A method for simulating starch content and its component contents

By constructing and calibrating the starch accumulation and component simulation model of wheat grains, the problem that the existing technology cannot predict the starch content and component content is solved, and the accurate prediction of the starch content and components of wheat grains is achieved, and the quality of grain is promoted Simulation and wheat variety breeding process.

CN118227929BActive Publication Date: 2024-11-22INSTITUTE OF ENVIRONMENT AND SUSTAINABLE DEVELOPMENT IN AGRICULTURE CAAS
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410203925.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-02-23
Publication Date
2024-11-22
Estimated Expiration
2044-02-23

AI Technical Summary

Technical Problem

The existing technology cannot effectively predict the starch content and component content in wheat grains, especially the failure to simulate the dynamic starch accumulation process using the APSIM-Wheat model, which limits the application of grain quality simulation and the process of wheat variety selection and breeding. .

Method used

The starch accumulation simulation model of wheat grains was constructed, and the starch component simulation model was established through Linear Regression. The model was calibrated and verified using experimental data, and the APSIM-Wheat model was embedded to predict starch content and component content. The variables were screened and Lasso regression and Linear were used. Regression performs model optimization.

Benefits of technology

Accurate prediction of the starch content and component content of wheat grains is achieved, the gap in APSIM-Wheat model simulation was filled, the application of grain quality simulation was promoted, and key information was provided for wheat variety selection and breeding.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118227929B_ABST
    Figure CN118227929B_ABST
Patent Text Reader

Abstract

The present invention provides a method for simulating starch content and its component content, and belongs to the technical field of crop starch content simulation. In view of the problem that the starch content and starch component content cannot be predicted in the prior art, the present invention constructs a wheat grain starch accumulation simulation model based on the APSIM-Wheat model, and uses the model to simulate the thousand-grain weight, yield, number of grains per ear and total starch content of wheat grains. At the same time, a wheat grain starch component simulation model is established based on Linear Regression with the variables selected in the model, and the two simulation models are calibrated and verified respectively by experimental data, and the simulated values ​​are compared with the measured values, and a linear evaluation equation is fitted to evaluate the two simulation models. The method of the present invention promotes the widespread application of food quality simulation, and can obtain grain starch information in the process of variety breeding, which is helpful to screen out excellent wheat varieties.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of crop starch content simulation, and in particular to a method for simulating starch content and the content of its components. Background Art

[0002] Wheat is one of the most important food crops in the world. Starch is an important component of wheat grains. According to its molecular composition structure, starch can be divided into amylose and amylopectin. The chemical properties of amylose molecules with less branching have better nutritional value. Since amylose is digested more slowly, it is beneficial to human health, can increase the body's fat burning ability, and help maintain a healthy weight. Based on this, the market will have a high demand for high-amylose grains in the future. The different ratios of amylose and amylopectin content in wheat will lead to different industrial processing uses of different varieties of wheat, which will also directly affect the choice of breeding work.

[0003] At present, many studies are mostly aimed at the detection of starch content, amylose and amylopectin in wheat grains. No study has publicly used the APSIM-Wheat model to simulate the dynamic accumulation process of starch in wheat grains or the content of starch components (amylose, amylopectin). However, the simulation study of starch content and its component contents can not only promote the widespread application of grain quality simulation, but also help wheat variety breeders obtain information on grain starch, which plays a positive role in the breeding of high-quality varieties. Summary of the invention

[0004] In view of the above-mentioned deficiencies, the present invention aims to propose a method for simulating starch content and the content of its components, thereby effectively solving the problem of being unable to predict starch content and the content of its components in the background technology.

[0005] In order to achieve the above technical objectives, the present invention adopts the following technical solutions:

[0006] A method for simulating starch content and the content of its components comprises the following steps:

[0007] S1: Construct a starch accumulation simulation model for wheat grains;

[0008] S2: Screen the variables output by the starch accumulation simulation model, and establish a starch component simulation model of wheat grains based on Linear Regression with the screened variables whose coefficients are not zero;

[0009] S3: The starch accumulation simulation model and the starch component simulation model were calibrated and verified using experimental data, the simulated values ​​obtained by the two models were compared with the measured values, and linear evaluation equations were fitted to evaluate the two simulation models.

