CN108680515B - Single-grain rice amylose quantitative analysis model construction and detection method thereof - Google Patents
Single-grain rice amylose quantitative analysis model construction and detection method thereof Download PDFInfo
- Publication number
- CN108680515B CN108680515B CN201810982153.9A CN201810982153A CN108680515B CN 108680515 B CN108680515 B CN 108680515B CN 201810982153 A CN201810982153 A CN 201810982153A CN 108680515 B CN108680515 B CN 108680515B
- Authority
- CN
- China
- Prior art keywords
- sample
- spectrum
- amylose
- grain rice
- rice
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/75—Systems in which material is subjected to a chemical reaction, the progress or the result of the reaction being investigated
- G01N21/77—Systems in which material is subjected to a chemical reaction, the progress or the result of the reaction being investigated by observing the effect on a chemical indicator
- G01N21/78—Systems in which material is subjected to a chemical reaction, the progress or the result of the reaction being investigated by observing the effect on a chemical indicator producing a change of colour
Abstract
The invention discloses a method for constructing a single-grain rice amylose quantitative analysis model, which comprises the following steps: s1, collecting a plurality of parts of single-grain rice samples with different amylose contents, drying, and balancing moisture to be used as a correction set; s2, collecting the near infrared spectrum of each single-grain rice sample in the correction set; s3, processing each single-grain rice sample in the correction set into rice flour respectively, obtaining the reference value of the amylose content of each single-grain rice sample, and constructing a correction set reference value matrix; s4, construction of a near-infrared single-particle amylose model: and screening a spectrum interval from the near infrared spectrum obtained in the step S2 to obtain a corrected light collection spectrum matrix, and performing regression correlation analysis on the spectrum matrix and the reference value matrix in the step S3 to obtain a near infrared quantitative analysis model of the amylose content of the single-grain rice.
Description
Technical Field
The invention relates to a method for detecting the component content of a single-grain crop, in particular to a construction method and a detection method of a single-grain rice amylose quantitative analysis model.
Background
In order to shorten the breeding cycle and accelerate the breeding process, the rice breeding requirement is that seeds with specific characters meeting the needs of breeders can be detected and even sorted out in the early generation of seed cultivation. In rice quality breeding, amylose is one of the most important quality indexes of rice, and the content of amylose determines the eating quality and the nutritional quality of rice, so that the amylose is well paid attention to breeders and consumers. The traditional amylose detection methods such as iodine chromogenic photometry and iodine affinity determination (including an ampere-titration method and a potentiometric titration method) have the defects of seed damage, reagent consumption, complex operation, time and labor waste at different degrees, and cannot meet the requirements of rapid nondestructive detection and sorting of early-generation seeds. The near infrared spectrum analysis technology is a rapid, nondestructive, simple and environment-friendly analysis technology and is widely applied to various fields of agriculture, biology, medicines, food, chemical industry and the like. The Single grain near-infrared detection technology (SK-NIRS) refers to a near-infrared spectrum analysis technology on the level of Single grain crops, is expected to realize the rapid nondestructive detection of the amylose content of Single grain rice, and can be combined with a certain automatic device to realize sorting so as to meet the requirement of breeding proper rice varieties meeting the quality requirement in early generations.
However, near infrared is a two-stage analysis technique, and a chemical method is required as a reference during analysis, and the precision of the reference method greatly influences the analysis result of the near infrared. In the existing literature reports, the chemical reference value is mainly obtained from the average chemical value of a multi-seed mixed sample. However, this method has great limitations: even if the seeds are of the same variety, the chemical value and the chemical value of the seeds with different sizes and growth conditions are different. Therefore, the chemical value obtained by the method can only be used for detecting the content of the variety components, the accurate reference value of the single grain cannot be obtained, and an accurate single grain near-infrared analysis model is more difficult to construct. L.e. agelet et al indicate that the grain traits and sizes, the type of near-infrared spectrometer, the collection mode and modeling method of the spectrum, and the determination of the reference value are crucial to establishing an accurate near-infrared model. The determination of the chemical reference value of the single grain is an important factor influencing the success of model establishment: if the error of the chemical reference value of a single grain is large, the error of model prediction becomes large. Aiming at the problem that the single particle detection reference method is not enough in precision, different optimization methods are adopted by experts at home and abroad. For example, the amylose reference technology is improved in J.G.Wu and the like, and a half-grain method is used as a reference method, so that the amylose content of single-grain rice is detected; and for example, Armstrong, P.R and the like average a plurality of single grain spectrums, analyze reference values of the mixed grains, and construct a component analysis model of the near-infrared single-grain crops according to the average spectrums and the reference values.
