Intermediate infrared rapid batch detection method for arginine content in milk
Through mid-infrared spectroscopy technology and second-order derivative partial least squares regression model, combined with the selection of characteristic bands, the problem of low accuracy and sensitivity of arginine detection in milk is solved, and fast and accurate detection is achieved, suitable for China Cow herd.
Patent Information
- Application Number
- CN202411022810.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-29
- Publication Date
- 2024-11-12
AI Technical Summary
The prior art cannot effectively and quickly detect the arginine content in milk, especially in Chinese dairy cows. The accuracy and sensitivity of mid-infrared spectral detection are low, making it difficult to achieve fast, efficient and accurate detection.
Using mid-infrared spectroscopy technology, MIR data with characteristic bands of 925.92cm-1-2997.666cm-1 were selected, and predictions were made through second-order derivatives and partial least squares regression models, combining adjacency constraint hierarchical clustering and manual selection of feature bands. The complexity of the model and multicollinearity problems are reduced, and the detection accuracy is improved.
It realizes rapid, batch and accurate detection of the arginine content in milk. The detection time takes only 10-15 seconds, improving the detection efficiency and is suitable for dairy cow production performance measurement and quality testing in the dairy industry.
Smart Images

Figure BDA0004967776470000031 
Figure BDA0004967776470000061 
Figure BDA0004967776470000062
Abstract
Description
Technical Field
[0001] The invention belongs to the field of dairy cow performance measurement and milk quality detection, and particularly relates to a mid-infrared rapid batch detection method for arginine content in milk. Background Art
[0002] Amino acids can be divided into two types according to their form of existence. One is hydrolyzed amino acids, which exist in the form of peptide chains in proteins or peptide molecules; the other is free amino acids, which exist in a separate, free form. According to their nutritional properties, amino acids can be divided into essential amino acids, semi-essential amino acids and non-essential amino acids.
[0003] The molecular formula of arginine (Arg) is C6H 14 N4O2 exists in both hydrolyzed milk and free amino acids. It is a semi-essential amino acid. Although the human body can synthesize it, it usually cannot meet normal needs, especially in infancy and in human disease states. It has a wide range of uses and plays an important role in protein synthesis, physiological regulation, cell repair and regeneration, and maintenance of immune function. Studies have shown that arginine can stimulate fibroblasts to synthesize collagen and promote matrix deposition, angiogenesis and remodeling, thereby promoting wound healing. In addition, arginine is also the main component of sperm protein and plays an important role in the development and function maintenance of the male reproductive system.
[0004] At present, the arginine content in milk is detected by amino acid analyzer (ninhydrin post-column derivatization ion exchange chromatograph). In addition, there are pre-column derivatization reversed-phase high-performance liquid chromatography and so on. However, these methods require a lot of pre-treatment work, which is time-consuming, complicated to operate, and difficult to detect in large quantities. Mid-infrared spectroscopy MIRS (Mid-Infrared Spectroscopy) technology is based on the study of the interaction between matter and electromagnetic waves. In the mid-infrared region, when a substance is passed through by electromagnetic radiation, the bonds of the molecules will move (such as vibration and rotation), significantly absorbing the energy provided, and the chemical composition and related compounds of the sample can be determined based on the absorption characteristics formed by the substance. The use of mid-infrared spectroscopy technology can achieve rapid and non-destructive batch detection of sample substances for qualitative or quantitative purposes, such as the detection of cow milk and dairy product composition and dairy product adulteration, but there is currently no high-precision mid-infrared spectroscopy rapid batch detection model for milk amino acids.
[0005] At present, the research on arginine content in milk amino acid based on mid-infrared detection has been started, but it is mostly in free amino acids or colostrum, and there are problems such as low accuracy and sensitivity. The prediction model established based on the current herd is also not suitable for Chinese herds, because the accuracy of the mid-infrared spectrum detection result can be affected by various factors. Therefore, it is urgent to establish a detection method for the arginine content of milk suitable for Chinese dairy cows, so as to achieve fast, efficient, accurate and non-invasive determination of arginine content in milk milk, an important performance index, promote milk milk production performance determination and genetic breeding improvement, and provide reference for dairy processing industry and consumers, and provide technical guarantee for the development of China's dairy industry. The purpose of the present invention is to solve the current problems, establish a rapid detection technology for arginine content in milk amino acids, and provide technical guarantee for the development of China's dairy industry. Summary of the invention
[0006] The purpose of the present invention is to provide a mid-infrared rapid batch detection method for arginine content in milk. The method is simple, rapid and has a high accuracy rate compared with the true value.
