Analysis method for high-throughput evaluation of tobacco nicotine synthesis capability

Abstract

The invention provides an analysis method for high-throughput evaluation of tobacco nicotine synthesis capability, and belongs to the technical field of nicotine synthesis capability analys.The analysis method comprises the steps that tobacco seeds are disinfected, germinated and cultured for 14 days to obtain seedlings, the seedlings are transferred to a control medium and an induction medium added with 3 [mu] M of methyl jasmonate to be cultured for 15 days, and the seedlings are obtained; the method comprises the following steps: removing water from an overground part for 15 minutes at 105 DEG C, drying at 65 DEG C until the weight is constant, grinding the overground part into powder, adding a NaOH solution and ethyl acetate containing an internal standard, carrying out vortex mixing uniformly, carrying out ultrasonication for 15 minutes, taking supernate, filtering to obtain sample extract liquor, inputting peak area data into an intelligent prediction model based on a hierarchical weight mechanism, and processing to obtain prediction correction data, the nicotine concentration of a sample is calculated by combining gas chromatography detection with a standard curve, the concentration is corrected by utilizing predicted correction data, finally, nicotine content data is calculated according to the corrected concentration and the dry weight of the sample, and high-precision evaluation of the tobacco nicotine synthesis capability is realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of nicotine synthesis ability analysis, and in particular relates to a high-throughput analysis method for evaluating the nicotine synthesis ability of tobacco. Background Art

[0002] Nicotine content testing in tobacco, a key technical tool for tobacco breeding and quality evaluation, has traditionally relied on field cultivation combined with chemical analysis. Existing techniques typically use organic solvent extraction and high-performance liquid chromatography or gas chromatography to determine nicotine content in mature tobacco leaf samples. This method is widely used in tobacco variety selection, quality control, and genetic improvement research, providing fundamental testing technology support for the tobacco industry.

[0003] However, traditional detection methods suffer from significant precision deficiencies, primarily manifested in severe environmental interference, complex sample matrix effects, unstable extraction efficiency, and large instrument response variability. Field environmental factors such as temperature, humidity, light, and soil conditions significantly impact the biosynthesis and accumulation of nicotine in tobacco, leading to significant discrepancies in test results for the same variety under different environments. Furthermore, the complex matrix components of tobacco leaves can cause matrix interference, affecting the accuracy of chromatographic analysis and lacking an effective correction mechanism. This means that existing methods for assessing tobacco nicotine synthesis capacity suffer from low detection precision. Summary of the Invention

[0004] In view of this, the present invention provides a high-throughput analytical method for evaluating the nicotine synthesis capacity of tobacco, which can solve the technical problem of low detection accuracy of the existing methods for evaluating the nicotine synthesis capacity of tobacco.

[0005] The present invention is achieved as follows: the present invention provides a high-throughput analysis method for evaluating the nicotine synthesis capacity of tobacco, comprising the steps of disinfecting and germinating tobacco seeds, culturing them, inducing them with methyl jasmonate, drying and preparing samples, extracting and processing them, ultrasonically fragmenting and detecting them, and calculating the nicotine content. In the methyl jasmonate induction step, the germinated tobacco seedlings are transferred to an induction medium supplemented with methyl jasmonate for culturing, and methyl jasmonate is used to activate the defense response pathway in the tobacco plant to promote nicotine biosynthesis. In the sample extraction step, the tobacco powder sample is added with methyl jasmonate. The solution and ethyl acetate containing the internal standard are vortex-mixed; in the ultrasonic fragmentation detection step, the mixed sample is ultrasonically fragmented and then filtered to obtain a sample extract, and the peak area data is input into a nicotine synthesis intelligent prediction model to obtain prediction and correction data; the nicotine synthesis intelligent prediction model is a multi-layer perception network architecture based on a hierarchical weight mechanism, and the hierarchical weight parameters are dynamically adjusted according to the methyl jasmonate concentration data, the nutrient concentration classification, and the internal standard concentration classification; the sample nicotine concentration is calculated based on the peak area and the standard curve, and the tobacco nicotine content data is corrected in combination with the prediction and correction data.

[0006] Among them, the tobacco seed disinfection, germination and cultivation steps are specifically to disinfect the tobacco seeds with a 10% sodium hypochlorite solution containing 0.1% Tween for 7 minutes, rinse with sterile water 3 times, inoculate onto 1 / 2MS solid culture medium, and culture for 14 days at a temperature of 23°C, a light intensity of 5000 lux, and a photoperiod of 14 hours of light and 10 hours of darkness to obtain germinated tobacco seedlings.

[0007] Among them, the methyl jasmonate induction treatment step specifically involves transferring 6 of the germinated tobacco seedlings to a control medium and an induction medium, wherein the control medium is a 1 / 2 MS solid medium or an MS solid medium without the addition of methyl jasmonate, and the induction medium is a 1 / 2 MS solid medium or an MS solid medium with the addition of 3 μM methyl jasmonate, and the culture is continued for 15 days.

[0008] Among them, the methyl jasmonate is a nicotine synthesis inducer, and its concentration is divided into a low concentration range of 1μM to 2μM, a medium concentration range of 3μM to 5μM and a high concentration range of 6μM to 10μM. The medium concentration range has a significant activation effect on nicotine synthesis and the nicotine synthesis conversion rate is 300% to 400%. Controlling it within the medium concentration range reduces the risk of plant toxicity.

[0009] Among them, the tobacco sample drying preparation step specifically includes taking the above-ground part of the treated tobacco seedlings to obtain a tobacco sample, fixing the tobacco sample at 105°C in an oven for 15 minutes and then drying it at 65°C to constant weight to obtain a dry tobacco sample, weighing the dry weight of the dry tobacco sample to obtain the sample dry weight data, and grinding the dry tobacco sample into powder on a grinder to obtain a tobacco powder sample.

[0010] Among them, the 1 / 2MS solid culture medium is a solid culture medium prepared by reducing the inorganic salt concentration in the MS culture medium by half, containing 2.2g / L MS powder, 15g / L sucrose and 12g / L agar, and the pH is adjusted to 5.8 to 6.0. The nutrient concentration of the culture medium is divided into three nutrient concentration categories: low nutrient concentration 1 / 4MS, standard nutrient concentration 1 / 2MS and high nutrient concentration MS. The standard nutrient concentration has a suitable effect on the growth of tobacco seedlings.

[0011] The MS solid culture medium is a complete concentration of Murashige and Skoog culture medium, containing 4.4 g / L MS powder, 30 g / L sucrose and 12 g / L agar, and the pH is adjusted to 5.8 to 6.0.

[0012] The sample extraction step is to add 1 mL of 10% The solution and 5 mL of ethyl acetate containing internal standard were vortexed for 2 minutes on a vortex shaker to obtain a mixed sample.

[0013] Among them, the ultrasonic fragmentation and gas chromatography detection steps specifically include placing the mixed sample in an ultrasonic processor with a frequency of 300 Hz and ultrasonically fragmenting it for 15 minutes, taking 1 mL of the supernatant and passing it through a fat-soluble filter membrane with a pore size of 0.22 μm to obtain a sample extract, and placing the sample extract into a 2 mL gas chromatography automatic sampling bottle. At the same time, the peak area data of the sample extract is input into the nicotine synthesis intelligent prediction model for processing to obtain predicted correction data.

[0014] Among them, the nicotine synthesis intelligent prediction model is a multi-layer perception network architecture based on a hierarchical weight mechanism, which includes an input layer, multiple hidden layers and an output layer. The input layer receives peak area data, sample dry weight data and methyl jasmonate concentration data as feature vectors. The hidden layer uses a hierarchical attention mechanism to process feature information at different abstraction levels. The output layer generates prediction correction data and confidence assessment data.

[0015] The present invention significantly improves detection accuracy and the reproducibility of results by establishing a standardized in vitro culture induction system and an intelligent data processing model. By using methyl jasmonate to specifically activate the nicotine biosynthesis pathway at a controlled concentration, combined with standardized in vitro culture conditions using 1 / 2MS medium, the method effectively eliminates the interference of environmental variation on detection results, addressing the drawback of traditional methods that suffer from unstable precision due to environmental factors. Furthermore, an ethyl acetate extraction system containing an internal standard is employed, and matrix effects and extraction efficiency variation are corrected using 2,4-bipyridine as the internal standard, significantly improving the standardization of sample processing and the consistency of detection results. A multi-layer perception network intelligent prediction model based on a hierarchical weighting mechanism dynamically identifies and corrects systematic errors under different experimental conditions, achieving intelligent correction of detection data and optimizing precision, thereby addressing the technical issue of low detection precision in tobacco nicotine synthesis capacity assessment methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a flow chart of the method of the present invention.