However, in the former method, the half-grain method detects the half-grain rice without embryo, which does not completely correspond to the collected complete spectrum of the single-grain rice, thereby affecting the analysis result; for the latter method, the spectrum finally used for modeling is the spectrum after the arithmetic mean of the spectrum of each grain, but the quality and the size of each grain are different, so that the contribution of each grain to the finally calculated chemical value of the mixed sample is different, and the average spectrum is not completely corresponding, so that the prediction result is influenced to a certain extent. Therefore, only by improving the amylose reference method to enable the amylose reference method to detect single-grain rice, the near-infrared method can be adopted to accurately realize the rapid nondestructive detection of the amylose content of the single-grain rice.
The method is characterized in that during near-infrared modeling, diffuse transmission spectra of samples are collected, an agricultural ministry standard iodine colorimetric method (NY/T2639) is improved, the amount of the samples is reduced from 50mg to 10mg, a centrifugal tube is used for replacing a volumetric flask, so that chemical detection of single-grain rice amylose is realized, and then on the basis of the method, the spectra and chemical values are associated to construct a single-grain amylose near-infrared amylose content detection model.
Disclosure of Invention
The invention aims to provide a construction method of a near-infrared quantitative analysis model of amylose content of single-grain rice and an amylose micro-detection method.
The invention solves the technical problems through the following technical scheme:
a method for constructing a single-grain rice amylose quantitative analysis model comprises the following steps:
s1, collecting a plurality of parts of single-grain rice samples with different amylose contents, drying, and balancing moisture to be used as a correction set;
s2, collecting the near infrared spectrum of each single-grain rice sample in the correction set;
s3, processing each single-grain rice sample in the correction set into rice flour respectively, obtaining the reference value of the amylose content of each single-grain rice sample, and constructing a correction set reference value matrix;
s4, construction of a near-infrared single-particle amylose model: and screening a spectrum interval from the near infrared spectrum obtained in the step S2 to obtain a corrected light collection spectrum matrix, and performing regression correlation analysis on the spectrum matrix and the reference value matrix in the step S3 to obtain a near infrared quantitative analysis model of the amylose content of the single-grain rice.
Further, each single-grain rice sample in the step S1 is a single-grain rice sample which is full and complete in appearance, free of disease spots, mildew and worm damage.
Further, the near infrared spectrum of each single-grain rice sample in the step S2 is an average value of at least one group of near infrared spectra collected on the front surface and the back surface of each single-grain rice sample.
Further, the near infrared spectrum is a near infrared diffuse transmission spectrum.
Further, in the step S3, the rice flour is brown rice flour or polished rice flour.
Further, the determination of the reference value of the amylose content in the corrected rice collection powder in the step S3 adopts an improved iodine colorimetric method, and the improved iodine colorimetric method specifically comprises the following steps:
(1) and (3) preparing a standard curve: selecting a plurality of amylose standard samples, and carrying out the following steps to obtain a standard curve:
a) pasting: weighing a proper amount of sample powder for each sample, placing the sample in a container with the volume not less than 2ml, adding absolute ethyl alcohol to disperse the sample, slightly beating the container to ensure that the sample powder is fully dispersed, then adding 1mol/L NaOH, then placing the sample in a water bath with the temperature of 95-100 ℃ for heating, taking out the sample and shaking the sample for several times, heating the sample for 15-20 minutes, taking out the sample, cooling the sample, and then adding water to shake the sample uniformly, wherein the weight-volume ratio of the sample powder to the absolute ethyl alcohol, the 1mol/L NaOH solution and the water is 10mg:100 μ L:900 μ L:1 ml;
b) color development: putting the gelatinized liquid, the first added water, HAc, I and the like into a container with the capacity of not less than 5ml, sequentially adding water, 1mol/LHAc and 0.2% I2-KI staining solution, shaking up, adding water again, continuing to shake up, developing for 5-20 minutes, measuring the absorbance at 620nm, repeating the developing process at least twice for each sample, and taking the average value of the repeatedly measured absorbances as the absorbance of the sample, wherein the gelatinized liquid, the first added water, HAc, I and the like2The volume ratio of the KI staining solution to the second addition of water is 40 mul 860 mul 40 mul 60 mul 3000 mul;
c) calculating a standard curve: fitting a standard curve according to the amylose content of the standard sample and the ratio of the absorbance to the mass of the standard sample;
(2) the detection of the amylose content of the calibration set comprises the following steps:
d) pasting: and (3) carrying out shelling and grinding treatment on each correction set single-grain rice sample, sieving the rice sample with a 100-mesh sieve, and drying the rice sample in a 75-85 ℃ oven to constant weight to obtain the required rice flour. The subsequent steps are the same as step a);
e) color development: the process is the same as the step b);
f) calculating the amylose content of the calibration set: calculating the amylose content of the calibration set samples according to the ratio of the absorbance to the mass of the calibration set samples according to the standard curve obtained in step c).
Further, the container is a centrifuge tube.