[0007] Another object of the present invention is to provide an application of a mid-infrared rapid batch detection method for arginine content in milk. In order to achieve the above object, the present invention adopts the following technical measures:
[0008] The mid-infrared rapid batch detection method for arginine content in milk comprises the following steps:
[0009] 1. The characteristic band of the infrared spectrum of the collected milk sample is: 925.92cm -1 -979.932cm -1 、1003.08cm -1 -1010.796cm -1 、1026.228cm -1 、1045.518cm -1 -1064.808cm -1 、1103.388cm -1 -1130.394cm -1 、1215.27cm -1 -1230.702cm -1 、1253.85-1257.708cm -1 、1319.436cm -1 、1358.016-1381.164cm -1 、1415.886cm -1 、1458.324-1462.182cm -1 、1473.756cm -1、1477.614-1481.472cm -1 、1493.046cm -1 、1581.78-1585.638cm -1 、1728.384cm -1 、1759.248cm -1 、1770.822cm -1 、1793.97cm -1 、1817.118-1944.432cm -1 1971.438-1998.444cm -1 、2349.522cm -1 、2816.34-2827.914cm -1 、2862.636cm -1 、2966.802cm -1 、2970.66cm -1 、2982.234cm -1 and 2997.666cm -1 MIR data in;
[0010] 2. Substitute the MIR data obtained by the measurement into the second-order derivative + partial least squares regression model to output the prediction result of arginine content;
[0011] The principal component of the partial least squares regression is 10.
[0012] The protection scope of the present invention also includes:
[0013] Application of mid-infrared rapid batch detection method for arginine content in milk in milk quality testing;
[0014] In the above application, preferably, the milk quality detection is arginine content detection.
[0015] The arginine content involved in the present invention refers to the total content of free arginine and hydrolyzed arginine in milk.
[0016] Compared with the prior art, the present invention has the following advantages:
[0017] 1. In terms of the selection of characteristic bands, we broke the commonly used algorithm to screen features, and instead used the adjacency constrained hierarchical clustering (HAC) + manual selection method. The characteristic waves finally selected not only improved the prediction performance of the model, but also overcame the multicollinearity problem of the spectrum, reduced the complexity of the model (the input variables were reduced from the original 1060 wave points to 113 wave points), and ultimately saved the time cost, storage space cost and electricity cost of the model operation.
[0018] 2. The preprocessing and algorithm combination corresponding to the optimal model for predicting the arginine content in milk was screened out, the optimal parameters were determined, and the accuracy of the model was improved.
[0019] 3. It realizes the rapid, batch, accurate and low-cost detection of arginine content in raw milk. The measurement time of each sample is only 10-15 seconds, which improves the detection efficiency and has strong practicality. It lays a foundation for the determination of dairy cow production performance (DHI) and the expansion of milk quality detection indicators in the dairy industry, the screening of high-quality milk sources for the dairy processing industry, and the breeding of cows rich in arginine milk required by consumers by breeders. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] Figure 1 (a) The infrared spectrum of the untreated milk sample and the average (± standard deviation) spectrum (b) are shown.
[0021] Figure 2 This is the mid-infrared spectrum of the milk sample after second-order derivative pretreatment.
[0022] Figure 3 This is the correlation between the true value and predicted value of the model in milk data and the fitted straight line graph. DETAILED DESCRIPTION
[0023] The technical solutions described in the present invention, unless otherwise specified, are all conventional solutions in the art; the reagents or materials described, unless otherwise specified, are all from commercial channels.