[0017] Figure 2 This is a bar chart comparing the test results of the method of the present invention and field topping treatment on the nicotine content of tobacco.

[0018] Figure 3 This is a graph comparing tobacco growth between the method of the present invention and field topping treatment on the effect of methyl jasmonate on inducing nicotine synthesis in tobacco under different culture medium conditions.

[0019] Figure 4 This is a bar graph comparing the effects of the method of the present invention and field topping treatment on methyl jasmonate-induced nicotine synthesis in tobacco under different culture medium conditions.

[0020] Figure 5 This is a tobacco growth comparison diagram showing that the method of the present invention can accurately evaluate the nicotine synthesis capacity of different tobacco varieties.

[0021] Figure 6 The figure is a bar chart showing that the method of the present invention can accurately evaluate the nicotine synthesis capacity of different tobacco varieties.

[0022] Figure 7 A diagram showing the high-throughput evaluation of the nicotine synthesis capacity of different tobacco strains under methyl jasmonate induction using the method of the present invention.

[0023] Figure 8 This figure shows the high-throughput evaluation of the nicotine synthesis capacity of different tobacco strains after traditional topping treatment using the method of the present invention. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0025] like Figure 1 FIG. 1 is a flow chart of a high-throughput analysis method for evaluating the nicotine synthesis capacity of tobacco provided by the present invention. The method comprises the following steps: S01, sterilizing tobacco seeds with a 10% sodium hypochlorite solution containing 0.1% Tween for 7 minutes, rinsing with sterile water three times, inoculating the sterilized tobacco seeds onto 1 / 2 MS solid culture medium, and culturing the 1 / 2 MS solid culture medium with the tobacco seeds at a temperature of 23° C., a light intensity of 5000 lux, and a photoperiod of 14 hours light and 10 hours dark for 14 days to obtain germinated tobacco seedlings; S02, transferring 6 tobacco seedlings from the germinated tobacco seedlings to a control medium and an induction medium, respectively, wherein the control medium is 1 / 2 MS solid medium or MS solid medium without the addition of methyl jasmonate, and the induction medium is 1 / 2 MS solid medium or MS solid medium with the addition of 3 μM methyl jasmonate, and the transferred tobacco seedlings are further cultured under the culture conditions for 15 days to obtain treated tobacco seedlings; S03, taking the above-ground part of the treated tobacco seedlings to obtain a tobacco sample, fixing the tobacco sample in an oven at 105° C. for 15 minutes, and then drying at 65° C. to constant weight to obtain a dried tobacco sample; S04, weighing the dry weight of the dried tobacco sample to obtain the sample dry weight data, grinding the dried tobacco sample into powder on a grinder to obtain a tobacco powder sample, and adding 1 mL of 10% The solution and 5 mL of ethyl acetate containing internal standard were vortexed for 2 min on a vortex shaker to obtain a mixed sample; S05. Ultrasonicate the mixed sample in a 300 Hz ultrasonic processor for 15 minutes, filter 1 mL of the supernatant through a 0.22 μm pore size fat-soluble filter to obtain a sample extract, and place the sample extract into a 2 mL gas chromatography autosampler vial. Simultaneously, input the peak area data of the sample extract into a nicotine synthesis intelligent prediction model for processing to obtain prediction and correction data. S06. Prepare a nicotine standard solution to establish a standard curve, place the gas chromatography autosampler bottle containing the sample extract into the gas chromatograph injector, calculate the sample nicotine concentration based on the peak area of ​​the sample extract and the standard curve, and calibrate the sample nicotine concentration based on the predicted calibration data to obtain a calibrated sample nicotine concentration; S07. Based on the corrected sample nicotine concentration and the sample dry weight data, the tobacco nicotine content data is obtained according to the calculation formula: nicotine content equals the corrected sample nicotine concentration multiplied by 5 mL divided by the sample dry weight, thereby achieving high-throughput assessment of tobacco nicotine synthesis capacity.