Further, the step S4 further includes a preprocessing step of the spectrum obtained in S2, where the preprocessing step is: performing first derivative processing on the spectrum obtained in S2, smoothing the points to obtain 17 points, and then cutting out the spectrum range of 11146.8cm-1-9812.3cm-1、9133.5cm-1-8470.1cm-1And 7791.3cm-1-7127.9cm-1And obtaining a corrected collection spectrum matrix after pretreatment.
Further, in the step S4, a partial least square method is used to construct a near-infrared quantitative analysis model of amylose content in the single-grain rice, and the number of principal components in the partial least square method is set to 14.
The invention also provides a single-grain rice amylose detection method, which comprises the following steps:
1) selecting a plurality of single-grain rice which is full and complete in appearance, free of disease spots, mildew and worm damage as a to-be-detected set, and respectively collecting average values of at least one group of near infrared spectrum data on the front side and the back side of each single-grain rice;
2) preprocessing the near infrared spectrum data average value obtained in the step 1) and intercepting the spectrum range, and calculating the amylose content of each sample of the to-be-detected set by using the model constructed by the single-grain rice amylose quantitative analysis model construction method of any one of claims 1 to 9.
Further, the step of preprocessing the near infrared spectrum data average value and intercepting the spectrum range in the step 2) comprises the following steps: performing first derivative processing on the obtained spectrum, smoothing the points to obtain 17 points, and then cutting out the spectrum range of 11146.8cm-1-9812.3cm-1、9133.5cm-1-8470.1cm-1And 7791.3cm-1-7127.9cm-1Spectrum of (a).
The invention also provides an automatic sorting device for single-kernel rice, which automatically sorts the single-kernel rice by adopting the method as claimed in any one of claims 1 to 10.
During modeling, the evaluation indexes of the established quantitative model are mainly as follows: coefficient of Determination (R)2) Cross-Validation Root Mean Square Error (RMSECV); in the case of external verification, the following evaluation indexes are used for the prediction results: prediction of correlation coefficient (R), Prediction of Root Mean Square Error (RMSEP).
The detailed algorithm is shown in the following formula:
(1) determining the coefficient (R)2)
In the formula, yi,actualThe reference value of the ith correction set sample; y isi,predictedPredicting the near infrared model value of the ith correction set sample; y isi,actualThe average value of the reference values of all calibration set samples; n is the number of samples in the calibration set. R2Is used to evaluate the effect of the model fit established by the calibration. Under the premise of same concentration range, R2The closer to 1, the closer to the reference value the predicted value is, namely the higher the accuracy is; if R is2Equal to 1, a complete fit is indicated; if R is2A negative value indicates that the model fitting effect is very poor. In addition, R2Has a great relationship with the distribution range of the to-be-measured object, and for the to-be-measured object with a wide distribution range, R may appear2Close to 1, but with poor accuracy.
(2) Cross validation Root Mean Square Error (RMSECV)
In the formula, yi,actualTo correct the reference value of the ith sample in the set; y isi,predictedModel prediction value of the ith sample in the cross validation process of the correction set is obtained; n is the number of samples in the calibration set; the smaller the RMSECV, the better the model predicts the samples of the correction set.
(3) Prediction correlation coefficient (R)
In the formula, yi,actualThe reference value of the sample of the ith verification set; y isi,predictedPredicting the near infrared model value of the ith verification set sample; y isi,actualThe average value of the reference values of all the verification set samples; and m is the number of samples in the verification set. The closer the prediction correlation coefficient R is to 1, the closer the predicted value is to the reference value, i.e., the higher the accuracy.
(4) Predicting Root Mean Square Error (RMSECP)
In the formula, yi,actualTo verify the reference value of the ith sample in the set; y isi,predictedTo verify the model prediction value of the ith sample in the set; m is the number of samples in the validation set; the smaller the RMSEP value is, the stronger the prediction capability of the established model is, and the more accurate the prediction result is.
According to the method, a set of near-infrared nondestructive testing automatic sorting device can be designed and established, and the single-grain crops to be tested can be sorted according to the difference of the detected components.
Compared with the prior art, the invention has the advantages that: the rapid nondestructive detection of the amylose content of the single-grain rice is realized by adopting a near-infrared technology; on the basis, during near-infrared analysis, an accurate amylose reference value is obtained by improving an iodine colorimetric method, so that the constructed near-infrared model is more accurate than the past report. Therefore, the method can realize the rapid, nondestructive and accurate detection of the amylose content of the single-grain rice.
Drawings
FIG. 1 is a raw spectrum collected in example 1;
FIG. 2 is a standard curve prepared according to the NY/T2639 method in example 1;
FIG. 3 is a standard curve according to the modified iodometric method of example 1;
FIG. 4 is a scatter plot of the predicted values versus reference values for the near-infrared single-grain rice amylose model versus the validation set of samples in example 1.