[0010] A further preferred embodiment of the above scheme is as follows: in step S1, the specific process of constructing the starch accumulation simulation model includes:

[0011] S101: obtaining the starch accumulation rate of einkorn wheat;

[0012]

[0013] Among them: STR i is the starch accumulation rate of einkorn wheat on day i, STR m is the maximum rate of starch accumulation of individual grains, f(A i ) is the factor of starch synthesis capacity of grains on day i, f(T i ) is temperature, f(N i ) is the nitrogen supply for plants, f(W i ) is the water status of the plant, GCA i is the available carbon per grain on day i;

[0014] S102: Determine the starch accumulation of the single grain wheat using the starch accumulation rate, the formula is as follows:

[0015]

[0016] Where: GST0 is the initial starch content of einkorn wheat, GST i and GST i-1 represent the starch accumulation of einkorn wheat on the i-th day and the i-1-th day of grain filling, respectively, and STR1 is the starch accumulation rate of einkorn wheat on the first day.

[0017] Further: In step S101, f(A i ) is related to the soluble carbohydrate content in wheat grains. It increases exponentially and then decreases linearly with the increase of accumulated temperature after flowering. Its expression is as follows:

[0018]

[0019] Among them: GDD i is the accumulated temperature after flowering on the ith day, GDD m It means that when f(A i ) and the accumulated temperature when the grain filling rate reaches the maximum;

[0020] f(T i ) is as follows:

[0021]

[0022] Where: T maxi is the maximum daily temperature, T miniis the daily minimum temperature, T i is the daily average temperature, T o The optimal temperature for wheat starch accumulation;

[0023] f(N i ) is as follows:

[0024]

[0025] Among them: CNP i is the critical nitrogen concentration for plants, MNP i is the minimum plant nitrogen concentration, ANP i is the actual dry nitrogen concentration of the plant;

[0026] f(W i ) is as follows:

[0027]

[0028] Where: T a is the actual transpiration of wheat plants, T p is potential transpiration;

[0029] GCA i Determined by the immediate assimilates transported directly to the kernel and the remobilized assimilates from storage, it is expressed as follows:

[0030] GCA i =GCP i +GCT i

[0031] Among them: GCT i GCT is an immediate assimilate for grain utilization. i To remobilize assimilates.

[0032] Going a step further: the actual dry nitrogen concentration of the plant ANP i The critical nitrogen concentration of plants, CNP, is determined by the amount of nitrogen absorbed and the total dry matter. i Affected by the accumulated aboveground biomass, the expressions are as follows:

[0033]

[0034]

[0035] Among them: NUP i is the N absorption, TW i is the total dry matter mass, GB i To accumulate aboveground biomass.

[0036] Going a step further: Instant Assimilate GCP iGCT i The expressions are:

[0037]

[0038]

[0039] Among them: TOPWT i and TOPWT i-1 DLTDM is the total aboveground weight of wheat plants on the i-th day and the i-1th day after anthesis. i is the actual dry matter accumulation on the ground on day i, VWT i and VWT i-1 is the total weight of vegetative organs on the i-th day and the i-1th day after anthesis;

[0040]

[0041] Among them: VWT i and VWT i-1 It is the total weight of vegetative organs on the i-th day and i-1 day after flowering.

[0042] A further preferred embodiment of the above scheme is as follows: in step S2, when screening the variables output by the starch accumulation simulation model, the coefficients of the variables with weak correlation output by the starch accumulation simulation model are reduced to zero by adding an L1 regularization term using the Lasso regression method, and the variables whose coefficients are not zero are GB, TOPWT and VWT.