[0026] The methyl jasmonate is a nicotine synthesis inducer that promotes nicotine biosynthesis by activating defense response pathways in tobacco plants. The concentration of the methyl jasmonate is divided into three concentration categories: a low concentration range of 1 μM to 2 μM, a medium concentration range of 3 μM to 5 μM, and a high concentration range of 6 μM to 10 μM. The low concentration range is obtained by diluting a standard methyl jasmonate solution, the medium concentration range is obtained by preparing a basic methyl jasmonate solution, and the high concentration range is obtained by concentrating a methyl jasmonate solution. The low concentration range has a slight activation effect on nicotine synthesis and a nicotine synthesis conversion rate of 150% to 200%. The medium concentration range has a significant activation effect on nicotine synthesis and a nicotine synthesis conversion rate of 300% to 400%. The high concentration range has an excessive activation effect on nicotine synthesis and a nicotine synthesis conversion rate of 500% to 600%, but produces negative phytotoxic effects. The risk of phytotoxicity is reduced by controlling the methyl jasmonate concentration within the medium concentration range. The 1 / 2MS solid culture medium is a solid culture medium prepared by reducing the inorganic salt concentration in the MS culture medium by half, containing 2.2g / L MS powder, 15g / L sucrose and 12g / L agar, and the pH is adjusted to 5.8 to 6.0. The nutrient concentration of the culture medium is divided into three nutrient concentration categories: low nutrient concentration 1 / 4MS, standard nutrient concentration 1 / 2MS and high nutrient concentration MS. The low nutrient concentration is obtained by diluting the MS powder concentration, the standard nutrient concentration is obtained by a conventional preparation method, and the high nutrient concentration is obtained by using full-concentration MS powder. The low nutrient concentration has a restrictive effect on the growth of tobacco seedlings and a growth delay rate of 20% to 30%. The standard nutrient concentration has a suitable effect on the growth of tobacco seedlings and a growth delay rate of 5% to 10%. The high nutrient concentration has an excessive nutritional effect on the growth of tobacco seedlings and a growth delay rate of 15% to 25%, but produces a negative effect of nutritional excess. The risk of nutritional imbalance is reduced by adopting the standard nutrient concentration. The MS solid medium is a complete concentration of Murashige and Skoog medium containing 4.4 g / L MS powder, 30 g / L sucrose and 12 g / L agar, and the pH is adjusted to 5.8 to 6.0; Wherein, the ethyl acetate containing internal standard is an extraction solvent in which 2,4-bipyridine is added to ethyl acetate as an internal standard substance, which is used to correct matrix effects and extraction efficiency in gas chromatography-mass spectrometry analysis. The concentration of the 2,4-bipyridine is divided into three internal standard concentration categories: low concentration internal standard 0.1 mg / mL to 0.3 mg / mL, standard concentration internal standard 0.5 mg / mL and high concentration internal standard 0.8 mg / mL to 1.0 mg / mL. The low concentration internal standard is obtained by diluting the standard internal standard solution, the standard concentration internal standard is obtained by directly preparing, and the high concentration internal standard is obtained by concentrating the standard internal standard solution. The low concentration internal standard has an insufficient correction effect on the analysis accuracy and a correction efficiency of 60% to 70%, the standard concentration internal standard has a sufficient correction effect on the analysis accuracy and a correction efficiency of 85% to 95%, and the high concentration internal standard has an excessive correction effect on the analysis accuracy and a correction efficiency of 70% to 80%, but produces a negative effect of matrix interference. By using the standard concentration internal standard, the risk of analysis error is reduced; The high-throughput assessment utilizes the high homogeneity and high seedling density characteristics of MS solid culture medium to complete the nicotine synthesis capacity assessment of a small breeding population through a single test, significantly improving the detection efficiency compared to traditional field testing methods. The detection throughput is divided into three detection scale categories: small-scale testing of 10 to 30 samples, medium-scale testing of 50 to 100 samples, and large-scale testing of 150 to 300 samples. The small-scale test is obtained through single-batch culture dish cultivation, the medium-scale test is obtained through multi-batch synchronous cultivation, and the large-scale test is obtained through an automated culture system. The small-scale test has a basic impact on resource utilization and a resource utilization rate of 40% to 60%. The medium-scale test has an optimizing impact on resource utilization and a resource utilization rate of 70% to 85%. The large-scale test has a maximized impact on resource utilization and a resource utilization rate of 90% to 95%, but it produces the negative effect of excessive equipment load. By rationally planning the test batches, the risk of equipment failure is reduced within the medium-scale detection range. The nicotine synthesis intelligent prediction model is a multi-layer perception network architecture based on a hierarchical weight mechanism, comprising an input layer, multiple hidden layers, and an output layer. The input layer receives the peak area data, the sample dry weight data, and the methyl jasmonate concentration data as feature vectors. The hidden layer uses a hierarchical attention mechanism to process feature information at different abstraction levels. The output layer generates the prediction correction data and confidence assessment data. The hierarchical weight parameters of the hierarchical weight mechanism are dynamically adjusted according to three key parameters: the methyl jasmonate concentration data, the nutrient concentration classification, and the internal standard concentration classification. When the methyl jasmonate concentration data is in the medium concentration range, the hierarchical weight parameters are tilted toward features related to nicotine synthesis. When the nutrient concentration classification is in the standard nutrient concentration, the hierarchical weight parameters are balanced toward features related to plant growth. When the internal standard concentration classification is in the standard concentration internal standard, the hierarchical weight parameters are concentrated toward features related to analysis accuracy. Among them, the step of establishing a training data set for the nicotine synthesis intelligent prediction model includes collecting peak area data of different tobacco varieties under various methyl jasmonate concentration conditions as input features, collecting corresponding measured nicotine content data as label data, and establishing a basic training set containing 3,000 tobacco sample data. The basic training set contains 250 samples of data from 12 different tobacco varieties, covering a gradient change of methyl jasmonate concentration from 1 μM to 10 μM, and the culture medium types include two preparation methods: the 1 / 2MS solid culture medium and the MS solid culture medium. The culture time is a time series from 10 days to 20 days. The basic training set is expanded to 10,000 sample data through data enhancement technology, and the data enhancement technology includes noise addition, time shift and concentration interpolation. A validation set is established containing 1,000 independent sample data for model performance evaluation, and a test set is established containing 500 new tobacco variety sample data for model generalization ability testing; The training step of the nicotine synthesis intelligent prediction model includes end-to-end training of the nicotine synthesis intelligent prediction model using a supervised learning method, using a mean square error loss function to measure the difference between the predicted nicotine content and the measured value, and using an Adam optimizer for gradient descent optimization. The learning rate is set to 0.001 and a learning rate decay strategy is adopted. The training process is divided into a pre-training stage and a fine-tuning stage. The pre-training stage uses all training data for 200 epochs of basic training. The fine-tuning stage performs fine adjustments on tobacco variety data for 50 epochs. An early stopping mechanism is used during training to prevent overfitting. Training is stopped when the validation set loss does not decrease for 10 consecutive epochs. Batch normalization and dropout techniques are used to improve the generalization ability of the model. Finally, the trained nicotine synthesis intelligent prediction model achieves a prediction accuracy of more than 95% on the test set. The concentration adaptation function is used to adjust the hierarchical weight parameters of the nicotine synthesis intelligent prediction model. The concentration adaptation function calculates a comprehensive concentration adaptation index based on three input data: the methyl jasmonate concentration data, the nutrient concentration classification data, and the internal standard concentration classification data. When the comprehensive concentration adaptation index falls within the low adaptation range of 0.2 to 0.4, a conservative weight adjustment function is used to reduce model sensitivity. When the comprehensive concentration adaptation index falls within the standard adaptation range of 0.5 to 0.7, a balanced weight adjustment function is used to maintain model stability. When the comprehensive concentration adaptation index falls within the high adaptation range of 0.8 to 1.0, an aggressive weight adjustment function is used to improve model responsiveness. The comprehensive concentration adaptation index is calculated using a weighted average method, wherein the weight of the methyl jasmonate concentration data is 0.5, the weight of the nutrient concentration classification data is 0.3, and the weight of the internal standard concentration classification data is 0.2. The concentration adaptation function implements adaptive adjustment of model parameters under different experimental conditions, ensuring the accuracy and stability of the predicted correction data.

[0027] The specific implementation of the above steps is described in detail below.

[0028] The specific implementation of step S01 involves surface disinfection of tobacco seeds, using chemical disinfection to remove microorganisms attached to the seed surface. First, a 10% sodium hypochlorite disinfectant solution containing 0.1% Tween-20 surfactant is prepared. The addition of Tween-20 reduces the solution's surface tension, improving the disinfectant's wettability and penetration of the seed surface. The tobacco seeds are completely immersed in the disinfectant solution and disinfected at room temperature for 7 minutes. The disinfection time is selected based on the killing kinetics of sodium hypochlorite against bacteria and fungi to ensure sufficient destruction of surface microorganisms while avoiding damage to the seed embryo. Immediately after disinfection, the seeds are rinsed with sterile water three times, each for 30 seconds, to remove residual disinfectant. The disinfected seeds are then inoculated onto the surface of a 1 / 2 MS solid culture medium at a density of 20 seeds per culture dish. The inoculated culture dishes are then placed in an artificial climate chamber with a temperature of 23°C, a light intensity of 5000 lux, and a photoperiod of 14 hours light and 10 hours dark, mimicking the optimal environmental conditions for tobacco seed germination. After 14 days of cultivation, germinated tobacco seedlings with well-developed root systems and unfolded cotyledons were obtained. The seedling height reached 2 to 3 cm. At this time, the seedlings had the physiological basis for subsequent treatment.

[0029] The specific implementation of step S02 involves establishing two culture conditions: a control treatment and an induction treatment. Using the principle of comparative experimental design, the induction effect of methyl jasmonate on nicotine biosynthesis is evaluated. Seedlings with consistent growth status are selected from the germinated tobacco seedlings obtained in step S01, with six seedlings selected for each treatment group to ensure statistical significance of the experimental results. The control culture medium uses either 1 / 2 MS solid medium or MS solid medium without methyl jasmonate to provide a reference baseline for basal levels of nicotine biosynthesis. The induction culture medium is based on the same culture medium supplemented with 3 micromolar methyl jasmonate, a concentration within the medium range that significantly activates the jasmonic acid signaling pathway in tobacco plant defense response pathways, thereby upregulating the expression of genes for key enzymes in nicotine biosynthesis. Methyl jasmonate, a fat-soluble plant hormone, regulates the expression of genes for key enzymes in the nicotine biosynthesis pathway, such as nicotine demethylase and ornithine decarboxylase, by activating transcription factors. The treated seedlings were further cultured under the same culture conditions for 15 days, during which the seedlings experienced a rapid growth period, nicotine synthesis metabolism was active, and finally treated tobacco seedlings with sufficient aboveground biomass were obtained.

[0030] Step S03 involves dehydrating and drying the tobacco sample using hot air drying to remove moisture from the plant tissue and terminate enzyme activity. First, the above-ground portions of the treated tobacco seedlings, including stem and leaf tissue, are collected, and the roots are removed to prevent interference with nicotine content determination by root secondary metabolites. The collected tobacco sample is placed in an oven and initially subjected to a high-temperature drying process at 105°C for 15 minutes. This high-temperature drying process rapidly inactivates enzymes in the plant tissue, preventing enzymatic degradation of nicotine during the subsequent drying process. This drying process, based on the principle of protein thermal denaturation, disrupts the tertiary structure of enzyme molecules, rendering them catalytically inactive. Following drying, the temperature is lowered to 65°C for constant-temperature drying, a temperature that effectively removes moisture without degrading heat-sensitive compounds like nicotine. The drying process continues until the sample weight remains constant, typically taking 12 to 24 hours. The constant weight requirement is a difference of less than 0.5 mg between two consecutive weighings. After drying, the moisture content of the tobacco sample is reduced to below 5%, facilitating subsequent grinding and extraction.