Detailed Description
The following examples are given for the detailed implementation and specific operation of the present invention, but the scope of the present invention is not limited to the following examples.
Embodiment of the invention relates to a method for constructing a single-grain rice amylose quantitative analysis model
The specific detection steps of this embodiment are as follows:
s1, collecting samples
In the embodiment, rice varieties with different germplasm sources are collected, including rice indica type rice variety '9311', japonica rice 'Wuyujing No. 7' mutant subjected to heavy ion irradiation, and hundreds of other japonica rice varieties, 160 full, complete and mature single rice seeds without disease spots, mildew, worm erosion and other phenomena on the surface are screened from the rice varieties, and 120 rice seeds are randomly selected to serve as a correction set for modeling.
S2, collecting the near infrared diffuse transmission spectrum of the correction set sample;
the instrument used for spectrum collection is MPA type Fourier transform near infrared spectrometer of Bruker company in Germany, and is provided with an integrating sphere, a PbS detector and OPUS 7.0 data processing and analyzing software.
Fixing an aluminum sheet with a diameter of 30mm and a small hole with a diameter of 2mm in the middle at the detection window, and using diffuse transmission scanning parameters of the sample cup, wherein the spectrum scanning range is 5793cm-1-12489cm-1Resolution of 16cm-1The detection window was scanned 64 times as a background spectrum. After background scanning, a rice sample is flatly placed on an aluminum sheet, and the average spectrum is calculated after 1 spectrum is collected on the front surface and the back surface of the rice respectively to be used as the spectrum of the sample. The raw spectra collected are shown in FIG. 1.
S3, manually shelling the collected single rice grains, grinding the rice grains into powder by using a mortar and a pestle, and then sieving the powder by using a 100-mesh sieve to prepare rice flour. And determining the reference value of the amylose content of the sample in the correction set by adopting an improved iodine colorimetric method, and constructing a reference value matrix of the sample in the correction set according to the reference value of the amylose content of the sample in the correction set, wherein each row in the reference value matrix of the sample in the correction set represents the reference value of the amylose content of one sample, and different rows represent the reference values of the amylose content of different samples.
The improved iodine colorimetric method comprises the following specific steps:
(1) and (3) preparing a standard curve: selecting 4 parts of proper amylose standard samples, wherein the amylose content of the amylose standard samples is 1.5%, 10.4%, 16.2% and 26.5%, and repeating the steps for 2 times, and obtaining a standard curve by performing the following steps:
a) pasting: weighing about 10mg of powder for each sample, recording the mass of the powder, placing the powder into a centrifuge tube (the capacity of the centrifuge tube is not less than 2ml), adding 100 mu L of absolute ethyl alcohol to disperse the sample, slightly beating the centrifuge tube to ensure that the powder is fully dispersed, adding 900 mu L of 1mol/L NaOH, then placing the sample into a water bath at 95-100 ℃ to heat, taking out the sample during the heating process, shaking the sample for several times, taking out the sample after the heating time is 15-20 minutes, cooling the sample, and then adding 1ml of water to shake the sample uniformly.
b) Color development: adding 40 μ L of gelatinized liquid into new centrifuge tube (the capacity of the centrifuge tube is not less than 5ml), sequentially adding 860 μ L of water, 40 μ L of 1mol/L HAc, and 60 μ L of 0.2% I2The KI staining solution was shaken, 3ml of water was added thereto, shaken, and after 10 minutes of color development, the absorbance at 620nm was measured with a spectrophotometer. Each sample was set to 2 or more repetitions in the development step, and the absorbance of the sample was determined as the average of the absorbance of the 2 repetitions.
c) Calculating a standard curve: and fitting a standard curve according to the amylose content of the standard sample and the ratio of the absorbance to the mass of the standard sample.
(2) The detection of amylose content of calibration set samples comprises the following steps:
d) pasting: each correction set single-grain rice sample is processed into rice flour through hulling, grinding and sieving, and the subsequent steps are the same as the step a);
e) color development: the process is the same as the step b);
f) calculating the amylose content of the calibration set: calculating the amylose content of the calibration set samples according to the ratio of the absorbance to the mass of the calibration set samples according to the standard curve obtained in step c).
The iodine colorimetric method mainly comprises the steps of replacing a metal bath with a water bath to meet the requirements of most laboratories, increasing the dosage of a sample solution to be developed and a corresponding reagent in the developing step, and repeating for 2 times to ensure the precision of the developing step. To verify that the iodine colorimetric method has the same effect on amylose as the traditional iodine colorimetric method (NY/T2639)The detection precision is that 10mg of rice flour is weighed by the method and 50mg of rice flour is weighed by the method in NY/T2639 respectively for the 4 parts of amylose standard samples, the amylose content is detected (repeated for 1 time), standard curves established by the two methods are respectively shown in figures 2 and 3, and the determination coefficients r of the absorbance/mass and the true value of the amylose are respectively shown in figure 2 and figure 320.9987 and 0.9972, respectively. The correlation r between the absorbance/weight values of the 4 standards measured in both methods was 0.9950 (not shown), and it can be seen that the modified iodometric method is also feasible for amylose detection. Wherein r is2The two methods (traditional method and modified iodine colorimetric method) establish the determination coefficient of a standard curve, and r refers to the correlation coefficient (correlation coefficient) between two results measured by the two methods.