[0043] A further preferred embodiment of the above scheme is as follows: in step S2, the steps of establishing a starch component simulation model of wheat grains are as follows:

[0044] S201: Normalize GB, TOPWT and VWT. The formula is as follows:

[0045]

[0046] Where: x ND is the normalized variable, x is the measured value of the variable, and x max and x min are the highest and lowest values ​​of the variable measured, respectively;

[0047] S202: Based on Linear Regression, a starch component simulation model of wheat grains is established with normalized variables, and its expression is as follows:

[0048] f(x) amylose =8.108-3.733x1+6.429x2-3.853x3

[0049] f(x) amylopectin =23.135-1.093x1+6.250x2-1.093x3

[0050] Where: f(x) amylose is the amylose content, f(x) amylopectin is the amylopectin content, and x1, x2, and x3 are the normalized GB, TOPWT, and VWT, respectively.

[0051] A further preferred embodiment of the above scheme is as follows: In step S3, the specific process of calibrating and verifying the starch accumulation simulation model and the starch component simulation model respectively using the test data is as follows:

[0052] S301: calibrating and verifying the starch accumulation simulation model and the starch component simulation model using the first test data, and comparing the simulation values ​​with the measured values;

[0053] S302: Evaluate the starch accumulation simulation model and the starch component simulation model using the second test data, and obtain a linear evaluation equation by fitting.

[0054] Further: In step S3, by determining the index R 2 , normalized root mean square error RMSE and Wilmot's consistency index D value The accuracy of the simulated values ​​of the two models was evaluated, including:

[0055]

[0056]

[0057]

[0058] in: is the predicted value, y i is the measured value, is the average of the measured values, and n is the number of samples.

[0059] Compared with the prior art, the present invention has the following beneficial effects:

[0060] 1. The present invention embeds a wheat starch content simulation module in the APSIM-Wheat model, constructs a starch accumulation simulation model of plant carbon dynamics during the filling process of wheat grains, and uses the three variables GB, TOPWT and VWT output by the module to construct a wheat grain starch component simulation model based on Linear Regression, providing information for wheat variety breeders to obtain grain starch and its component content. Field trials were conducted using 17 wheat varieties from 2018 to 2020, and data sets including phenotypes, climate, soil data and farm management records were collected to calibrate and evaluate the starch content simulation module based on the APSIM-Wheat model. The results show that the starch content simulation module based on the APSIM-Wheat model can accurately predict the grain weight, yield, number of grains and starch content of wheat grains of different varieties, and the wheat grain starch component simulation model can reliably predict the starch component content of different wheat varieties.

[0061] 2. The starch content simulation module based on the APSIM-Wheat model of the present invention can also accurately simulate the "S" dynamic accumulation trend of starch content in wheat grains. At the same time, the two starch component models can also accurately predict the starch component content of different varieties of wheat grains. This study fills the gap that the APSIM-Wheat model cannot simulate wheat starch and its component content at the current stage, and greatly promotes the widespread application of grain quality simulation. In addition, this study can also provide key information for new variety breeding and provide a new direction for screening excellent wheat varieties. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments.

[0063] Figure 1 is the geographical location of the integrated test station of the present invention;

[0064] Figure 2 The temperature radiation and precipitation of the comprehensive test station of the present invention;

[0065] Figure 3 Basic information and distribution of wheat materials of the present invention;

[0066] Figure 4 To verify the simulated and measured values ​​of grain weight, yield, grain number and total starch content using the test data of 17 wheat varieties from 2018 to 2019;

[0067] Figure 5 To evaluate grain weight, yield, number of grains and total starch content using trial data of 17 wheat varieties from 2019 to 2020;

[0068] Figure 6 To simulate the wheat amylose content and amylopectin content using 17 wheat varieties from 2018 to 2019. DETAILED DESCRIPTION

[0069] The technical solution of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application described and shown in the accompanying drawings can be arranged and designed in various different configurations.