[0024] 1. Experimental Materials
[0025] The experimental materials came from 317 Chinese Holstein cows from 7 dairy farms in 6 regions of my country. A milk sample was collected from each cow. The milk sample collection was completed using an automatic milking device. First, the milk room was wiped with a sterilized towel, and then the udder was disinfected with an iodine-glycerin mixed solution. After squeezing out the first three handfuls of milk, milk samples were collected from the entire milking process. 40 ml of each milk sample was collected and packaged into new cylindrical sampling bottles with a diameter of 3.5 cm and a height of 9 cm. They were numbered in sequence, and bronopol preservative was immediately added to each sampling bottle and shaken slowly to fully dissolve it. Ice bags (2-4°C) were placed around the milk samples on the way back to prevent deterioration. Spectra were collected immediately after the samples arrived at the laboratory.
[0026] Sample information statistics table
[0027]
[0028] 2. Mid-infrared spectrum measurement and collection
[0029] The sample was poured into a cylindrical sample tube with a diameter of 3.5 cm and a height of 9 cm, and then placed in a water bath at 42°C for 15-20 min. The sample was scanned using the MilkoScan TM 7RM milk composition detector, the solid fiber probe is extended into the liquid, the sample is mixed and then scanned. The specific collection steps are: put the milk samples in batches into a 45℃ electric constant temperature water bath for 5 minutes, put the preheated milk samples on the detection rack and shake them up and down several times to mix the milk colloidal solution evenly, put the detection rack on the detection track, open the bottle cap, and test in turn. After the spectrum is collected, the milk samples are stored at 4℃ for the subsequent determination of the amino acid content in milk.
[0030] 3. Detection method of the true (reference) value of arginine content in milk
[0031] 3.1 Instruments, equipment and reagents
[0032] Fully automatic amino acid analyzer (German Sykam S433D); syringe filter; 0.22μm nylon filter membrane; electric heating blast drying oven; nitrogen blower; electronic analytical balance; vortex oscillator; centrifuge; injection bottle; hydrolysis tube; parallel evaporator; volumetric flask.
[0033] Hydrochloric acid (HCl): concentration ≥36%, high purity; phenol (C6H5OH); nitrogen: purity 99.9%; sodium citrate (Na3C6H5O7·2H2O); sodium hydroxide (NaOH): high purity; mixed amino acid standard solution; amino acid mixed standard solution; ninhydrin solution; buffer A, buffer B, buffer C, regeneration solution D; LiOH×H2O.
[0034] 3.2 Experimental methods
[0035] (1) Preparation of standard solution
[0036] Mixed amino acid standard stock solution (1 μmol / mL): Accurately weigh individual amino acid standards (accurate to 0.00001 g) in the same 50 mL beaker, dissolve with 8.3 mL 6 mol / L hydrochloric acid solution, accurately transfer to a 250 mL volumetric flask, dilute to volume with water, and mix well.
[0037] Mixed amino acid standard working solution (100nmol / mL): Accurately pipette 1.0mL of mixed amino acid standard stock solution into a 10mL volumetric flask, add pH 2.2 sodium citrate buffer solution to the scale, mix well, and use it as the standard liquid for the machine.
[0038] (2) Milk sample processing
[0039] Pipette about 1ml of milk sample into the hydrolysis tube, add 10mL-15mL of 6mol / L hydrochloric acid solution and 3-4 drops of phenol into the hydrolysis tube. Put the hydrolysis tube into the refrigerant, freeze for 3min-5min, connect it to the vacuum pump's exhaust pipe, evacuate (close to 0Pa), and then fill it with nitrogen. Repeat the evacuation-filling with nitrogen 3 times, and seal or tighten the screw cap under the nitrogen filling state. Put the sealed hydrolysis tube in an electric blast thermostat or hydrolysis furnace at 110℃±1℃, hydrolyze for 22h, take it out, and cool it to room temperature. Open the hydrolysis tube, filter the hydrolyzate into a 50mL volumetric flask, rinse the hydrolysis tube with a small amount of water several times, transfer the water washing solution into the same 50mL volumetric flask, and finally dilute to the scale with water, and shake to mix. Accurately pipette 1.0mL of filtrate into a 15mL or 25mL test tube, and use a test tube concentrator or parallel evaporator to reduce pressure and dry under a heating environment of 40℃-50℃. Dissolve the residue after drying with 1mL-2mL of water, and then reduce pressure and dry it, and finally evaporate it to dryness. Add 1.0mL-2.0mL of pH2.2 sodium citrate buffer solution to the dried test tube to dissolve it, shake and mix it, pipette the solution through a 0.22μm filter membrane, and transfer it to the instrument injection bottle as the sample test solution for instrument measurement.