As can be seen from the figure, the standard curve constructed by the improved method has the precision close to that of the method in NY/T2639, which shows that the adopted improved iodine colorimetric method can accurately detect the amylose chemical value of 10mg of samples.
The amylose reference value statistics for the calibration and validation sets after measurement using the modified iodometric method are shown in table 1. As can be seen from the table, the two have a closer average value, a lower standard error and standard deviation. The correction set has an amylose content range (1.35% -26.74%) greater than the verification set (1.63% -24.45%), indicating that the correction set can better represent the verification set and most rice varieties.
TABLE 1 improved iodine colorimetric method for detecting amylose content in single-grain rice
| Species of | Calibration set sample | Verification set sample |
| Number of samples | 120 | 40 |
| Range of amylose content | 1.35-26.74 | 1.63-24.45 |
| Average amylose content | 13.54 | 13.46 |
| Standard error of | 0.61 | 1.09 |
| Standard deviation of | 6.71 | 6.91 |
S4, construction of a near-infrared single-particle amylose model: processing the original spectrum of the correction set by using a proper pretreatment method, and screening a spectrum interval to obtain a pretreated spectrum matrix; and according to the obtained spectrum matrix and the reference value matrix of the correction set sample, a regression model is constructed by adopting a partial least square method. Model construction was implemented on matlab 2015b software (The Mathworks, Natick, MA, USA).
Different preprocessing methods, different spectral ranges and different numbers of selected principal components during partial least squares regression are adopted, and finally the models are different in performance. And screening an optimal correction model by comparing different preprocessing methods and different combinations of spectral ranges. The pretreatment method is selected from 4 methods including no pretreatment, first-order derivative (default 17-point smoothing), standard normal variable transformation (SNV), first-order derivative and SNV transformation; the spectral range is screened by equally cutting 10 segments of the data points in the full spectral range (for example, 870 spectral points exist in the spectrum of the full spectral range, corresponding to 870 data, each segment in the 10 segments contains 87 spectral points), and the spectral range is cut as shown in table 2.
Any combination of 1 segment, 2 segments, 3 segments and 4 segments is exhausted, and the optimal spectral range is screened from the combinations of 210 spectral ranges in total. The preprocessing method and the spectrum range are combined in 840 kinds, a partial least square model is established for the 840 kinds of combination, and R within 20 main components is selected2The highest model and the lowest RMSECV model are used as the optimal model, and the corresponding preprocessing method, the spectrum range and the principal component number are optimal partial least squares modeling parameters. The results of the partial comparison of the models are shown in table 3.
TABLE 2 Spectrum numbering and corresponding spectral ranges
TABLE 3 modeling performances under different preprocessing methods and different spectral ranges
As can be seen from Table 3, the spectral range of the first derivative in the pretreatment method is 11146.8cm-1-9812.3cm-1、9133.5cm-1-8470.1cm-1And 7791.3cm-1-7127.9cm-1(corresponding to the 3 rd, 4 th, 6 th and 8 th sections in Table 3) and the number of principal components is 14, the cross-validation result of the built model is best, and R is the cross-validation result2The RMSECV was 0.8563, 2.55. The model is therefore selected for subsequent validation, prediction.
In order to verify the prediction performance of the model, 40 full, complete and mature single rice seeds without surface disease spots, mildew, worm erosion and the like are selected from the mutant library in the step S1 again to serve as a verification set sample. Spectra were collected from these samples in the same manner as in step S2, in synchronization with step S3The method collects the reference values, in synchronization with the same spectral pre-processing (i.e. first derivative, spectral range 11146.8 cm) in step S4-1-9812.3cm-1、9133.5cm-1-8470.1cm-1And 7791.3cm-1-7127.9cm-1) The verification set raw spectrum is processed, and then the preprocessed spectrum is predicted using the model constructed in step S4, with the number of principal components set to 14. Model prediction was implemented on matlab 2015b software (The Mathworks, Natick, MA, USA). A scatter plot of the predicted values versus the reference values is shown in FIG. 4.