[0070] Reference Figure 1-6 The present invention provides a method for simulating starch content and the content of its components, comprising the following steps:

[0071] S1: Construct a starch accumulation simulation model for wheat grains;

[0072] S2: Screen the variables output by the starch accumulation simulation model, and establish a starch component simulation model of wheat grains based on Linear Regression with the screened variables whose coefficients are not zero;

[0073] S3: The starch accumulation simulation model and the starch component simulation model were calibrated and verified using experimental data, the simulated values ​​obtained by the two models were compared with the measured values, and linear evaluation equations were fitted to evaluate the two simulation models.

[0074] Based on the APSIM-Wheat crop model, the present invention studies and analyzes the dynamic accumulation process of plant carbon in wheat grains during the grain filling process, and finally establishes the starch accumulation simulation model of wheat grains of the present invention, which is as follows:

[0075] A: Obtain the starch accumulation rate of einkorn wheat;

[0076]

[0077] STR m =-0.22*GCA i 2 +2.2474*GCA i -1

[0078] Among them: STR i is the starch accumulation rate of the single grain wheat on day i (mg·grain -1 ·day -1 ), which is determined by the availability of carbon sources in the plant and the ability to synthesize starch in the grain. m is the maximum rate of starch accumulation of individual grains (mg·grain -1 ·day -1 ), f(Ai ) is the factor of starch synthesis capacity of grains on day i, f(T i ) is temperature, f(N i ) is the nitrogen supply for plants, f(W i ) is the water status of the plant, GCA i is the available carbon per grain on day i;

[0079] In this formula, f(A i ) is closely related to the soluble carbohydrate content in wheat grains, which increases exponentially with the increase of accumulated temperature after flowering and then decreases linearly. Its expression is as follows:

[0080]

[0081] Among them: GDD i is the accumulated temperature after flowering on the ith day (0℃ reference), GDD m It means that when f(A i ) and the accumulated temperature when the grain filling rate reaches the maximum can be estimated to be 203-250℃·d.

[0082] In this formula, f(T i ) is the direct effect of temperature on grain starch accumulation, which depends on the temperature conditions during grain filling and is expressed as follows:

[0083]

[0084] Where: T maxi is the maximum daily temperature, T mini is the daily minimum temperature, T i is the daily average temperature, T o The best temperature for wheat starch accumulation can be set at 23°C;

[0085] In this formula, f(N i ) is the effect of plant nitrogen status on the starch accumulation rate of grains, and its expression is:

[0086]

[0087] Among them: CNP i is the critical nitrogen concentration for plants, MNP i is the minimum plant nitrogen concentration, set as CNP i 60% of ANP i is the actual dry nitrogen concentration of the plant (%);

[0088] ANP i and CNP i It can be further expressed as follows:

[0089]

[0090]

[0091] Among them: NUP i is the N absorption (g / m 2 ), TW i is the total dry matter mass (g / m 2 ), GB i is the cumulative aboveground biomass (kg·ha -1 ).

[0092] In this formula, f(W i ) is the effect of water status on the accumulation rate of grain starch, and its expression is:

[0093]

[0094] Where: T a is the actual transpiration of wheat plants, T p is potential transpiration;

[0095] In this formula, GCA i Determined by the immediate assimilates transported directly to the kernel and the remobilized assimilates from storage, it is expressed as follows:

[0096] GCA i =GCP i +GCT i

[0097] Among them: GCP i GCT is an immediate assimilate for grain utilization. i To remobilize assimilates.

[0098] GCP i and GCT i It can be further expressed as follows:

[0099]

[0100]

[0101] Among them: TOPWT i and TOPWT i-1 is the total aboveground weight of wheat plants on the i-th day and the i-1th day after anthesis (g·m -2 ), DLTDM i is the actual dry matter accumulation on the ground on day i (g·m -2 ), VWT i and VWT i-1 is the total weight of the vegetative organs on the i-th day and the i-1th day after anthesis (g·m -2), which corresponds to the constant rate of dry matter accumulation in grains and the time of the initial decline in net assimilation rate. According to experimental data analysis, when the number of days i>GDD m , the assimilates stored in the sheath begin to transfer to the grains, and the weight of the nutritional organs gradually decreases.

[0102]

[0103] Among them: VWT i and VWT i-1 is the total weight of the vegetative organs on the i-th day and the i-1th day after anthesis (g·m -2 ).

[0104] B: The starch accumulation rate of einkorn wheat is used to determine the starch accumulation amount. The formula is as follows:

[0105]

[0106] Where: GST0 is the initial starch content of einkorn wheat (mg·grain -1 ), take 3.0mg·grain -1 , GST i and GST i-1 represents the starch accumulation of the einkorn wheat on the i-th day and the i-1-th day after grain filling, and STR1 is the starch accumulation rate of the einkorn wheat on the first day (mg·grain -1 ·day -1 ).

[0107] The corresponding relationship between the variables in the starch component simulation model of wheat grains of the present invention and the output variables of the APSIM-Wheat model is shown in Table 1.

[0108] Table 1 Starch simulation module variables and APSIM-Wheat model output variables

[0109]

[0110]

[0111] In order to predict the amylose content and amylopectin content of wheat grains, the model was calibrated and the regression parameters were verified using experimental data based on LinearRegression, and finally a starch component simulation model of wheat grains was established.

[0112] In the present invention, the Lasso regression method is used to screen the variables output by the model. By adding the L1 regularization term in the loss function, some characteristic coefficients with weak correlation are effectively reduced to zero (refer to Table 2). This method effectively reduces the risk of model overfitting and solves the problem of multicollinearity; finally, the characteristic variables with non-zero coefficients are used for modeling, and the final variables with non-zero coefficients include GB, TOPWT and VWT.

[0113] The L1 regularization added to the loss function of the present invention, also known as Lasso regularization (Least Absolute Shrinkage and Selection Operator), is a regularization method in machine learning and statistical modeling, which penalizes the complexity of the model by adding the L1 norm of the weight vector in the loss function; in models such as linear regression, the goal of L1 regularization is to minimize the sum of the loss function and the L1 norm of the weight vector; the main feature of L1 regularization is that it tends to make some weights zero, thereby achieving feature selection, which means that through L1 regularization, the model can more easily ignore irrelevant or redundant features, thereby improving the generalization ability of the model.

[0114] Table 2 Lasso regression variable correlation coefficient table

[0115]

[0116]

[0117] GB represents the active, photosynthetic part of the plant, usually including green tissues such as leaves and stems. This part of biomass is the basis of plant growth and development, and is also the main source of energy and nutrients; TOPWT represents the total aboveground dry weight of wheat plants; VWT represents the total dry weight of nutritional organs; the comprehensive consideration of these factors can help us more accurately predict the starch content of wheat grains.

[0118] Subsequently, the above three variables were normalized to eliminate the dimension effect and improve the generalization ability of the simulation model. The formula is as follows:

[0119]

[0120] Where: x ND is the normalized variable, x is the measured value of the variable, and x max and x min are the highest and lowest values ​​of the variable measured, respectively;

[0121] Based on Linear Regression, the starch component simulation model of wheat grains was established with normalized variables, and its expression is as follows:

[0122] f(x) amyoose =8.108-3.733x1+6.429x2-3.853x3

[0123] f(x) amylopectin =23.135-1.093x1+6.250x2-1.093x3

[0124] Where: f(x) amylose is the content of amylose (%), f(x) amylopectin is the amylopectin content (%), x1, x2, x3 are the normalized GB, TOPWT and VWT, respectively.

[0125] The test data used above are derived from wheat planting in the test field of the present invention. The test data are used as the basic data for calibrating and verifying the starch accumulation simulation model and starch component simulation model established by the present invention, and the test data are used as the measured value to compare with the simulation value of the present invention. The detailed design of the source of the test data is as follows, combined with Figure 1-3 As shown:

[0126] (1) Site description: Field experiments were conducted in the 2018-2019 and 2019-2020 growing seasons at the Comprehensive Experimental Station of the Chinese Academy of Agricultural Sciences in Xinxiang, Henan Province (35°18′N, 113°51′E, 78 m above sea level) (Figure 1). The region is characterized by a warm temperate semi-humid climate. The maximum and minimum temperatures over the years were 20.4°C and 10.0°C, respectively, and the total solar radiation was 4641.5 MJ / m 2 The total precipitation was 768.5 mm. The cumulative temperatures during the two growing seasons were 1906.2 °C·d and 1894.5 °C·d, respectively, and the total radiation was 2151.3 MJ / m 2 and 2116.9MJ / m 2 , the total precipitation was 103.9mm and 138.3mm respectively. Figure 1-2 ;

[0127] (2) Experimental design: 17 varieties were selected and sown on October 8, 2018 and October 23, 2019, and harvested on June 6, 2019 and June 2, 2020, respectively. At the same time, the data from 2018 to 2019 were used as the first experimental data of the present invention, and the data from 2019 to 2020 were used as the second experimental data of the present invention. The experiment was repeated twice using a randomized complete block design, and the sowing density was 270 plants / m 2, the row spacing is 10cm, the sowing depth is 5cm, and the planting area is 4.2m 2 (3m*1.4m), and the detailed water and fertilizer management in the two growing seasons is shown in Table 3. The number of ears, number of grains per ear, yield and thousand-grain weight were measured in the two growing seasons. From the filling to maturity of wheat, a 0.5m*0.5m sample frame was used to randomly sample twice in each planting area (except the marginal area), and the number of ears in the sample frame was recorded to obtain the average value; before the wheat matured, 20 representative ears of wheat were randomly selected from each planting area and threshed to obtain the average number of grains per ear; after the wheat matured, it was threshed, dried and weighed to obtain the yield per unit area, and then 1,000 representative grains were selected and weighed to obtain the thousand-grain weight; the prevention and control of pests and diseases and weeds during the growing season were carried out according to the standards of high-yield fields in the study area.

[0128] Table 3 Summary of water and fertilizer management in the 2018-2019 and 2019-2020 growing seasons

[0129]

[0130] (3) Material composition: This test selected 17 different types of test materials, covering various rib types (including strong ribs, medium-strong ribs, medium ribs, and weak ribs), different promotion years (from the 1980s to the 2010s), and multiple origins (at home and abroad). Among them, medium-rib type materials accounted for the highest proportion. The promotion of most materials was mainly concentrated in the 1980s to 2000s. These materials mainly came from five provinces in China, namely Anhui Province, Henan Province, Jiangsu Province, Shaanxi Province and Shandong Province, as well as Queensland, Australia abroad. These materials are rich in variety and widely distributed, with a reasonable genotype distribution. For details, please refer to Figure 3 ;

[0131] (4) Grain quality: Starch concentration and starch composition (amylose and amylopectin) were measured at maturity in 2019 and 2020. The total starch content, amylose content, and amylopectin content of wheat were measured using the starch detection kits Megazyme K-STAR and K-AMYL (Megazyme, Wicklow, Ireland, http: / / www.megazyme.com), respectively.

[0132] In the present invention, the model parameters that have direct and indirect effects on grain weight, yield, grain number and starch formation process are calibrated and evaluated, as shown in Table 4. These parameters include ecotype parameters, such as "eo_crop_factor_default", which reflects the characteristics of crop water demand; species type parameters, such as "node_sen_rate" indicating the node senescence rate, and "fr_lf_sen_rate" indicating the proportion of green leaf senescence corresponding to each senescent node; in addition, there is "y_rue", which reflects the efficiency of plant utilization of light energy during photosynthesis; "y_node_app_rate", which reflects the change in the number of nodes during plant growth; and "grains_per_gram_stem", "max_grain_size", "potential_grain_growth_rate" and "potential_grain_filling_rate" reflecting the size and number of grains after flowering, as well as the filling and growth rate of grains. The calibration and evaluation of the above parameters are all completed by batch parameter adjustment method. Through the calibration and evaluation of the parameters, the growth of wheat can be simulated more accurately.

[0133] Table 4 Variety and ecotype parameters related to grain yield (GY), thousand-grain weight (TGW) and number of kernels per spike (KNS) in the APSIM-Wheat model

[0134]

[0135] like Figure 4-5 As shown, the starch accumulation simulation model was calibrated and verified with the first test data (17 wheat varieties in 2018-2019). By comparing the simulated values ​​with the measured values, it can be seen that the grain weight, yield, number of grains and total starch content simulated by the starch accumulation simulation model of the present invention are highly consistent with the measured grain weight, yield, number of grains and total starch content. In addition, the model was used to evaluate the second test data (grain weight of 17 wheat varieties in 2019-2020), and the linear evaluation equations of grain weight, yield, number of grains and total starch content of wheat grains were fitted as follows: y=0.978x-0.31, y=1.27x-4005.05, y=1.087x-4242.83 and y=1.238x-8.36, respectively;

[0136] At the same time, the linear regression model was used to calibrate and evaluate the amylose content and amylopectin content with the first test data (17 wheat varieties in 2018-2019). The results of the simulated winter wheat amylose content and amylopectin content in 2018-2019 were as follows: Figure 6As shown, overall, the amylose content and amylopectin content simulated by the starch component simulation model based on Linear Regression are consistent with the actual amylose content and amylopectin content. The simulated values ​​of the amylose content and amylopectin content in 2019-2020 were compared with the measured values, and the linear evaluation equations of the amylose content and amylopectin content were fitted, which were y=0.31x+5.25 and y=0.43x+13.11, respectively, and the model was independently evaluated.

[0137] In the present invention, three statistical indices are used to evaluate the performance of the two models. The statistical indices used to measure the accuracy of the calibration and validation results of the starch component simulation model of wheat grains and the Linear Regression include the coefficient of determination (R 2 ), normalized root mean square error (RMSE) and Wilmot's consistency index D-index, expressed as follows:

[0138]

[0139]

[0140]

[0141] in: is the predicted value, y i is the measured value, is the average of the measured values, and n is the number of samples.

[0142] The evaluation results of the two simulation models using the above statistical indicators are shown in Tables 5 and 6.

[0143] Table 5 Evaluation results of simulation models for wheat grain weight, yield, grain number and starch content

[0144]

[0145]

[0146] Table 6 Evaluation results of starch component simulation model

[0147] Evaluation Items Amylose content Amylopectin content Fitted linear equation y=0.31x+5.25 y=0.43x+13.11 RMSE value 0.863 2.176 Dvalue 0.675 0.745 <![CDATA[R 2 Value]]> 0.6 0.61 p-value <0.05 <0.05

[0148] The evaluation results in Tables 5 and 6 above show that the calibrated starch simulation model of wheat grains can well predict the grain weight, yield, grain number and starch content of 17 varieties in the study area. At the same time, the starch component simulation model constructed based on LinearRegression can also accurately predict the starch component content of wheat grains of different varieties.