[0031] The specific implementation method of step S04 is to weigh and pre-treat the dry tobacco sample, and use solvent extraction to prepare the sample for chromatographic analysis. First, use an analytical balance with an accuracy of 0.1 mg to weigh the exact weight of the dry tobacco sample, and record the sample dry weight data for subsequent calculation of nicotine content. Place the dry sample in a high-speed grinder and grind it into powder. The grinding time is controlled within 30 seconds, and the sieve aperture is 80 mesh to ensure that the sample particle size is uniform and improve the extraction efficiency. The powdering process is based on the principle of surface area increase. Fine particles have a larger specific surface area, which is conducive to full contact between the solvent and the sample. Add 1 ml of 10% concentration to the ground tobacco powder sample. The sodium hydroxide solution is used to liberate bound nicotine. Nicotine often exists as a salt in plant tissues and needs to be converted to its free base form under alkaline conditions. Simultaneously, add 5 ml of ethyl acetate solution containing 0.5 mg / ml of 2,4-bipyridine as an internal standard. This internal standard is added based on the principle of quantitative internal standard analysis to correct for losses and matrix effects during the extraction process. Vortex the sample at 2000 rpm for 2 minutes to ensure thorough mixing of the sample and the extraction solvent, resulting in a homogeneous mixture.

[0032] The specific implementation of step S05 is to use ultrasound-assisted extraction technology to improve the extraction efficiency of nicotine, followed by sample purification. The mixed sample is placed in an ultrasonic processor with a frequency of 300 Hz and subjected to ultrasonic disruption for 15 minutes. The cavitation effect of ultrasound can destroy the plant cell wall structure, promoting the release of nicotine in the cells, while mechanical vibration accelerates the diffusion and mass transfer process of the solute into the solvent. After the ultrasonic treatment is completed, the sample is allowed to stand for 5 minutes to allow stratification. 1 ml of the upper clear ethyl acetate phase containing the extracted nicotine and internal standard substance is collected. The extract is filtered and purified through a fat-soluble polytetrafluoroethylene filter membrane with a pore size of 0.22 microns to remove suspended particles and impurities, obtaining a clear and transparent sample extract. The filtered sample extract is placed in a 2 ml gas chromatograph automatic sampling bottle and sealed with a cap to prevent evaporation loss. At the same time, the pre-processing information of the sample extract, including peak area data, sample dry weight data and methyl jasmonate concentration data, was input into the nicotine synthesis intelligent prediction model for data processing. The model used a machine learning algorithm to analyze the nonlinear relationship between the sample characteristic parameters and generate prediction correction data for subsequent concentration correction.

[0033] The specific implementation of step S06 involves establishing a calibration curve and performing gas chromatography quantitative analysis, employing a dual quantitative strategy combining an external standard method with an internal standard method. First, a nicotine standard solution is prepared with a concentration gradient of 0.1, 0.5, 1.0, 2.0, 5.0, and 10.0 μg / mL. Three replicates are prepared for each concentration point. The standard solution is also spiked with the same concentration of 2,4-bipyridine as the internal standard. Gas chromatography analyzes the standard solutions in ascending order of concentration, establishing a linear regression equation between nicotine concentration and peak area. The correlation coefficient is required to be greater than 0.999. The autosampler containing the sample extract is then placed in the gas chromatograph injector. The gas chromatography analysis conditions are set at an inlet temperature of 250°C, a detector temperature of 280°C, a nitrogen carrier gas flow rate of 1 mL / min, and an injection volume of 1 μL. A nonpolar capillary column is used, and the column temperature is programmed from 100°C to 280°C at a rate of 10°C / min. The initial sample nicotine concentration is calculated based on the retention time and peak area of ​​the nicotine peak in the sample extract, combined with the standard curve. Mathematical calculations are performed on the initial concentration data and the predicted correction data generated by the intelligent prediction model to correct for matrix effects and extraction efficiency, ultimately obtaining an accurate corrected sample nicotine concentration.

[0034] The specific implementation of step S07 is to calculate the actual nicotine content in the tobacco based on the mass balance principle to achieve a quantitative assessment of nicotine synthesis capacity. The input parameters include the corrected sample nicotine concentration and the sample dry weight data. The calculation principle is based on the dilution law and the concentration conversion relationship. Because 5 ml of ethyl acetate solvent was used in the extraction process, the measured solution concentration needs to be converted into nicotine content per unit dry weight of the sample. The calculation formula is: nicotine content equals the corrected sample nicotine concentration multiplied by the extraction solvent volume of 5 ml, divided by the sample dry weight, with units of micrograms per milligram dry weight or milligrams per gram dry weight. This calculation obtains the absolute nicotine content data in the tobacco sample, which reflects the nicotine synthesis capacity of the tobacco plant under specific culture conditions. The nicotine content data of the control group and the induced group are compared and analyzed to calculate the synergistic effect of the induction treatment and evaluate the promoting effect of methyl jasmonate on nicotine synthesis. The entire detection process, from sample preparation to result output, can be completed in 48 hours, achieving high-throughput and rapid assessment of the nicotine synthesis capacity of tobacco.

[0035] The intelligent nicotine synthesis prediction model utilizes a deep neural network architecture based on an attention mechanism. The model structure comprises a hierarchical input layer, multiple hidden layers, and an output layer. The input layer is designed as a multidimensional feature vector receiver, receiving three key input data types: peak area data, sample dry weight data, and methyl jasmonate concentration data. The input layer has 64 neurons, each corresponding to a normalized feature value. The hidden layer utilizes a multilayer perceptron architecture, comprising four hidden layers with 128, 256, 128, and 64 neurons, respectively. Each layer is fully connected, and the activation function is a rectified linear unit function to prevent the vanishing gradient problem. The key hierarchical attention mechanism is embedded between the second and third hidden layers. This mechanism uses a self-attention algorithm to calculate the correlation weights between different features. Eight attention heads are used, each responsible for capturing different types of feature interaction patterns. The output layer consists of two branches: a primary output branch that generates prediction correction data, and an auxiliary output branch that generates confidence assessment data. A linear activation function is used to ensure the continuity of the numerical output.

[0036] The hierarchical weighting mechanism is implemented using a dynamic weight adjustment algorithm, which adjusts network weights in real time based on three key input parameters: methyl jasmonate concentration data, nutrient concentration classification, and internal standard concentration classification. The weight adjustment function utilizes a piecewise linear interpolation algorithm. When the methyl jasmonate concentration is in the medium range of 3 to 5 micromolar, the weight parameters shift toward features related to nicotine synthesis, with the weight coefficient increasing by a factor of 0.3. When the nutrient concentration is 1 / 2 MS of the standard nutrient concentration, the weight parameters remain balanced toward features related to plant growth, with all features equally weighted. When the internal standard concentration reaches the standard concentration of 0.5 mg / mL, the weight parameters shift toward features related to analytical accuracy, with the relevant weight coefficient increasing by a factor of 0.2. The concentration adaptation function calculates a comprehensive concentration adaptation index based on a weighted average algorithm. The weight of the methyl jasmonate concentration data is 0.5, the weight of the nutrient concentration classification data is 0.3, and the weight of the internal standard concentration classification data is 0.2. This index drives the corresponding weight adjustment functions.

[0037] The detailed steps for establishing the training dataset began with basic data collection. Twelve representative tobacco varieties, including flue-cured, sun-cured, and mixed varieties, were selected. Each variety was cultured and samples collected under varying methyl jasmonate concentrations. A concentration gradient was set at 10 concentration points: 1, 2, 3, 4, 5, 6, 7, 8, 9, and 10 micromolar. Twenty-five replicates were collected for each variety at each concentration point, creating a basic training set containing 3,000 sample data. Culture media were prepared using both 1 / 2 MS and MS formulations, and incubation times were set at five time points: 10, 12, 15, 18, and 20 days, forming a multidimensional experimental design matrix. Data augmentation techniques employed three methods: Gaussian noise addition, time series shifting, and linear interpolation of concentrations. The noise standard deviation was set to 5% of the original data, the time shift range was ±2 days, and concentration interpolation was performed linearly between adjacent concentration points. Data augmentation was used to expand the basic training set to 10,000 sample data points. The validation set contains 1,000 independent samples from the same varieties but different batches as the training set. This is used to monitor model performance and optimize hyperparameters during training. The test set contains 500 new tobacco varieties, never seen during training, to evaluate the model's generalization and practical application.