As can be seen from FIG. 4, the model has a higher external verification correlation coefficient R2(0.9511), and lower RMSEP (2.1135), validation set predictor to reference value paired t-test p value of 0.926>0.05, which is superior to that reported by J.G.Wu et al (Field Crops Research, 2004) for single-grain rice (R.G.Wu et al20.66 and RMSEP 4.69), which indicates that the model can effectively predict the samples of the verification set, and thus can be popularized on single-grain rice of other similar germplasm.
When predicting unknown rice seed samples, the single rice spectrum to be tested is acquired by the same method as the step S2, and the same spectrum pretreatment (namely, the first derivative, the spectrum range of 11146.8 cm) in the step S4 is synchronized-1-9812.3cm-1、9133.5cm-1-8470.1cm-1And 7791.3cm-1-7127.9cm-1) And (4) processing the original spectra of the single-grain rice, and predicting the preprocessed spectra by using the model constructed in the step S4 to obtain the result, namely the amylose content of the single-grain rice to be detected.
It should be noted that the detection method of the present invention is not limited to the application scenario described in the above embodiment, and further, a set of near-infrared amylose content automatic sorting apparatus for single-grain rice with different components may be designed and established, and software in the apparatus may integrate the method to design a near-infrared model and perform spectrum collection and prediction on single-grain rice to be detected, so as to meet the requirements of rapid nondestructive detection and sorting of single-grain rice with different amylose contents in breeding and food industries.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
Claims (10)
1. A method for constructing a single-grain rice amylose quantitative analysis model is characterized by comprising the following steps:
s1, collecting a plurality of parts of single-grain rice samples with different amylose contents, drying, and balancing moisture to be used as a correction set;
s2, collecting the near infrared spectrum of each single-grain rice sample in the correction set;
s3, processing the single-grain rice samples in the correction set into rice flour, obtaining the reference value of the amylose content of the single-grain rice samples by an improved iodine colorimetric method, and constructing a reference value matrix of the correction set;
the improved iodine colorimetric method comprises the following specific steps:
(1) and (3) preparing a standard curve: selecting a plurality of amylose standard samples, and carrying out the following steps to obtain a standard curve:
a) pasting: weighing a proper amount of sample powder for each sample, placing the sample in a container with the volume not less than 2ml, adding absolute ethyl alcohol to disperse the sample, slightly beating the container to ensure that the sample powder is fully dispersed, then adding 1mol/L NaOH, then placing the sample in a water bath with the temperature of 95-100 ℃ for heating, taking out the sample and shaking the sample for several times, heating the sample for 15-20 minutes, taking out the sample, cooling the sample, and then adding water to shake the sample uniformly, wherein the weight-volume ratio of the sample powder to the absolute ethyl alcohol, the 1mol/L NaOH solution and the water is 10mg:100 μ L:900 μ L:1 ml;
b) color development: putting the gelatinized liquid into a container with the capacity of not less than 5ml, sequentially adding water, 1mol/L HAc and 0.2% I2-KI staining solution, shaking up, adding water again, continuing shaking up, developing for 5-20 min, measuring the absorbance at 620nm, repeating the developing process at least twice for each sample, and taking the average value of the repeatedly measured absorbances as the absorbance of the sample, wherein the gelatinized liquid, the HAc and the I2-KI staining solution,First addition of water, HAc, I2The volume ratio of the KI staining solution to the second addition of water is 40 mul 860 mul 40 mul 60 mul 3000 mul;
c) calculating a standard curve: fitting a standard curve according to the amylose content of the standard sample and the ratio of the absorbance to the mass of the standard sample;
(2) the detection of the amylose content of the calibration set comprises the following steps:
d) pasting: carrying out shelling and grinding treatment on each correction set single-grain rice sample, sieving the rice sample with a 100-mesh sieve, and drying the rice sample in a 75-85 ℃ oven to constant weight to obtain required rice flour; the subsequent steps are the same as step a);
e) color development: the process is the same as the step b);
f) calculating the amylose content of the calibration set: calculating the amylose content of the calibration set sample according to the ratio of the absorbance of the calibration set sample to the mass of the calibration set sample according to the standard curve obtained in the step c);
s4, construction of a near-infrared single-particle amylose model: and screening a spectrum interval from the near infrared spectrum obtained in the step S2 to obtain a corrected light collection spectrum matrix, and performing regression correlation analysis on the spectrum matrix and the reference value matrix in the step S3 to obtain a near infrared quantitative analysis model of the amylose content of the single-grain rice.
2. The method for constructing the single-grain rice amylose quantitative analysis model according to claim 1, wherein each single-grain rice sample in the step S1 is a single-grain rice with full and complete appearance, no disease spots, no mildew and no worm damage.