[0040] (3) Chromatographic conditions
[0041] Chromatographic column: sulfonic acid cationic resin; detection wavelength: 570nm and 440nm.
[0042] 4. Elimination of outliers and selection of valid samples
[0043] The outliers are removed according to the following criteria:
[0044] (1) Remove samples with reference values and spectra with null values
[0045] (2) Remove samples whose reference values are outside the range of mean ± 3 times standard deviation.
[0046] Finally, 4 abnormal samples were eliminated from 317 samples, and 313 samples were selected for model building and optimization.
[0047] Figure 1 The infrared spectra of 317 untreated milk samples (a) and the average (± standard deviation) spectra (b).
[0048] Embodiment 1:
[0049] Selection of the prediction model algorithm for arginine in milk:
[0050] The purpose of this application is to establish a quantitative prediction model for the arginine content in milk, using three common basic modeling algorithms.
[0051] Partial least squares regression (PLSR) is a multivariate data analysis method in statistics and machine learning, and is particularly suitable for dealing with situations where there is multicollinearity between dependent and independent variables. PLSR is the most basic and common machine learning algorithm. Its core idea is to find new orthogonal projection directions (principal components) so that the projected dependent and independent variables have the largest covariance, and then establish a prediction model. Its advantages mainly include the following three aspects: (1) It effectively overcomes the multicollinearity problem between the dependent and independent variables; (2) When the dimensions of both the independent and dependent variables are very high, PLS regression can extract the most important information and construct a prediction model through dimensionality reduction technology; (3) When the number of samples is small, a relatively ideal prediction effect can still be obtained. Therefore, this algorithm is selected as the candidate algorithm in this embodiment.
[0052] K-nearest neighbor algorithm (KNN) is an algorithm in supervised learning, which can be used for both classification and regression. The advantages of KNN algorithm are that it is simple and easy to use, can be used for nonlinear traits, has fast model training time and good prediction effect. It is one of the most basic and simplest algorithms in machine learning algorithms. Therefore, this embodiment selects this algorithm as the candidate algorithm.
[0053] Ridge regression is a least squares method with L2 regularization. It abandons the unbiasedness of the least squares method and obtains the regression coefficient at the cost of losing some information and reducing precision. It is a more practical and reliable regression method, and its fitting of data with outliers is stronger than the least squares method. Therefore, this embodiment selects this algorithm as the candidate algorithm.
[0054] Embodiment 2:
[0055] Establishment of a method for detecting arginine content in milk using mid-infrared spectroscopy:
[0056] 1. Division of modeling data set
[0057]
[0058] In this embodiment, the total data set is divided into two parts, namely, a training set and a test set, with a data ratio of about 4:1. The training set is used to train the model, and the test set is used to test the model performance. The training set is also called a cross-validation set, and a 10-fold cross-validation is performed during the model training process.
[0059] 2. Screening of MIR data preprocessing methods for modeling
[0060] This embodiment uses trial and error to select appropriate spectral preprocessing techniques, such as first derivative (1D), second derivative (1D), standard normal variable (SNV) and Savitzky-Galsy convolution smoothing (SG). The strategy of trial and error is to first apply each preprocessing to each model, and then determine the most successful model based on performance parameters. SNV is mainly used to eliminate the effects of particle size and surface scattered light on the spectrum. Derivative processing is generally used to improve the resolution of spectral samples and sharpen absorption bands.
[0061] The most appropriate spectral preprocessing technique was selected by trial and error. The modeling band used was the one with the water region removed (1593.35-1709.1 cm -1 and 3059.39-3641.95cm -1 ) and no information area (3,500-5,000cm -1 ) The remaining spectral region contains a total of 520 wave points. The modeling algorithm used is the most commonly used PLSR algorithm. The best preprocessing method is determined by the results: 2D, Figure 2 This is the mid-infrared spectrum of the milk sample after second-order derivative pretreatment.