[0038] The model training adopts an end-to-end supervised learning method. The loss function selects the mean square error function to measure the deviation between the predicted value and the true value. The optimizer uses the adaptive moment estimation algorithm for gradient descent optimization. The initial value of the learning rate is set to 0.001, and the cosine annealing learning rate decay strategy is adopted. The learning rate decays to 0.1 times the original value every 50 training cycles. The training process is divided into two stages: pre-training and fine-tuning. The pre-training stage uses all 10,000 training samples for basic training of 200 training cycles. The fine-tuning stage performs fine adjustments for 50 cycles based on the specificity of different tobacco varieties. In order to prevent overfitting, an early stopping mechanism is adopted. When the validation set loss does not decrease for 10 consecutive cycles, the training is automatically stopped. The model regularization technology includes batch normalization and random dropout. The batch normalization parameters are set to momentum 0.9 and epsilon 1× The random dropout probability was set to 0.3. The trained model achieved a prediction accuracy of 95.2% on an independent test set, with a root mean square error of 0.08 μg / mg, meeting the accuracy requirements for practical applications.

[0039] It should be noted that the core technical ideas of the present invention are mainly reflected in the following three aspects: The first core technical idea is the nicotine synthesis regulation technology using methyl jasmonate induction combined with culture medium optimization. Traditional tobacco nicotine detection methods usually rely on the basal nicotine content in the plant under natural conditions. The test results are often affected by environmental factors and fluctuate greatly, and cannot accurately reflect the plant's nicotine synthesis potential. The present invention introduces methyl jasmonate as a specific inducer, which can activate the defense response pathway in the tobacco plant, significantly enhance the activity of the nicotine biosynthesis enzyme system, and thus fully release the potential nicotine synthesis capacity. At the same time, combined with the graded regulation of nutrient concentration of 1 / 2MS and MS culture media, it provides the most suitable nutritional environment for tobacco seedlings at different growth stages, avoids the interference of nutritional imbalance on the nicotine synthesis process, and makes the test results more accurately reflect the genetic differences between varieties.

[0040] The second core technical approach is the intelligent prediction model technology based on a hierarchical weighting mechanism. While traditional chemical analysis methods can accurately determine nicotine content, they often overlook the impact of factors such as matrix effects during sample processing, differences in extraction efficiency, and instrument system errors on the results. The multi-layer perception network architecture constructed in this invention comprehensively considers multidimensional feature information such as peak area data, sample dry weight data, and methyl jasmonate concentration data. It automatically identifies feature patterns at different levels of abstraction through a hierarchical attention mechanism and dynamically adjusts weight parameters based on changes in experimental conditions. This intelligent data processing method effectively eliminates the subjectivity of human judgment in traditional methods and significantly improves the reliability and reproducibility of analytical results.

[0041] The precision analysis technology of high-throughput detection and internal standard correction represents the third core technical idea. Traditional field determination methods are not only long and costly, but also difficult to achieve simultaneous detection of large-scale samples, which seriously restricts the efficiency of breeding work. The present invention utilizes the high homogeneity characteristics of MS solid culture medium, combined with standardized culture condition control, to simultaneously process a large number of samples under a unified experimental environment, fundamentally solving the impact of environmental variation on test results. At the same time, the ethyl acetate extraction system using 2,4-bipyridine as the internal standard substance can correct the systematic error of each sample in the extraction and detection process in real time, ensuring the comparability of results between different batches.

[0042] The synergistic effect of these three core technical approaches yields significant combined advantages. Methyl jasmonate induction technology provides high-quality input data for the intelligent prediction model, enabling it to make accurate predictions based on a fully activated state of nicotine synthesis. The intelligent prediction model provides precise data correction for high-throughput testing, eliminating systematic biases that may arise during batch processing. High-throughput testing, in turn, provides a large amount of standardized data for model training, continuously improving prediction accuracy. This positive feedback mechanism between the technologies ensures that the entire analytical method combines the accuracy of traditional chemical analysis with the efficiency of modern biotechnology, providing strong technical support for tobacco breeding.

[0043] Specifically, the principle behind this invention is that the core principle behind its technical solution, which addresses the issue of low detection accuracy, lies in the construction of a three-tiered technical architecture: standardized induction, standardized extraction, and intelligent correction. First, methyl jasmonate, a key signaling molecule in plant defense responses, can specifically activate the jasmonic acid signal transduction pathway in tobacco plants. This, in turn, promotes the activation of the nicotine biosynthesis pathway by upregulating the expression of key enzyme genes such as nicotine synthase and putrescine N-methyltransferase. At an optimal concentration of 3 μM, methyl jasmonate significantly activates nicotine biosynthesis, achieving a conversion rate of 300-400% while avoiding the phytotoxic effects of high concentrations, ensuring the standardization and controllability of the induction process.

[0044] Secondly, the standardized in vitro culture system eliminates sources of environmental variation by precisely controlling culture conditions. The 1 / 2MS culture medium provides a constant nutritional environment for tobacco seedlings. The culture conditions of 23°C, 5000 lux light intensity, and a 14-hour photoperiod ensure comparability between different samples. Testing at the seedling stage avoids interference from complex environmental factors during field cultivation. Sample pretreatment at 105°C for 15 minutes effectively terminates enzyme activity. The standardized drying process of drying to constant weight at 65°C ensures consistent sample moisture content, effectively controlling key variables that affect test accuracy.

[0045] Crucially, the intelligent prediction model based on a hierarchical weighting mechanism achieved a fundamental improvement in detection accuracy through a deep learning algorithm. The model employs a multi-layer perceptron network architecture. The input layer receives peak area data, sample dry weight data, and methyl jasmonate concentration data as feature vectors, while the hidden layer uses a hierarchical attention mechanism to process feature information at different levels of abstraction. The hierarchical weight parameters are dynamically adjusted based on three key parameters: methyl jasmonate concentration, nutrient concentration classification, and internal standard concentration classification. When each parameter is within its optimal range, the model automatically optimizes the weight distribution to maximize prediction accuracy. The concentration adaptation function automatically selects a conservative, balanced, or aggressive weight adjustment strategy by calculating a comprehensive concentration adaptation index to ensure the model's stability under different experimental conditions. Using a large-scale training dataset consisting of 3,000 basic samples and expanded to 10,000 samples, the model learned complex nonlinear mapping relationships, accurately identifying and correcting systematic errors, and achieving a prediction accuracy of over 95%.

[0046] The logical rationality of this technical solution lies in the organic integration of biological induction mechanisms and intelligent algorithms. Methyl jasmonate induction conforms to the biological laws governing plant secondary metabolism, standardized culture conditions eliminate random interference, and the intelligent prediction model leverages machine learning to uncover implicit patterns in the data and accurately calibrate them. These three technical components work together to form a complete, high-precision detection technology system.

[0047] A specific embodiment 1 of the present invention is provided below. The specific implementation of each step in this embodiment 1 is described in detail as follows.

[0048] The specific implementation of steps S01-S05 is the same as above and will not be repeated here.

[0049] In step S06, the calculation process of nicotine concentration is described in detail as follows. The standard curve is established using linear regression analysis method, and the linear equation is specifically expressed as: ; Where, is the nicotine peak area; is the nicotine concentration in μg / mL; is the slope of the regression line; is the intercept of the regression line.

[0050] The parameter acquisition method is: Obtained by gas chromatography detection; It is obtained by preparing standard solutions with a concentration range of 0.1 to 10.0 μg / mL; and Obtained through least squares regression calculation.

[0051] The calculation formula of the sample nicotine concentration is specifically expressed as: ; Where, is the sample nicotine concentration, in μg / mL; is the nicotine peak area in the sample.

[0052] The internal standard correction calculation process adopts the peak area ratio method, and the correction formula is specifically expressed as follows: ; Where, is the nicotine concentration of the sample after internal standard correction, in μg / mL; is the peak area of ​​the internal standard in the sample; is the internal standard peak area in the standard solution; is the internal standard concentration in the standard solution, in mg / mL; is the internal standard concentration in the sample, in mg / mL.

[0053] The correction process of the intelligent prediction model adopts the multiplicative correction factor method, and the concentration calculation formula after correction is specifically expressed as: ; Where, is the final corrected sample nicotine concentration, in μg / mL; is the prediction correction factor, dimensionless.

[0054] In step S07, the calculation process of nicotine content is described in detail as follows. According to the mass balance principle and the concentration dilution law, the calculation formula of tobacco nicotine content is specifically expressed as: ; Where, is the nicotine content of tobacco, in μg / mg or mg / g; is the extraction solvent volume, which is 5 mL; is the dry weight of the sample in mg.

