Evaluation method for morphological characteristics of new variety of gene-edited juvenile soybean

Abstract

The invention discloses a method for evaluating morphological characteristics of a new variety of gene-edited juvenile soybean, which comprises the following steps: collecting ecological element data of a soybean planting area, and establishing a standardized environment database; analyzing an environmental factor periodic change rule of the standardized environmental database, and establishing an environmental dynamic change model; controlling environment parameters of the phytotron according to the environment dynamic change model, and simulating different environment conditions including target area environment conditions; planting gene editing soybeans of different genotypes, and collecting phenotype data of the soybeans of different genotypes under different environmental conditions; extracting environmental condition data of the phytotron and phenotype data of the gene editing soybean, and constructing a multiple regression model; and predicting soybean phenotypic characters under other environmental conditions through the multiple regression model, and obtaining suitable growth conditions of gene editing soybeans of different genotypes according to prediction results. According to the method, a breeder can be helped to make a more reasonable planting decision according to the genotype and environmental adaptability of soybeans.

Description

Technical Field

[0001] The present invention belongs to the technical field of soybean planting, and in particular relates to a method for evaluating the morphological characteristics of a new gene-edited long-juvenile soybean variety. Background Art

[0002] In the process of using artificial climate chambers to evaluate the phenotypic characteristics of new gene-edited long-juvenile soybean varieties, how to accurately simulate the complex and changeable natural environment of the target planting area is a key technical challenge. Although artificial climate chambers can control basic environmental factors such as light, temperature and humidity, these factors are not constant in the actual natural environment, but show dynamic characteristics such as circadian rhythms, seasonal changes and random fluctuations. How to accurately simulate these dynamic changes and couple them with the growth and development process of gene-edited soybeans poses a challenge to the accurate evaluation of phenotypic traits.

[0003] In addition, soil conditions in different planting areas, such as soil texture, fertility, and pH value, also vary significantly. These factors will directly affect the root development and nutrient absorption of soybeans, and thus affect the growth and yield of the aboveground parts. How to effectively simulate the soil characteristics of different regions in an artificial climate chamber to avoid soil factors from interfering with the phenotypic evaluation results is also an urgent problem to be solved. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a method for evaluating the morphological characteristics of new gene-edited long-juvenile soybean varieties to solve the problems existing in the above-mentioned prior art.

[0005] To achieve the above object, the present invention provides a method for evaluating the morphological characteristics of a new gene-edited long-juvenile soybean variety, comprising the following steps:

[0006] Collect ecological factor data in soybean-growing areas and establish a standardized environmental database;

[0007] Analyze the periodic change rules of environmental factors in the standardized environmental database and establish an environmental dynamic change model;

[0008] Controlling environmental parameters of the artificial climate chamber according to the environmental dynamic change model to simulate different environmental conditions including those of the target area;

[0009] Plant gene-edited soybeans of different genotypes and collect phenotypic data of soybeans of different genotypes under different environmental conditions;

[0010] Extract the environmental condition data of the artificial climate chamber and the phenotypic data of gene-edited soybeans, and construct a multivariate regression model;

[0011] The multivariate regression model is used to predict soybean phenotypic traits under other environmental conditions, and the suitable growth conditions for gene-edited soybeans of different genotypes are obtained based on the prediction results.

[0012] Preferably, the ecological element data includes meteorological element data and soil parameter data;

[0013] Meteorological element data include hourly temperature, light intensity, rainfall and air humidity;

[0014] Soil parameter data include soil temperature and moisture, soil texture, pH value and organic matter content.

[0015] Preferably, the ecological factor data is preprocessed before establishing the standardized environmental database, and the preprocessing includes:

[0016] Outlier detection and missing value processing are performed on ecological factor data, and missing data are filled by interpolation method to ensure data integrity and continuity; meteorological factor data and soil parameter data are temporally and spatially matched according to the data's timestamp and site coordinates to generate hourly, daily, and monthly spatiotemporal distribution data sets of environmental factors; data standardization methods are used to normalize environmental factor data and establish a standardized environmental database.

[0017] Preferably, the establishing of the environment dynamic change model comprises:

[0018] The periodic variation law of environmental factors is analyzed by Fourier transform method to obtain frequency components and amplitude information, random fluctuations of environmental factors are simulated by random functions, and an environmental dynamic change model is established based on the frequency components, amplitude information and random fluctuations.

[0019] Preferably, the multivariate regression model comprises a random forest regression model.