3. The method for constructing the single-grain rice amylose quantitative analysis model according to claim 1, wherein the method for acquiring the near infrared spectrum of each single-grain rice sample in the step S2 comprises the following steps:
an aluminum sheet with a diameter of 30mm and a small hole with a diameter of 2mm in the middle is fixed on the detection window, and the spectrum scanning range is 5793cm-1-12489cm-1Resolution of 16cm-1Scanning the detection window as a background spectrum; flatly placing the rice sample on aluminumOn the sheet, 1 spectrum was collected from each of the front and back sides of rice, and the average spectrum was calculated as the spectrum of the sample.
4. The method for constructing the single-grain rice amylose quantitative analysis model according to claim 1 or 3, wherein the near infrared spectrum is a near infrared diffuse transmission spectrum.
5. The method for constructing a single-grain rice amylose quantitative analysis model according to claim 1, wherein the rice flour is brown rice flour or polished rice flour in step S3.
6. The method for constructing the single-grain rice amylose quantitative analysis model according to claim 1, wherein the container is a centrifuge tube.
7. The method for constructing a single-grain rice amylose quantitative analysis model as claimed in claim 1, wherein the step S4 further comprises a pretreatment step of the spectrum obtained in S2, the pretreatment step is as follows: performing first derivative processing on the spectrum obtained in S2, smoothing the points to obtain 17 points, and then cutting out the spectrum range of 11146.8cm-1-9812.3cm-1、9133.5cm-1-8470.1cm-1And 7791.3cm-1-7127.9cm-1And obtaining a corrected collection spectrum matrix after pretreatment.
8. The method for constructing the single-grain rice amylose quantitative analysis model according to claim 1, wherein a partial least square method is adopted in the step S4 to construct the single-grain rice amylose content near-infrared quantitative analysis model, and the number of principal components in the partial least square method is set to be 14.
9. A single-grain rice amylose detection method is characterized by comprising the following steps:
1) selecting a plurality of single-grain rice which is full and complete in appearance, free of disease spots, mildew and worm damage as a to-be-detected set, and respectively collecting average values of at least one group of near infrared spectrum data on the front side and the back side of each single-grain rice;
2) preprocessing the near infrared spectrum data average value obtained in the step 1) and intercepting the spectrum range, and calculating the amylose content of each sample of the to-be-detected set by using the model constructed by the single-grain rice amylose quantitative analysis model construction method of any one of claims 1 to 8.
10. The method as claimed in claim 9, wherein the step of preprocessing the near infrared spectrum data average value and spectrum range clipping in the step 2) comprises the following steps: performing first derivative processing on the obtained spectrum, smoothing the points to obtain 17 points, and then cutting out the spectrum range of 11146.8cm-1-9812.3cm-1、9133.5cm-1-8470.1cm-1And 7791.3cm-1-7127.9cm-1Spectrum of (a).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810982153.9A CN108680515B (en) | 2018-08-27 | 2018-08-27 | Single-grain rice amylose quantitative analysis model construction and detection method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810982153.9A CN108680515B (en) | 2018-08-27 | 2018-08-27 | Single-grain rice amylose quantitative analysis model construction and detection method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108680515A CN108680515A (en) | 2018-10-19 |
| CN108680515B true CN108680515B (en) | 2021-01-26 |
Family
ID=63815518
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810982153.9A Active CN108680515B (en) | 2018-08-27 | 2018-08-27 | Single-grain rice amylose quantitative analysis model construction and detection method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108680515B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109323911A (en) * | 2018-12-14 | 2019-02-12 | 扬州大学 | A kind of starch iodine absorption spectrum method for the rice silty seed mutant that quick screening starch ingredients change |
| CN111122470B (en) * | 2019-12-27 | 2023-01-24 | 海南大学 | Single-grain wheat amylose content detection method |
| CN112595675B (en) * | 2020-09-29 | 2022-03-08 | 华中农业大学 | Hyperspectrum-based rice grain quality functional gene analysis method |
| CN112683840A (en) * | 2020-10-29 | 2021-04-20 | 河南工业大学 | Method for rapidly and nondestructively measuring amylose content of single wheat grain by utilizing near infrared spectrum technology |
| CN112345323A (en) * | 2020-10-31 | 2021-02-09 | 中国水稻研究所 | Method for developing standard substance of amylose content of rice |
| CN113484270A (en) * | 2021-06-04 | 2021-10-08 | 中国科学院合肥物质科学研究院 | Construction and detection method of single-grain rice fat content quantitative analysis model |