[0055] The induction effect evaluation adopts the relative synergistic multiple calculation method, and the synergistic multiple calculation formula is specifically expressed as: ; Where, It is the synergistic factor of induced treatment, dimensionless; is the nicotine content of the induced treatment group, in μg / mg; is the nicotine content of the control treatment group, in μg / mg.

[0056] The calculation process of the concentration adaptation function in the intelligent prediction model is described in detail as follows. The comprehensive concentration adaptation index is calculated using the weighted average method, and the calculation formula is specifically expressed as follows: ; Where, is the comprehensive concentration adaptation index, ranging from 0 to 1; is the methyl jasmonate concentration index, dimensionless; is the nutrient concentration index, dimensionless; is the internal standard concentration index, dimensionless; is the concentration weight of methyl jasmonate, which is 0.5; is the nutrient concentration weight, which is 0.3; is the internal standard concentration weight, which is set to 0.2.

[0057] The parameter acquisition method is: The data were normalized by methyl jasmonate concentration, and the calculation method was to divide the actual concentration value by the maximum concentration value of 10 μM; The nutrient concentration classification data was numerically obtained, with the low nutrient concentration value being 0.25, the standard nutrient concentration value being 0.5, and the high nutrient concentration value being 1.0; The internal standard concentration was obtained by numerically classifying the data: the low concentration internal standard value was 0.2, the standard concentration internal standard value was 0.5, and the high concentration internal standard value was 0.9.

[0058] The weight adjustment function adopts a piecewise function form according to the range of the comprehensive concentration adaptation index, which is specifically expressed as follows: ; Where, is the weight adjustment factor, dimensionless. When the adaptation range is 0.2 to 0.4, conservative weight adjustment is adopted; when When the standard adaptation range is 0.5 to 0.7, balance weight adjustment is used; when Aggressive weight adjustment is used when the value is in the high adaptability range of 0.8 to 1.0.

[0059] The prediction correction factor is calculated by multiplying the output value of the neural network and the weight adjustment factor. The calculation formula is specifically expressed as: ; Where, is the original output value of the neural network, dimensionless; It is the model bias correction item, and its value range is -0.05 to 0.05.

[0060] The high-throughput detection efficiency evaluation adopts the sample processing rate calculation method. The detection flux calculation formula is specifically expressed as: ; Where, is the detection throughput, the unit is the number of samples / hour; is the total number of samples; is the total processing time in hours.

[0061] The resource utilization rate is calculated using the equipment load assessment method, and the calculation formula is as follows: ; Where, is the resource utilization rate, in percentage; is the actual working time, in hours; The theoretical maximum working time, in hours.

[0062] The detection accuracy is evaluated by the relative standard deviation calculation method. The accuracy calculation formula is specifically expressed as: ; Where, is the relative standard deviation, expressed in percentage; is the standard deviation; is the average value.

[0063] In the evaluation of the induction effect of methyl jasmonate, the calculation process of the nicotine synthesis conversion rate is described in detail below. The formula for calculating the induction effect in the low concentration range is specifically expressed as: ; Where, The conversion rate of nicotine synthesis in the low concentration range is 150% to 200%. is the nicotine content of the low-concentration treatment group, in μg / mg; is the nicotine content of the control group, in μg / mg.

[0064] The calculation formula for the inductive effect in the medium concentration range is specifically expressed as: ; Where, The conversion rate of nicotine synthesis in the medium concentration range is 300% to 400%. is the nicotine content of the medium concentration treatment group, in μg / mg.

[0065] The calculation formula for the induction effect in the high concentration range is specifically expressed as: ; Where, The conversion rate of nicotine synthesis in the high concentration range is 500% to 600%. is the nicotine content of the high concentration treatment group, in μg / mg. The parameter acquisition method is: The method is obtained through a low-concentration methyl jasmonate treatment experiment, comprising step 1: culturing tobacco seedlings in a culture medium supplemented with 1-2 μM methyl jasmonate; step 2: collecting the above-ground parts after culturing for 15 days; and step 3: determining the nicotine content according to the process of steps S03 to S07. The method is obtained by treating tobacco seedlings with medium-concentration methyl jasmonate, including step 1: culturing tobacco seedlings in a culture medium supplemented with 3-5 μM methyl jasmonate; step 2: collecting samples after 15 days of cultivation; and step 3: determining the nicotine content according to the same process. The method is obtained by a high-concentration methyl jasmonate treatment experiment, comprising the steps of: 1: culturing tobacco seedlings in a culture medium supplemented with 6-10 μM methyl jasmonate; 2: collecting samples after the same culturing time; and 3: determining the nicotine content using the same method.

[0066] In the evaluation of the effect of nutrient concentration on tobacco seedling growth, the calculation process of the growth delay rate is described in detail below. The formula for calculating the growth delay rate at low nutrient concentration is specifically expressed as: ; Where, is the growth delay rate at low nutrient concentration, ranging from 20% to 30%; is the growth time of the low nutrient concentration treatment group, in days; is the growth time of the control group with standard nutrient concentration, in days.

[0067] The calculation formula for the growth delay rate at high nutrient concentration is specifically expressed as: ; Where, is the growth delay rate at high nutrient concentration, ranging from 15% to 25%; is the growth time of the high nutrient concentration treatment group, in days. The parameter acquisition method is: It is obtained through experimental observation, including step 1: inoculating tobacco seedlings into a low nutrient concentration 1 / 4 MS culture medium; step 2: cultivating under standard culture conditions until the seedlings reach a preset growth standard; step 3: recording the number of days required from inoculation to reaching the standard. The method is obtained through a control experiment, including step 1: inoculating the same batch of tobacco seedlings into a standard nutrient concentration 1 / 2MS culture medium; step 2: cultivating under the same culture conditions to the same growth standard; and step 3: recording the number of days required for the growth of the control group. It is obtained through experimental observation, including step 1: inoculating tobacco seedlings into high nutrient concentration MS culture medium; step 2: culturing to a preset standard under standard culture conditions; step 3: recording the growth days of the high nutrient concentration treatment group.

[0068] In the evaluation of the effect of internal standard concentration on analytical accuracy, the calculation process of correction efficiency is described in detail below. The calculation formula for low concentration internal standard correction efficiency is specifically expressed as: ; Where, is the low concentration internal standard correction efficiency, ranging from 60% to 70%; This is the accuracy after correction for low-concentration internal standards; The theoretical maximum accuracy.

[0069] The calculation formula of internal standard correction efficiency of standard concentration is specifically expressed as: ; Where, is the internal standard correction efficiency of the standard concentration, ranging from 85% to 95%; The accuracy is after correction for the internal standard concentration.

[0070] The calculation formula for high concentration internal standard correction efficiency is specifically expressed as: ; Where, is the high concentration internal standard correction efficiency, ranging from 70% to 80%; is the accuracy after high concentration internal standard correction. The parameter acquisition method is: The method is obtained by a low-concentration internal standard calibration experiment, including step 1: preparing an extraction solvent containing 0.1 to 0.3 mg / mL of the internal standard; step 2: performing sample processing and detection according to the process of steps S04 to S06; and step 3: calculating the degree of conformity between the measured value after calibration and the true value. The results were obtained by standard concentration internal standard calibration experiment, including step 1: preparing a standard extraction solvent containing 0.5 mg / mL internal standard; step 2: using the same processing and detection process; step 3: evaluating the accuracy level of the standard internal standard calibration. The method is obtained by a high-concentration internal standard calibration experiment, including step 1: preparing an extraction solvent containing 0.8 to 1.0 mg / mL of the internal standard; step 2: processing and testing the sample according to the standard process; and step 3: calculating the accuracy index of the high-concentration internal standard calibration. It is the theoretical maximum accuracy, and the value is 100%, which represents an ideal detection state of complete accuracy.

[0071] To better understand and implement the present invention, Example 2, a specific application scenario, is provided below. Researchers used the method presented here to evaluate the nicotine synthesis capacity of different tobacco varieties and hybrid strains, validating its effectiveness and practicality. Four representative tobacco varieties, NC95, MAFC5, MLAF34, and LAFC53, were selected for study. High-throughput screening was also performed using F5 populations generated by hybridization between NC95 and LAFC53.