[0020] Preferably, the method further comprises:

[0021] The differences in sensitivity to environmental factors at different growth stages were analyzed through the gradient boosting decision tree. When a growth stage was more sensitive to a specific environmental factor, the environmental factor parameters were controlled more accurately at that growth stage.

[0022] The present invention also provides a computer device, comprising a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method.

[0023] The present invention provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the method are implemented.

[0024] Compared with the prior art, the present invention has the following advantages and technical effects:

[0025] The present invention discloses a method for evaluating the morphological characteristics of a new variety of gene-edited long-juvenile soybeans. The method first constructs an environmental database of the target planting area, including data on ecological factors such as temperature and light. Then, Fourier transform and other methods are used to analyze the periodic changes of environmental factors and establish a mathematical model that describes the dynamic changes of the environment. According to the model, environmental parameters are precisely controlled in an artificial climate chamber to simulate the climate characteristics of the target area. At the same time, different types of soil matrices are configured to simulate the soil characteristics of the target area. Gene-edited soybeans are planted in a simulated environment, and phenotypic data are collected regularly. A multivariate regression model of environmental parameters and soybean phenotypes is constructed using a machine learning method to predict phenotypic traits under different environmental conditions. In addition, the differences in sensitivity to environmental factors at different growth stages are analyzed, and key environmental factors are more accurately controlled in sensitive stages. The present invention realizes the accurate evaluation of the phenotypic traits of gene-edited soybeans under specific environments, providing important technical support for soybean variety breeding. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0027] Figure 1 The figure is a flow chart of a method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0028] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0029] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0030] Embodiment 1

[0031] like Figure 1 As shown, this embodiment provides a method for evaluating the morphological characteristics of a new gene-edited long-juvenile soybean variety, comprising the following steps:

[0032] Collect ecological factor data in soybean-growing areas and establish a standardized environmental database;

[0033] Analyze the periodic change rules of environmental factors in the standardized environmental database and establish an environmental dynamic change model;

[0034] Controlling environmental parameters of the artificial climate chamber according to the environmental dynamic change model to simulate different environmental conditions including those of the target area;

[0035] Plant gene-edited soybeans of different genotypes and collect phenotypic data of soybeans of different genotypes under different environmental conditions;

[0036] Extract the environmental condition data of the artificial climate chamber and the phenotypic data of gene-edited soybeans, and construct a multivariate regression model;

[0037] Predicting soybean phenotypic traits under other environmental conditions by using the multivariate regression model, and obtaining suitable growth conditions for gene-edited soybeans of different genotypes based on the prediction results;

[0038] The differences in sensitivity to environmental factors at different growth stages were analyzed through the gradient boosting decision tree. When a growth stage was more sensitive to a specific environmental factor, the environmental factor parameters were controlled more accurately at that growth stage.

[0039] Furthermore, establishing a standardized environmental database includes: constructing an environmental database for the target planting area, which includes data on ecological factors in the area throughout the year, such as hourly temperature, light intensity, rainfall, air humidity, as well as soil temperature and humidity, soil texture, pH value, organic matter content, etc., cleaning and preprocessing the collected data, removing outliers and missing values, and establishing a standardized environmental database.

[0040] According to the geographical location of the target area and the distribution of meteorological stations, the meteorological stations and soil monitoring points where environmental data needs to be collected are determined, and the latitude and longitude coordinate information of the stations is obtained. Through the automatic observation system of the meteorological station, meteorological element data such as hourly temperature, light intensity, rainfall data and air humidity are collected, and the data is transmitted to the environmental database. Soil temperature and humidity sensors and soil moisture sensors are deployed in the target area to collect soil temperature and soil moisture data in real time, and the data is uploaded to the environmental database through the wireless transmission module. The soil texture, pH value and organic quality of the target area are sampled and analyzed to obtain soil physical and chemical property parameters and enter them into the environmental database. Outlier detection and missing value processing are performed on the meteorological element data and soil parameter data in the environmental database, and the missing data are filled by interpolation method to ensure the integrity and continuity of the data. According to the timestamp and station coordinates of the data, the meteorological element data and soil parameter data are matched in time and space to generate hourly, daily and monthly environmental element spatiotemporal distribution data sets. The environmental element data are normalized by data standardization method to eliminate the dimensional differences of different element data and establish a standardized environmental database.