[0062] The results are shown in the table:
[0063]
[0064] 3. Selection of modeling algorithms
[0065] In the screening process of the modeling algorithm, the modeling band used was the one that removed the water area (1593.35-1709.1cm -1 and 3059.39-3641.95cm -1 ) and no information area (3,500-5,000cm -1 ) The remaining spectral region contains a total of 520 wave points, and the spectral preprocessing technology used is 2D. The best modeling algorithm is determined by the prediction accuracy of the training set and the test set: PLSR, and the results are shown in the table:
[0066]
[0067] 4. Screening of modeling feature bands (adjacent constraint hierarchical clustering combined with manual selection)
[0068] The feature band method includes two methods: algorithmic feature selection and manual feature selection. The basis for algorithmic feature selection is mainly the correlation between each wave point and the reference value. Although it is fast and efficient, it ignores the synergy between adjacent wave points. The advantage of manually selecting features is that the role of the band (i.e., adjacent wave points) can be strengthened in the selection process, and the original information state of the spectrum can be retained more in the process of improving the model, with stronger inclusiveness and generalization ability. However, the speed of manual feature selection is relatively slow and the efficiency is also low. The milk spectrum is a high-dimensional data with 1060 wave points. One of the notable features is that adjacent wave points have multicollinearity, that is, they may contain the same or similar information. If these wave points are used directly for modeling, a large amount of similar information will exist in the model, thereby affecting the modeling effect. Taking the above points into consideration, the present application uses manual feature selection based on the wave points after adjacency constrained hierarchical clustering (HAC). This method not only handles the multicollinearity characteristics of the spectrum, but also more fully retains the original information state of the spectrum to improve the accuracy and effect of modeling.
[0069] In this embodiment, the feature bands are selected by combining adjacency constrained hierarchical clustering (HAC) with manual selection, and the selection steps are as follows:
[0070] (1) Remove 1593.35cm -1 -1709.1cm -1 and 3059.39cm -1 -3641.95cm -1 Two spectral regions associated with water absorption.
[0071] (2) Remove 3,500-5,000cm -1 region of the spectrum because it is believed to contain very little valuable information.
[0072] (3) Spatial clustering: The remaining 520 spectral points are subjected to HAC analysis, and the optimal number of windows is determined based on the slope heuristic method. In this embodiment, HAC is implemented using the adjclust package in R, and the slope heuristic method for determining the number of windows is implemented using the Capushe function in the adjclust package. Finally, the HAC algorithm clusters the 520 points into 64 groups, with the number of points in each group ranging from 1 to 46. The results are shown in the table:
[0073]
[0074]
[0075]
[0076] (4) Using the partial least squares regression algorithm, first increase or decrease the wave points at both ends of the critical point of the first group of bands by one wave point (increase one wave point for the lowest wave number and decrease one wave point for the highest wave number) to find the optimal effect, and then perform similar operations on the second group of bands based on this. If the model performance of the second group of bands is not improved or even reduced after similar operations, then remove this band from the model. Finally, after all 64 bands have completed a round of operations until the model accuracy does not change or improve, it is the optimal feature band.
[0077] The results of the model established by the optimal characteristic band obtained after 64 rounds of screening are shown in the table:
[0078]
[0079] After HAC combined with manual band selection, the model accuracy has been improved. The final characteristic band selected is: 925.92cm -1 -979.932cm -1 、1003.08cm -1 -1010.796cm -1 、1026.228cm -1 、1045.518cm -1 -1064.808cm -1 、1103.388cm -1 -1130.394cm -1 、1215.27cm -1 -1230.702cm -1 、1253.85-1257.708cm -1 、1319.436cm -1 、1358.016-1381.164cm -1 、1415.886cm -1 、1458.324-1462.182cm -1 、1473.756cm -1 、1477.614-1481.472cm -1 、1493.046cm -1 、1581.78-1585.638cm -1 、1728.384cm -1 、1759.248cm -1 、1770.822cm -1 、1793.97cm -1 、1817.118-1944.432cm -1 1971.438-1998.444cm-1 、2349.522cm -1 、2816.34-2827.914cm -1 、2862.636cm -1 、2966.802cm -1 、2970.66cm -1 、2982.234cm -1 and 2997.666cm -1 , containing a total of 113 dots.
[0080] 5. Screening and determination of model parameters
[0081] Model parameters include parameters of the preprocessing method and parameters of the algorithm;
[0082] Since the model uses a 2D preprocessing algorithm, it is only necessary to select the algorithm parameters. The parameters of the partial least squares regression algorithm are: principal components (n_component). The comparison of parameter selection results is as follows:
[0083]
[0084]
[0085] According to the comprehensive comparison results of the training set and the test set, the principal component (n_component) was finally selected as 10.
[0086] After comparative analysis, the best regression model for arginine content in milk is: 2D+PLSR (n_component=10) model. The correlation coefficients of the training set and the test set are 0.9058 and 0.8304 respectively; the root mean square errors of the training set and the test set are 0.0063 and 0.0070 respectively; the performance deviation ratios of the training set and the test set are 2.3641 and 1.7964 respectively. Figure 3 This is the correlation between the true value and predicted value of the model in milk data and the fitted straight line graph.
[0087] Embodiment 3:
[0088] Application of rapid batch detection method of mid-infrared spectroscopy MIR for arginine content in milk:
[0089] The established optimal regression model 2D+PLSR (n_component=10) for arginine content was used to predict 10 randomly selected milk samples (not one of the 317 experimental materials), and the predicted results were compared with the true values.
[0090] How to use the model:
[0091] 1. The characteristic band of the infrared spectrum of the collected milk sample is: 925.92cm -1 -979.932cm -1 、1003.08cm -1 -1010.796cm -1 、1026.228cm -1 、1045.518cm -1 -1064.808cm -1 、1103.388cm -1 -1130.394cm -1 、1215.27cm -1 -1230.702cm -1 、1253.85-1257.708cm -1 、1319.436cm -1 、1358.016-1381.164cm -1 、1415.886cm -1 、1458.324-1462.182cm -1 、1473.756cm -1 、1477.614-1481.472cm -1 、1493.046cm -1 、1581.78-1585.638cm -1 、1728.384cm -1 、1759.248cm -1 、1770.822cm -1 、1793.97cm -1 、1817.118-1944.432cm -1 1971.438-1998.444cm -1 、2349.522cm -1 、2816.34-2827.914cm -1 、2862.636cm -1 、2966.802cm -1 、2970.66cm -1 、2982.234cm -1 and 2997.666cm -1 MIR data in;
[0092] At the same time, a fully automatic amino acid analyzer is used to detect the true value of the arginine content in the same batch of milk.
[0093] 2. Substitute the MIR data obtained by the measurement into the 2D+PLSR (n_component=10) model constructed in Example 2 to output the prediction result of the arginine content.
[0094] The results predicted by the model are close to the actual results (as shown in the following table), so the model has high accuracy and can be used to predict the arginine content in milk.
[0095]
[0096]
Claims
1. A mid-infrared rapid batch detection method for arginine content in milk, comprising the following steps:
1. The characteristic band in the infrared spectrum of the collected milk sample is: 925.92 cm -1 -979.932 cm -1 、1003.08cm -1 -1010.796cm -1 、1026.228 cm -1 、1045.518 cm -1 -1064.808cm -1 、1103.388 cm -1 -1130.394cm -1 、1215.27 cm -1 -1230.702cm -1 、1253.85-1257.708 cm -1 、1319.436cm -1 、1358.016-1381.164 cm -1 、1415.886cm -1 、1458.324-1462.182 cm -1 、1473.756cm -1 、1477.614-1481.472cm -1 、1493.046cm -1 、1581.78-1585.638cm -1 、1728.384cm -1 、1759.248cm -1 、1770.822cm -1 、1793.97cm -1 、1817.118-1944.432cm -1 1971.438-1998.444cm -1 、2349.522cm -1 、2816.34-2827.914cm -1 、2862.636cm -1 、2966.802cm -1 、2970.66cm -1 、2982.234cm -1 and 2997.666cm -1 MIR data in; 2. Substitute the MIR data obtained by the measurement into the second-order derivative + partial least squares regression model to output the prediction result of arginine content; The principal component of the partial least squares regression is 10.
2. Application of the detection method according to claim 1 in milk quality detection.
3. The use according to claim 2, wherein the milk quality detection is arginine content detection.