[0149] The above-described embodiments are only preferred implementations of the present application, which are used to illustrate the technical solutions of the present application, rather than to limit them. The protection scope of the present application is not limited thereto. Any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments or easily conceive of changes within the technical scope disclosed in the present application, or perform equivalent replacements for some of the technical features therein, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A method for simulating starch content and its component contents, characterized in that: The following steps are involved: S1: Construct a starch accumulation simulation model for wheat grains; S2: Screen the variables output by the starch accumulation simulation model, and establish a starch component simulation model of wheat grains based on Linear Regression with the screened variables whose coefficients are not zero; S3: The starch accumulation simulation model and the starch component simulation model were calibrated and verified using experimental data, the simulated values ​​obtained by the two models were compared with the measured values, and linear evaluation equations were fitted to evaluate the two simulation models; In step S1, the specific process of constructing the starch accumulation simulation model includes: S101: obtaining the starch accumulation rate of einkorn wheat; ; ; in: STR i For the i The starch accumulation rate of einkorn wheat per day, is the maximum rate of starch accumulation of individual grains, is the factor of starch synthesis capacity of grains on day i, is the temperature, Supply nitrogen to plants. The water status of the plant, For the i Available carbon per grain per day; S102: Determine the starch accumulation of the single grain wheat using the starch accumulation rate, the formula is as follows: ; in: GST 0 is the initial starch content of einkorn wheat, GST i and GST i-1 Respectively represent the first i Tianhedi i- The starch accumulation of einkorn wheat in one day, STR 1 is the starch accumulation rate of einkorn wheat on the first day; In step S101, It is related to the soluble carbohydrate content in wheat grains. It increases exponentially and then decreases linearly with the increase of accumulated temperature after flowering. Its expression is as follows: ; in: For the i The accumulated temperature after flowering, Indicates when and the cumulative temperature when the grain filling rate reaches the maximum; The expression is as follows: ; in: The maximum daily temperature, The lowest daily temperature, is the daily average temperature, The optimal temperature for wheat starch accumulation; The expression is as follows: ; in: is the critical nitrogen concentration for plants, is the minimum plant nitrogen concentration, is the actual dry nitrogen concentration of the plant; The expression is as follows: ; in: is the actual transpiration of the wheat plant, is potential transpiration; Determined by the immediate assimilates transported directly to the kernel and the remobilized assimilates from storage, it is expressed as follows: ; in: As immediate assimilates for grain utilization, To remobilize assimilates; Actual dry nitrogen concentration of plants The critical nitrogen concentration of plants is determined by the nitrogen absorption and total dry matter. Affected by the accumulated aboveground biomass, the expressions are as follows: ; ; in: is the N absorption, is the total dry matter mass, To accumulate aboveground biomass; Instant Assimilate Remobilized assimilates The expressions are: ; ; in: and For the flower queen i Tianhedi i -Total above-ground weight of wheat plants in 1 day, For the j The actual amount of dry matter accumulated on the ground, and For the flower queen i Tianhedi i -Total weight of vegetative organs in 1 day; ; in: and For the flower queen i Tianhedi i -Total weight of vegetative organs in 1 day; In step S2, when screening the variables output by the starch accumulation simulation model, the Lasso regression method is used to add an L1 regularization term to the loss function to screen the variables output by the starch accumulation model, and the variables whose coefficients are not zero are GB, TOPWT and VWT; In step S2, the steps of establishing the starch component simulation model of wheat grains are as follows: S201: Normalize GB, TOPWT and VWT. The formula is as follows: ; in: is the normalized variable, x is the measured value of the variable, and are the highest and lowest values ​​of the variable measured, respectively; S202: Based on Linear Regression, a starch component simulation model of wheat grains is established with normalized variables, and its expression is as follows: ; ; in: The content of amylose, is the amylopectin content, , , They are the normalized GB, TOPWT and VWT respectively.

2. The method for simulating starch content and its component contents according to claim 1, characterized in that: In step S3, the specific process of calibrating and verifying the starch accumulation simulation model and the starch component simulation model respectively using the test data is as follows: S301: calibrating and verifying the starch accumulation simulation model and the starch component simulation model using the first test data, and comparing the simulation values ​​with the measured values; S302: Evaluate the starch accumulation simulation model and the starch component simulation model using the second test data, and obtain a linear evaluation equation by fitting.

3. The method for simulating starch content and its component contents according to claim 2, characterized in that: In step S3, by determining the index R 2 , normalized root mean square error RMSE and Wilmot's consistency index D value The accuracy of the simulated values ​​of the two models was evaluated, including: ; ; ; in: is the predicted value, is the measured value, is the average of the measured values.