| CN114018859B (en) * | 2021-10-13 | 2023-03-17 | 中国水稻研究所 | Method for rapidly and synchronously measuring apparent amylose, amylose and amylopectin contents of rice flour |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0260942A2 (en) * | 1986-09-19 | 1988-03-23 | Satake Engineering Co., Ltd. | Measuring apparatus for amylose and/or amylopectin content in rice |
| JP2000352555A (en) * | 1992-03-08 | 2000-12-19 | Iseki & Co Ltd | Apparatus for detecting internal quality of rice |
| CN1590983A (en) * | 2003-09-02 | 2005-03-09 | 中国农业大学 | Method of detecting amylose content of rice |
| KR20090047997A (en) * | 2007-11-09 | 2009-05-13 | 대한민국(관리부서:농촌진흥청) | Component measuring apparatus for grain and measuring method using the same |
| CN104833671A (en) * | 2015-01-09 | 2015-08-12 | 中国水稻研究所 | Measurement method of absolute amylase content of rice |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106706554A (en) * | 2016-03-17 | 2017-05-24 | 西北农林科技大学 | Method for rapidly and nondestructively determining content of straight-chain starch of corn single-ear grains |
| CN106706553A (en) * | 2016-03-17 | 2017-05-24 | 西北农林科技大学 | Method for quick and non-destructive determination of content of amylase in corn single grains |
| CN107515203A (en) * | 2017-07-19 | 2017-12-26 | 中国农业大学 | The research of near infrared technology quantitative analysis rice single grain amylose content |
-
2018
- 2018-08-27 CN CN201810982153.9A patent/CN108680515B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0260942A2 (en) * | 1986-09-19 | 1988-03-23 | Satake Engineering Co., Ltd. | Measuring apparatus for amylose and/or amylopectin content in rice |
| JP2000352555A (en) * | 1992-03-08 | 2000-12-19 | Iseki & Co Ltd | Apparatus for detecting internal quality of rice |
| CN1590983A (en) * | 2003-09-02 | 2005-03-09 | 中国农业大学 | Method of detecting amylose content of rice |
| KR20090047997A (en) * | 2007-11-09 | 2009-05-13 | 대한민국(관리부서:농촌진흥청) | Component measuring apparatus for grain and measuring method using the same |
| CN104833671A (en) * | 2015-01-09 | 2015-08-12 | 中国水稻研究所 | Measurement method of absolute amylase content of rice |
Non-Patent Citations (1)
| Title |
|---|
| 冯光." 江淮稻区精米直链淀粉含量近红外测定模型的建立及稻米食味品质的研究".《中国优秀硕士学位论文全文数据库 农业科技辑》.2012,第1-36页. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108680515A (en) | 2018-10-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108680515B (en) | Single-grain rice amylose quantitative analysis model construction and detection method thereof | |
| CN106706553A (en) | Method for quick and non-destructive determination of content of amylase in corn single grains | |
| CN113049530B (en) | Single-seed corn seed moisture content detection method based on near-infrared hyperspectrum | |
| CN109470648B (en) | Rapid nondestructive determination method for imperfect grains of single-grain crops | |
| CN108613943B (en) | Near-infrared single-grain crop component detection method based on spectrum morphology transfer | |
| CN108169165B (en) | Maltose mixture quantitative analysis method based on terahertz spectrum and image information fusion | |
| Wang et al. | Quantitative analysis of fat content in rice by near‐infrared spectroscopy technique | |
| WO2020248961A1 (en) | Method for selecting spectral wavenumber without reference value | |
| CN102393376A (en) | Support vector regression-based near infrared spectroscopy for detecting content of multiple components of fish ball | |
| CN108169168A (en) | Test and analyze rice grain protein content mathematical model and construction method and application | |
| CN111912793A (en) | Method for measuring cadmium content in tobacco by hyperspectral and establishment of prediction model | |
| CN115201148A (en) | Method for rapidly detecting content of fatty acid in rice based on near infrared spectrum | |
| CN110231302A (en) | A kind of method of the odd sub- seed crude fat content of quick measurement | |
| Rathna Priya et al. | Characterising corn grain using infrared imaging and spectroscopic techniques: a review | |
| CN106706554A (en) | Method for rapidly and nondestructively determining content of straight-chain starch of corn single-ear grains | |
| CN106770016B (en) | NIR (near infrared) transmission spectrum measurement method for protein quantitative analysis of single rice seeds | |
| CN111693487A (en) | Fruit sugar degree detection method and system based on genetic algorithm and extreme learning machine | |
| CN107796779A (en) | The near infrared spectrum diagnostic method of rubber tree LTN content | |
| CN102519903B (en) | Method for measuring whiteness value of Agaricus bisporus by using near infrared spectrum | |
| CN110231306A (en) | A kind of method of lossless, the quick odd sub- seed protein content of measurement | |
| CN112213281A (en) | Comprehensive evaluation method for rapidly determining freshness of freshwater fish based on transmission near infrared spectrum | |
| CN109540837A (en) | The method that near-infrared quickly detects Boehmeria nivea leaves wood fibre cellulose content | |
| CN109520964A (en) | Seed Vigor Index, system and storage medium based on NIRS | |
| CN102175620A (en) | Method for measuring lysine in corn through near infrared spectrum | |
| CN112129726B (en) | Method, system, medium and equipment for evaluating uniformity of tobacco stem shreds |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |