solution preparation

[0071] Two parallel solutions were prepared for the above samples to reduce experimental errors;

[0072] a. Prepare a 3% protein solution, stir at 50℃ for 45 min, and hydrate for 1.5 h;

[0073] b. Homogenization: primary pressure 160 bar, secondary pressure 30 bar, stand for 30 min;

[0074] (4) Whiteness detection

[0075] The whiteness of the samples was recorded using a KONICA MINOLTA CM-3600d spectrophotometer, and each parallel solution was measured twice to reduce measurement errors;

[0076] a. Install the target mask, select CM-A106, measuring aperture MANφ8 mm, lighting aperture φ11 mm;

[0077] b. Install the zero calibration box and perform zero calibration;

[0078] c. Place the white calibration plate for calibration;

[0079] d. Stir the sample evenly, use a 15 mL syringe to take 12 mL of the sample solution and transfer it into a 40×40×10 mm quartz cuvette, and cover the cuvette with a lid;

[0080] e. Use the sample holder to fix the cuvette containing the sample;

[0081] f. Use D65 light source and 10° observer angle to measure the reflectance color mode and record the average whiteness of the samples (Table 4).

[0082] Table 4

[0083]

[0084] (5) Data statistical analysis

[0085] Use the Excel plug-in XLSTAT 2019 to perform statistical analysis on all data, including correlation analysis and regression analysis:

[0086] a. Correlation analysis

[0087] Correlation analysis was performed based on the α-lactalbumin content (mg / g protein), β-lactoglobulin content (mg / g protein) in Table 3 and all whiteness data in Table 4. The results showed that within the 95% confidence interval, the whiteness of the whey protein powder solution was extremely significantly negatively correlated with the α-lactalbumin and β-lactoglobulin contents of the raw materials.

[0088] The correlation matrix (Pearson) between whiteness of whey protein powder raw material solution and active whey protein is shown in Table 5.

[0089] Table 5

[0090]

[0091] Note: * indicates that there is a significant difference in the data, that is, p-values ​​< 0.05.

[0092] The results of the significance test are shown in Table 6.

[0093] Table 6

[0094]

[0095] b. Regression analysis

[0096] Based on the β-lactoglobulin content (mg / g protein) in Table 3 and all whiteness data in Table 4, a prediction model for β-lactoglobulin content based on whiteness index was established, and the regression equation was obtained as shown in Table 7;

[0097] Determination coefficient R of the embodiment 2 Greater than 0.8, the prediction performance is good, and the determination coefficient R 2 Less than 0.8, the prediction performance is poor.

[0098] Table 7

[0099]

[0100] The results show that in the prediction model of β-lactoglobulin content, the regression equation y1=896.57-12.68×W(R457) obtained by using the whiteness of W(R457) corresponding to Example 3 has a higher fitting degree, R 2 =0.99, with a standard error of 0.374.

[0101] Based on the α-lactalbumin content (mg / g protein) in Table 3 and all whiteness data in Table 4, a prediction model for α-lactalbumin content based on whiteness index was established, and the regression equation was shown in Table 8;

[0102] Determination coefficient R of the embodiment 2 Greater than 0.8, the prediction performance is good, and the determination coefficient R2 Less than 0.8, the prediction performance is poor.

[0103] Table 8

[0104]

[0105] The results show that in the α-lactalbumin content prediction model, the regression equation y2=257.04-3.19×W(R457) obtained by using the whiteness standard W(R457) corresponding to Example 2 has a higher fitting degree, R 2 =0.97, with a standard error of 0.163.

[0106] (6) Model validation

[0107] Fifteen non-modeled whey protein samples (numbered WP1-WP15) were collected to detect the active whey protein content in the samples. The W(R457) whiteness of each sample was also detected. The test data was brought into the model to predict the active whey protein. The Excel plug-in XLSTAT 2019 was used to perform linear regression analysis on the measured content and the predicted content to obtain the multiple correlation coefficient R 2 If R is within the 95% confidence interval 2 ≥0.90, it indicates that the model is accurate and reliable.

[0108] The model validation results for β-lactoglobulin and α-lactalbumin contents are shown in Tables 9 and 10, respectively.

[0109] Table 9

[0110]

[0111] The above data results show that the goodness of fit of the prediction model of the present invention can reach 0.990, indicating that the fitting accuracy of the evaluation model of the present invention is relatively high.

[0112] Table 10

[0113]

[0114] The above data results show that the goodness of fit of the prediction model of the present invention can reach 0.979, indicating that the fitting accuracy of the evaluation model of the present invention is relatively high.

[0115] The applicant declares that the above is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Those skilled in the art should understand that any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention shall fall within the protection scope and disclosure scope of the present invention.

Claims

1. A prediction model for whey protein content, characterized in that: The prediction models include a β-lactoglobulin content prediction model and an α-lactalbumin content prediction model; The β-lactoglobulin content prediction model is: y1=b1-k1×W1; The α-lactalbumin content prediction model is: y2=b2-k2×W2; Wherein, y1 represents the β-lactoglobulin content, the unit is mg / g protein, y2 represents the α-lactalbumin content, the unit is mg / g protein; W1 and W2 represent the whiteness values ​​detected by a spectrophotometer; b1 is a constant, the value is 600-1400, k1 is a constant, the value is 8-14; b2 is a constant, the value is 200-350, k2 is a constant, the value is 2.5-4.

2. The prediction model according to claim 1, characterized in that The detection methods of W1 and W2 are independently based on Hunter whiteness formula, blue light whiteness formula, GBT5950-96 standard or QBT1503-92 standard.

3. The prediction model according to claim 2, characterized in that When the detection method of W1 is based on the Hunter whiteness formula, b1 takes a value of 1000-1300, and k1 takes a value of 11-14; When the detection method of W1 is based on the blue light whiteness formula, b1 takes a value of 880-920 and k1 takes a value of 12-13; When the detection method of W1 is based on GBT5950-96 standard, b1 takes a value of 600-800, and k1 takes a value of 9-12; When the detection method of W1 is based on the QBT1503-92 standard, b1 takes a value of 800-900, and k1 takes a value of 12-14.

4. The prediction model according to claim 2, characterized in that When the detection method of W2 is based on the Hunter whiteness formula, b2 takes a value of 300-350 and k2 takes a value of 3.2-3.8; When the detection method of W2 is based on the blue light whiteness formula, b2 takes a value of 240-280, and k2 takes a value of 3-4; When the detection method of W2 is based on GBT5950-96 standard, b2 takes a value of 220-260, and k2 takes a value of 2.8-3.4; When the detection method of W2 is based on the QBT1503-92 standard, b2 takes a value of 200-290, and k2 takes a value of 2.5-4.

5. A method for predicting whey protein content, characterized in that: The prediction method comprises: The whiteness value of the sample is detected by a spectrophotometer, and the value is substituted into the prediction model described in any one of claims 1 to 4 to obtain the α-lactalbumin content and / or β-lactoglobulin content of the sample.

6. The prediction method according to claim 5, characterized in that: The whiteness value detection process includes: preparing a sample solution, placing the solution in a colorimetric dish, and detecting the solution in a reflective color mode using a spectrophotometer.

7. The prediction method according to claim 6, characterized in that: The sample solution is further homogenized and allowed to stand before detection.

8. The prediction method according to claim 7, characterized in that: The primary pressure of the homogenization is 150-200 bar, and the secondary pressure of the homogenization is 25-50 bar; The standing time is 20-40 min.

9. The prediction method according to claim 5, characterized in that: The measurements were performed under D65 illuminant and 10° observer angle.

Images