[0041] Furthermore, establishing a dynamic environmental change model includes: constructing a time series data set for each environmental factor. Performing Fourier transform on the time series data of each environmental factor, converting the time domain signal into a frequency domain signal, and obtaining the amplitude and phase information of each frequency component. According to the Fourier transform results, the main frequency components of the environmental factor changes are extracted, usually including daily frequency (corresponding to circadian rhythm) and annual frequency (corresponding to seasonal changes) and their frequency multiples. For each extracted main frequency component, a mathematical model in the form of a sine function is established, wherein the amplitude, frequency and phase parameters of the sine function are determined by the Fourier transform results of the corresponding frequency components. Combining the sine function models corresponding to each main frequency component, a mathematical model that can describe the periodic change law of environmental factors (including circadian rhythm and seasonal changes) is constructed. On the basis of the periodic change mathematical model, a random function is introduced to simulate the random fluctuation of environmental factors. The random function can use Gaussian white noise or other suitable random process models. Combining the periodic change model with the random fluctuation model, a comprehensive mathematical model that can fully reflect the dynamic change characteristics of environmental factors in the target planting area is obtained.

[0042] Furthermore, simulating different environmental conditions including the environmental conditions of the target area includes: obtaining data on changes in parameters such as light intensity, spectrum, temperature and humidity over time in the model according to a pre-established mathematical model of the environmental parameters of the target area; inputting the obtained light intensity and spectrum data into the LED light source control module, dynamically controlling the intensity and spectrum distribution of the LED light source by adjusting the power supply current and PWM signal of the LED lamp to simulate the day and night and seasonal light changes in the target area; inputting the obtained temperature change data into the temperature control system, adjusting the working state of the refrigeration and heating equipment through the PID algorithm, and dynamically controlling the temperature in the artificial climate room to simulate the day and night and seasonal temperature changes in the target area; inputting the obtained humidity change data into the temperature control system, dynamically controlling the temperature in the artificial climate room by adjusting the working state of the refrigeration and heating equipment through the PID algorithm, and dynamically controlling the temperature in the artificial climate room to simulate the day and night and seasonal temperature changes in the target area; and inputting the obtained humidity change data into the temperature control system. The humidification and dehumidification control module dynamically controls the air humidity in the artificial climate room by adjusting the working status of the humidifier and dehumidifier to simulate the humidity changes in the target area; in the process of environmental parameter control, the light, temperature and humidity sensor data in the artificial climate room are collected in real time, compared with the target value of the mathematical model, and the deviation value is calculated; if the deviation value exceeds the preset threshold range, the deviation value is input into the corresponding control module, and the working parameters of the LED light source, temperature control system and humidification and dehumidification equipment are adjusted through the feedback control algorithm to reduce the deviation between the actual environmental parameters and the target value; the above-mentioned environmental parameter collection, comparison and feedback control process is repeated until the parameters such as light, temperature and humidity in the artificial climate room are stabilized within the target value range.

[0043] According to the soil texture characteristics of the target planting area, determine the mixing ratio of sand, loam and clay, and simulate the soil texture similar to that of the target area by adjusting the ratio of the three. According to the soil fertility level of the target planting area, select the appropriate fertilizer type and amount, and add a certain amount of organic matter to adjust the fertility of the matrix to a level similar to that of the target area. Obtain the pH value data of the soil in the target planting area, and adjust the pH value of the matrix to a range similar to that of the soil in the target area by adding acidic or alkaline regulators. Mix the sand, loam and clay evenly according to the determined ratio, add the selected fertilizer and organic matter, and stir them thoroughly to make them evenly distributed. During the mixing process, continuously measure the pH value of the matrix, and dynamically adjust the amount of acid-base regulator according to the measurement results until the pH value stabilizes within the target range. Use the soil texture triangle diagram to determine the texture type of the mixed matrix. If it does not match the soil type of the target area, return to readjust the ratio of sand, loam and clay until the target texture is reached. The physical and chemical properties of the configured substrate are measured, and based on the measurement results, it is determined whether it meets the soil characteristic requirements of the target planting area. If not, return to the corresponding step to adjust the parameters according to the deviation until the requirements are met.

[0044] Furthermore, collecting phenotypic data of soybeans of different genotypes under different environmental conditions includes: obtaining gene-edited long-juvenile soybean seeds, selecting suitable substrates for sowing, and placing them in an artificial climate chamber for cultivation. According to the environmental conditions of the target area, the corresponding temperature, humidity, light and other parameters are set in the artificial climate chamber to simulate the real environment for soybean growth. During the growth of soybeans, image recognition technology is used to regularly take pictures of soybeans to obtain phenotypic data such as plant height and leaf area. Computer vision algorithms are used to segment and extract features from the captured soybean images to obtain phenotypic parameters such as stem thickness and root morphology. Biomass sensors are used to measure the fresh weight and dry weight of soybean plants at different growth periods to obtain soybean biomass data. The collected phenotypic data and biomass data are integrated to establish a phenotypic database of soybeans of different genotypes. Using machine learning algorithms, the phenotypic data of soybeans of different genotypes are compared and analyzed to explore the association between genotype and phenotype, providing data support for soybean variety improvement.