[0072] Material preparation and culture condition setting The researchers first prepared the culture medium. The method for preparing the MS solid culture medium was to add 4.4g of MS powder, 30g of sucrose, and 12g of agar per liter of culture medium. Adjust the pH to 5.8-6.0 and sterilize at 121°C for 15 minutes. Prepare 1 / 2 MS solid medium by adding 2.2 g MS powder, 15 g sucrose, and 12 g agar per liter of medium. The pH adjustment and sterilization conditions are the same. The stock concentration of methyl jasmonate is set to 3 mM, and the working concentration is 3 μM.

[0073] Tobacco seeds were disinfected with 10% Soak in the solution for 7 minutes, rinse three times with sterile water, and then inoculate onto the surface of 1 / 2 MS solid culture medium. Culture conditions were set at 23°C, 5000 lux of light intensity, and a photoperiod of 14 hours of light and 10 hours of darkness. Germinated seedlings were obtained after 14 days of culture.

[0074] Induction treatment and sample preparation After 14 days, the researchers transferred well-germinated tobacco seedlings to control and induction media, with six seedlings per treatment. The control media consisted of MS or 1 / 2 MS solid medium without methyl jasmonate, while the induction media consisted of the corresponding medium supplemented with 3 μM methyl jasmonate. After 15 days of culture under the same conditions, the aerial parts of the seedlings were collected for sample preparation.

[0075] The sample was dried using a graded drying method. First, the sample was fixed in an oven at 105°C for 15 minutes, then cooled to 65°C and dried to constant weight. After accurately weighing the dry sample, it was ground into 80 mesh powder using a high-speed grinder. 1 mL of 10% concentration of The solution and 5 mL of ethyl acetate solution containing internal standard were vortexed for 2 minutes and then ultrasonically extracted.

[0076] Nicotine extraction and detection analysis Ultrasonic extraction conditions were set at a frequency of 300 Hz and a processing time of 15 minutes. After extraction, 1 mL of the supernatant was filtered through a 0.22 μm lipid-soluble filter membrane and transferred into a 2 mL gas chromatography autosampler vial for analysis. A nicotine standard solution was simultaneously prepared with a concentration gradient of 5, 10, 20, 30, 40, and 50 μg / mL. A standard curve was established for quantitative analysis.

[0077] Gas chromatography detection conditions were an inlet temperature of 250°C, a detector temperature of 280°C, and nitrogen carrier gas at a flow rate of 1 mL / min. An HP-5 capillary column was used, and the column temperature program was initially 100°C for 2 minutes, then increased at 10°C / min to 280°C for 5 minutes. Calibration was performed using 2,4-bipyridine as the internal standard at a concentration of 0.5 mg / mL.

[0078] Evaluation of nicotine-induced effects under different culture medium conditions The researchers first verified the consistency of the method of the present invention with the traditional field topping treatment, corresponding Figure 2-4 In the field experiment, nicotine content was measured on the fifth leaf from the bottom to the top of tobacco leaves at the bud stage. The nicotine content was 11.4 μg / mg before topping, and increased to 18.2 μg / mg two weeks after topping, an increase of 59.6%. Figure 2 The significant changes in nicotine content before and after topping were clearly shown, verifying the promoting effect of topping on nicotine synthesis.

[0079] The comparative experiment was carried out using the method of the present invention, as shown in Table 1: Table 1 Effect of methyl jasmonate on nicotine synthesis in tobacco under different culture medium conditions

[0080] Experimental results showed that in MS solid medium, methyl jasmonate treatment increased nicotine content from 3.7 μg / mg to 11.9 μg / mg, a 221.6% increase relative to the control. In 1 / 2 MS solid medium, induction treatment increased nicotine content from 6.7 μg / mg to 23.9 μg / mg, a 256.7% increase. Both media effectively activated the nicotine biosynthesis pathway in tobacco, with the 1 / 2 MS medium exhibiting a more pronounced induction effect.

[0081] Analysis of differences in nicotine synthesis capacity among different tobacco varieties To verify the accuracy of the method of the present invention in detecting differences between varieties, the researchers selected four tobacco varieties with different nicotine content characteristics for comparative analysis. Figure 5-6 research content. Figure 6The nicotine content trends of the four strains (NC95, MAFC5, MLAF34, and LAFC53) under control and methyl jasmonate treatment conditions are visually displayed, clearly demonstrating the differences and consistency in nicotine synthesis capacity among the strains. The experiment used the same culture and treatment conditions, and the results are shown in Table 2: Table 2 Comparative analysis of nicotine synthesis capacity of different tobacco varieties

[0082] The test results showed that the nicotine content of the four varieties showed a gradient distribution of NC95>MAFC5>MLAF34>LAFC53, which was consistent with the results of field material determination, fully verifying the Figure 5-6 The accuracy of the nicotine synthesis capacity ranking among tobacco varieties demonstrated here is significant. NC95, a high-nicotine variety, achieved an induction level of 28.6 μg / mg, while LAFC53, a low-nicotine variety, achieved only 9.7 μg / mg. The methyl jasmonate-induced conversion rates for each variety ranged from 334.5% to 356.5%, indicating that the response mechanisms of different tobacco genotypes to methyl jasmonate are generally similar. Figure 6 The histogram clearly shows that the nicotine content ranking of the four varieties remains consistent regardless of whether they are treated with methyl jasmonate, which proves the reliability of the method of the present invention in evaluating the differences between varieties.

[0083] High-throughput screening and validation of hybrid populations The researchers used the F5 stable population obtained by hybridizing NC95 (female parent) and LAFC53 (male parent) to conduct high-throughput screening experiments. This part of the work corresponds to Figure 7-8 research content. Figure 7-8 The method demonstrated the ability to rapidly screen a large number of hybrid strains, with a histogram clearly showing the distribution of nicotine content variation among the different strains. Sixty of the 120 strains were randomly selected for preliminary screening, and 30 representative strains were ultimately selected for detailed analysis. The screening results are shown in Table 3: Table 3 Nicotine content distribution characteristics of F5 hybrid population

[0084] High-throughput screening results showed that nicotine content in the F5 generation showed a continuous distribution, with the NL55 strain reaching the highest nicotine content of 27.3 μg / mg, close to the level of its high-nicotine parent, NC95. The NL36 strain had the lowest content, 12.6 μg / mg, slightly higher than its low-nicotine parent, LAFC53. The NL279 strain, with a content of 18.7 μg / mg, was in the middle range, demonstrating the segregation characteristics of the hybrid progeny. Figure 7-8The data distribution diagram in the figure clearly demonstrates the characteristics of this continuous variation, proves the genetic diversity of nicotine synthesis ability in the hybrid population, and verifies the practicality and efficiency of the method of the present invention in large-scale population screening.

[0085] Field validation and method accuracy assessment To verify the reliability of the laboratory screening results, researchers planted 30 representative strains in a test field and measured nicotine content using the traditional topping method. The field verification results are shown in Table 4: Table 4 Comparison of laboratory screening and field validation results

[0086] Field validation results showed a good correlation between laboratory screening and field measurements, with an overall correlation coefficient of 0.88. Among high-nicotine strains, NL55 had a nicotine content of 36.5 μg / mg under field conditions, confirming it as the strain with the highest nicotine content. Among low-nicotine strains, NL36 had a field content of 20.2 μg / mg, the lowest. The average nicotine content of medium-nicotine strains was 27.7 μg / mg, falling between high- and low-nicotine strains, validating the accuracy of laboratory screening results.

[0087] Detection efficiency and time cost analysis The method of this invention takes only 48 hours from sample preparation to results, while traditional field methods require a full growing season (120-150 days). In terms of sample processing capacity, a single batch can process 96 samples simultaneously, while field methods are limited by the planted area and typically do not exceed 20 samples per test. In terms of detection accuracy, the relative standard deviation of the method of this invention is 3.2%, while the relative standard deviation of the field method, due to environmental factors, reaches 8.5%.

[0088] Traditional assessment of tobacco nicotine synthesis capacity mainly relies on biological methods of field planting and topping treatment. This method requires a complete tobacco growth cycle and is affected by multiple factors such as climatic conditions, soil environment, pests and diseases. The detection cycle is long, the cost is high, and the throughput is low. The present invention realizes the rapid assessment of tobacco nicotine synthesis capacity through the method of methyl jasmonate induction combined with in vitro culture. Compared with traditional methods, the present invention shortens the detection time by 75%, increases the sample processing throughput by 380%, and improves the detection accuracy by 62%, providing an efficient technical means for tobacco breeding and quality evaluation. This method not only maintains a high degree of consistency with field results, but also significantly improves detection efficiency and accuracy, providing important technical support for tobacco variety improvement and germplasm resource evaluation.

[0089] The above description is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed by the present invention, which should be covered by the scope of protection of the present invention.

Claims

1. A high-throughput analytical method for evaluating tobacco nicotine synthesis capacity, characterized in that: The method includes the steps of disinfecting and germinating tobacco seeds and culturing them, inducing treatment with methyl jasmonate, drying and preparing samples, extracting and processing samples, ultrasonic fragmentation and detection, and calculating nicotine content. In the methyl jasmonate induction treatment step, the germinated tobacco seedlings are transferred to an induction culture medium supplemented with methyl jasmonate for culturing, and methyl jasmonate is used to activate the defense response pathway in the tobacco plant to promote nicotine biosynthesis. In the sample extraction and processing step, a solution and ethyl acetate containing an internal standard are added to the tobacco powder sample and vortex-mixed. In the ultrasonic fragmentation and detection step, the mixed sample is ultrasonically fragmented and then filtered to obtain a sample extract, and the peak area data is input into a nicotine synthesis intelligent prediction model to obtain prediction correction data. The nicotine synthesis intelligent prediction model is a multi-layer perception network architecture based on a hierarchical weight mechanism, and the hierarchical weight parameters are dynamically adjusted according to the methyl jasmonate concentration data, the nutrient concentration classification, and the internal standard concentration classification. The nicotine concentration of the sample is calculated according to the peak area and the standard curve, and the tobacco nicotine content data is obtained by correction based on the prediction correction data.

2. The method according to claim 1, characterized in that The tobacco seed disinfection, germination and cultivation steps are as follows: the tobacco seeds are disinfected with a 10% sodium hypochlorite solution containing 0.1% Tween for 7 minutes, rinsed with sterile water three times, inoculated onto a 1 / 2MS solid culture medium, and cultured for 14 days at a temperature of 23°C, a light intensity of 5000 lux, and a photoperiod of 14 hours light and 10 hours dark to obtain germinated tobacco seedlings.

3. The method according to claim 2, characterized in that The methyl jasmonate induction treatment step specifically comprises transferring 6 of the germinated tobacco seedlings to a control medium and an induction medium, wherein the control medium is a 1 / 2 MS solid medium or an MS solid medium without adding methyl jasmonate, and the induction medium is a 1 / 2 MS solid medium or an MS solid medium with adding 3 μM methyl jasmonate, and continuing to culture for 15 days.

4. The method according to claim 3, characterized in that The methyl jasmonate is a nicotine synthesis inducer, and its concentration is divided into a low concentration range of 1μM to 2μM, a medium concentration range of 3μM to 5μM, and a high concentration range of 6μM to 10μM. The medium concentration range has a significant activation effect on nicotine synthesis and the nicotine synthesis conversion rate is 300% to 400%. Controlling within the medium concentration range reduces the risk of phytotoxicity.

5. The method according to claim 4, characterized in that The tobacco sample drying preparation step specifically comprises taking the above-ground part of the treated tobacco seedlings to obtain a tobacco sample, fixing the tobacco sample at 105° C. in an oven for 15 minutes, and then drying the tobacco sample at 65° C. to constant weight to obtain a dried tobacco sample, weighing the dry weight of the dried tobacco sample to obtain the sample dry weight data, and grinding the dried tobacco sample into powder on a grinder to obtain a tobacco powder sample.

6. The method according to claim 5, characterized in that The 1 / 2MS solid culture medium is a solid culture medium prepared by reducing the inorganic salt concentration in the MS culture medium by half, containing 2.2g / L MS powder, 15g / L sucrose and 12g / L agar, and the pH is adjusted to 5.8 to 6.

0. The nutrient concentration of the culture medium is divided into three nutrient concentration categories: low nutrient concentration 1 / 4MS, standard nutrient concentration 1 / 2MS and high nutrient concentration MS. The standard nutrient concentration has a suitable effect on the growth of tobacco seedlings.

7. The method according to claim 6, characterized in that The MS solid medium is a complete concentration Murashige and Skoog medium containing 4.4 g / L MS powder, 30 g / L sucrose and 12 g / L agar, with the pH adjusted to 5.8 to 6.

0.

8. The method according to claim 7, characterized in that The sample extraction and processing steps are as follows: adding 1 mL of a 10% solution and 5 mL of ethyl acetate containing an internal standard to the tobacco powder sample, and vortexing and mixing for 2 minutes using a vortex oscillator to obtain a mixed sample.

9. The method according to claim 8, characterized in that The ultrasonic fragmentation and gas chromatography detection steps specifically include placing the mixed sample in an ultrasonic processor at a frequency of 300 Hz and ultrasonically fragmenting it for 15 minutes, taking 1 mL of the supernatant and passing it through a fat-soluble filter membrane with a pore size of 0.22 μm to obtain a sample extract, and placing the sample extract into a 2 mL gas chromatography automatic sampling vial. At the same time, the peak area data of the sample extract is input into the nicotine synthesis intelligent prediction model for processing to obtain predicted correction data.

10. The method according to claim 9, characterized in that The nicotine synthesis intelligent prediction model is a multi-layer perception network architecture based on a hierarchical weight mechanism, comprising an input layer, multiple hidden layers, and an output layer. The input layer receives peak area data, sample dry weight data, and methyl jasmonate concentration data as feature vectors. The hidden layer uses a hierarchical attention mechanism to process feature information at different abstract levels. The output layer generates prediction correction data and confidence assessment data. The ethyl acetate containing internal standard is an extraction solvent in which 2,4-bipyridine is added to ethyl acetate as an internal standard substance. The concentration of 2,4-bipyridine is divided into a low concentration internal standard of 0.1 mg / mL to 0.3 mg / mL, a standard concentration internal standard of 0.5 mg / mL, and a high concentration internal standard of 0.8 mg / mL to 1.0 mg / mL. g / mL three internal standard concentration classifications, standard concentration internal standard has a sufficient correction effect on the analysis accuracy and the correction efficiency is 85% to 95%; among them, the hierarchical weight parameters of the hierarchical weight mechanism are dynamically adjusted according to the three key parameters of methyl jasmonate concentration data, nutrient concentration classification and internal standard concentration classification. When the methyl jasmonate concentration data is in the medium concentration range, the hierarchical weight parameters are inclined to the nicotine synthesis related characteristics. When the nutrient concentration classification is in the standard nutrient concentration, the hierarchical weight parameters are balanced to the plant growth related characteristics. Among them, the training data set of the nicotine synthesis intelligent prediction model includes collecting the peak area data of different tobacco varieties under various methyl jasmonate concentration conditions as input features, and collecting the corresponding measured nicotine The nicotine content data is used as label data to establish a basic training set containing 3,000 tobacco sample data, and the basic training set is expanded to 10,000 sample data through data enhancement technology; the nicotine content calculation step is to prepare nicotine standard solution to establish a standard curve, put the gas chromatography automatic injection bottle containing the sample extract into the gas chromatograph injector, calculate the sample nicotine concentration according to the peak area of ​​the sample extract and the standard curve, and calibrate the sample nicotine concentration with the predicted calibration data to obtain the calibrated sample nicotine concentration; the high-throughput evaluation step is to calculate the nicotine content of the sample according to the calibrated sample nicotine concentration and sample dry weight data, and calculate the nicotine content of the sample according to the nicotine content equal to the calibrated sample nicotine concentration. The nicotine content data for tobacco was calculated by multiplying the nicotine concentration by 5 mL divided by the dry weight of the sample. Taking advantage of the high homogeneity and high seedling density of MS solid culture medium, the nicotine synthesis capacity of a small breeding population was assessed in a single test. A concentration adaptation function was used to adjust the hierarchical weight parameters of the intelligent prediction model for nicotine synthesis. The concentration adaptation function calculated a comprehensive concentration adaptation index based on three input data: methyl jasmonate concentration data, nutrient concentration classification data, and internal standard concentration classification data. The comprehensive concentration adaptation index was calculated using a weighted average method, with a weight of 0.5 for methyl jasmonate concentration data, 0.3 for nutrient concentration classification data, and 0.2 for internal standard concentration classification data.

Images