[0045] Furthermore, the suitable growth conditions for gene-edited soybeans of different genotypes are obtained based on the prediction results, including: selecting a suitable machine learning multivariate regression model, such as support vector regression, random forest regression or multilayer perceptron regression; dividing the preprocessed environmental parameter data and phenotypic trait data into a training set and a test set, training the selected machine learning model with the training set data, and evaluating the model performance with the test set data; using the trained machine learning model, inputting environmental parameter data under different environmental conditions, and predicting the phenotypic trait data of gene-edited soybeans.

[0046] Furthermore, the method also includes: obtaining plant phenotypic data and corresponding environmental factor parameters at different growth stages to construct a training data set. The training data is modeled using a random forest algorithm, and the sensitivity of each growth stage to different environmental factors is determined through feature importance analysis. According to the results of the sensitivity analysis, more accurate parameter control thresholds are set for growth stages and environmental factors with higher sensitivity. During plant growth, environmental factor parameters are obtained in real time to determine the current growth stage. If the current growth stage is highly sensitive to a certain environmental factor, the control parameters of the environmental factor are dynamically adjusted according to a preset threshold. Plant phenotypic data is continuously obtained, and the control threshold of the environmental factor parameter is dynamically updated through the deviation from the model prediction value. The model and parameter control strategy are continuously iterated and optimized to improve the overall accuracy of plant phenotypic evaluation and growth regulation.

[0047] In this embodiment, plant phenotypic data and corresponding environmental factor parameters at different growth stages are obtained to construct a training data set. It is necessary to collect corresponding phenotypic data at different growth stages of the plant, such as the seedling stage, the vegetative growth stage, the flowering stage and the pod-setting stage. For example, in the seedling stage, plant height, leaf number and leaf area can be measured; in the vegetative growth stage, stem thickness, branch number and leaf area index can be measured; in the flowering stage, inflorescence size, flowering number and flowering duration can be measured; in the pod-setting stage, fruit number, fruit size and fruit weight can be measured. At the same time, environmental factor parameters corresponding to each growth stage, such as temperature, humidity, light intensity and carbon dioxide concentration, are recorded. These data will constitute a training data set. For example, data can be collected every other day for one month, and finally a data set containing multiple growth stages and corresponding environmental parameters is obtained. The training data is modeled using a random forest algorithm. Random forest is an ensemble learning algorithm that improves the accuracy and stability of the model by constructing multiple decision trees and combining their prediction results. Since there may be complex nonlinear relationships between environmental factors and phenotypic data, random forests can effectively capture these relationships. Using the collected training dataset, a random forest model can be trained to predict the phenotypic traits of plants under different environmental conditions. For example, the random forest algorithm can be implemented using the scikit-learn library in Python. Through feature importance analysis, the sensitivity of each growth stage to different environmental factors is determined. The random forest algorithm can provide a measure of feature importance, which can help understand which environmental factors have the greatest impact on the phenotypic traits of plants. For example, the degree of influence of temperature and humidity on plant height during the seedling stage can be analyzed; the degree of influence of light intensity and carbon dioxide concentration on the number of flowers during the flowering stage can be analyzed. Through feature importance analysis, it can be determined which environmental factors are critical to the growth and development of plants at which growth stages. Suppose that it is found that temperature has the greatest impact on plant height during the seedling stage, while light intensity has the greatest impact on the number of flowers during the flowering stage. According to the results of the sensitivity analysis, more precise parameter control thresholds are set for growth stages and environmental factors with higher sensitivity. After determining the sensitivity of plants to environmental factors at different growth stages, more precise parameter control thresholds can be set for growth stages and environmental factors with higher sensitivity. For example, if the seedling stage is very sensitive to temperature, the temperature can be controlled within the optimal range, such as 25°C ± 1°C. If the flowering stage is very sensitive to light intensity, the light intensity can be controlled within a more precise range according to the needs of the plant, such as 1000 μmol / m 2 / s±100μmol / m 2 / s. During the plant growth process, the environmental factor parameters are obtained in real time to determine the current growth stage. It is necessary to monitor the environmental parameters in the artificial climate room in real time, such as temperature, humidity, light intensity and carbon dioxide concentration. At the same time, according to the morphological characteristics of the plant, the current growth stage of the plant is determined, such as the seedling stage, the vegetative growth stage, the flowering stage or the pod stage. For example, the growth stage of the plant can be automatically identified by image recognition technology and used as the model input. If the current growth stage is highly sensitive to a certain environmental factor, the control parameter of the environmental factor is dynamically adjusted according to the preset threshold. If the current growth stage is highly sensitive to a certain environmental factor, such as the seedling stage is sensitive to temperature, the control parameter of the environmental factor is dynamically adjusted according to the preset threshold. For example, if the temperature monitored in real time exceeds the preset threshold of 26°C, the cooling device is automatically started to reduce the temperature to a suitable range. Continuously obtain plant phenotypic data, and dynamically update the control threshold of the environmental factor parameter by the deviation from the model prediction value. During the plant growth process, it is necessary to continuously obtain the phenotypic data of the plant. Compare the real-time acquired phenotypic data with the model prediction value to calculate the deviation. According to the size of the deviation, the control threshold of the environmental factor parameter is dynamically updated. For example, if the plant height predicted by the model is much lower than the actual measured plant height, it means that the current environmental conditions are not conducive to plant growth and environmental parameters need to be adjusted, such as increasing the temperature or light intensity. Continuously iterate and optimize the model and parameter control strategy to improve the overall accuracy of plant phenotypic evaluation and growth regulation. As data accumulates, the model and parameter control strategy can be continuously iterated and optimized. For example, the model can be retrained with new data, or the parameters of the model can be adjusted. At the same time, the control thresholds of environmental parameters are continuously adjusted according to the actual growth conditions. Through continuous iterative optimization, the overall accuracy of plant phenotypic evaluation and growth regulation can be improved. For example, after iterative optimization of several growth cycles, the prediction accuracy of the model can be improved, and the growth of plants in artificial climate chambers can be made more stable and efficient.

[0048] This embodiment further provides a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method.

[0049] This embodiment also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the method are implemented.

[0050] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for evaluating the morphological characteristics of a new gene-edited long-juvenile soybean variety, characterized in that: The following steps are involved: Collect ecological factor data in soybean-growing areas and establish a standardized environmental database; Analyze the periodic change rules of environmental factors in the standardized environmental database and establish an environmental dynamic change model; Controlling environmental parameters of the artificial climate chamber according to the environmental dynamic change model to simulate different environmental conditions including those of the target area; Plant gene-edited soybeans of different genotypes and collect phenotypic data of soybeans of different genotypes under different environmental conditions; Extract the environmental condition data of the artificial climate chamber and the phenotypic data of gene-edited soybeans, and construct a multivariate regression model; The multivariate regression model is used to predict soybean phenotypic traits under other environmental conditions, and the suitable growth conditions for gene-edited soybeans of different genotypes are obtained based on the prediction results.

2. The method according to claim 1, characterized in that The ecological element data include meteorological element data and soil parameter data; Meteorological element data include hourly temperature, light intensity, rainfall and air humidity; Soil parameter data include soil temperature and moisture, soil texture, pH value and organic matter content.

3. The method according to claim 1, characterized in that Before establishing the standardized environmental database, the ecological factor data is preprocessed, and the preprocessing includes: Outlier detection and missing value processing are performed on ecological factor data, and missing data are filled by interpolation method to ensure data integrity and continuity; meteorological factor data and soil parameter data are temporally and spatially matched according to the data's timestamp and site coordinates to generate hourly, daily, and monthly spatiotemporal distribution data sets of environmental factors; data standardization methods are used to normalize environmental factor data and establish a standardized environmental database.

4. The method according to claim 1, characterized in that The establishment of the environment dynamic change model comprises: The periodic variation law of environmental factors is analyzed by Fourier transform method to obtain frequency components and amplitude information, random fluctuations of environmental factors are simulated by random functions, and an environmental dynamic change model is established based on the frequency components, amplitude information and random fluctuations.

5. The method according to claim 1, characterized in that The multiple regression model includes a random forest regression model.

6. The method according to claim 1, characterized in that The method further comprises: The differences in sensitivity to environmental factors at different growth stages were analyzed through the gradient boosting decision tree. When a growth stage was more sensitive to a specific environmental factor, the environmental factor parameters were controlled more accurately at that growth stage.

7. A computer device comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 6.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images