A gordon euryale breeding database construction system based on big data collection

Abstract

The present application relates to the field of agricultural information technology, and more particularly to a gordon euryale breeding database construction system based on big data collection, comprising: a data collection module for collecting germplasm resource data and market demand data, and collecting final agronomic performance data; a processing scheme optimization module for receiving germplasm resource data, determining the best pretreatment scheme, and the best pretreatment scheme aims to maximize seed shell permeability and minimize embryo damage; a genetic value evaluation module for genotyping seeds treated according to the best pretreatment scheme and constructing a breeding value index; a decision feedback module for storing the related data of excellent single plants in a successful breeding event record, generating a breeding strategy based on historical successful breeding event records to guide new breeding practices, and responding to the final agronomic performance data. The present application constructs a data-driven closed-loop breeding optimization system that can continuously learn and optimize itself.

Description

Technical Field

[0001] This invention relates to the field of agricultural information technology, specifically to a system for constructing a water chestnut breeding database based on big data collection. Background Technology

[0002] As an important aquatic economic crop that is both medicinal and edible, the demand for high-quality seedlings and diversified varieties in the market is becoming increasingly urgent as the scale of its industrial planting continues to expand. In order to accelerate the breeding process, modern biotechnologies such as transgenic breeding technology and molecular marker-assisted breeding have been adopted.

[0003] However, the application of these advanced breeding technologies to Euryale ferox is severely limited by its own biological characteristics. Its thick, hard seed coat forms a strong physical barrier, greatly hindering the efficiency of key breeding operations. Although molecular marker-assisted breeding technology can provide rapid and accurate genotyping methods for screening and verifying successful transgenic individuals or the aggregation of desirable traits, its effectiveness depends on successfully obtaining a sufficient number of transformants or hybrid offspring in the early stages. Due to the physical barrier of the seed coat, the initial success rate of gene introduction or tissue culture is extremely low, which constitutes a bottleneck in the entire efficient breeding process.

[0004] Therefore, how to systematically optimize processing methods in a data-driven manner to overcome the physical obstacle of the hard seed shell of Euryale ferox, and efficiently integrate the optimization scheme with the subsequent molecular breeding process to support the breeding of new Euryale ferox varieties with higher yield, better quality and stronger stress resistance, is a technical problem that urgently needs to be solved in the field of Euryale ferox breeding. Summary of the Invention

[0005] The purpose of this invention is to provide a database construction system for gorgon fruit breeding based on big data collection, in order to solve the problems existing in the background technology. The system includes:

[0006] The data acquisition module is used to collect germplasm resource data and market demand data, and to recollect the final agronomic trait performance data;

[0007] The processing scheme optimization module is used to receive the germplasm resource data and determine the optimal pretreatment scheme, which aims to maximize seed coat permeability and minimize embryo damage.

[0008] The genetic value assessment module is used to perform genotyping on seeds treated according to the optimal pretreatment scheme, construct a breeding value index, and screen out superior individual plants based on the breeding value index.

[0009] a decision feedback module configured to store the data associated with the elite individual into a record of successful breeding events, generate a breeding strategy based on the record of successful breeding events to guide new breeding practice, and respond to the final agronomic performance data.

[0010] Preferably, the treatment scheme optimization module comprises:

[0011] an index acquisition unit configured to acquire original permeability indexes and original damage degree indexes through pre-treatment experiments;

[0012] a score construction unit configured to convert the original permeability indexes into permeability scores and convert the original damage degree indexes into damage degree scores;

[0013] a scheme determination unit configured to perform weighted operation on the permeability scores and the damage degree scores in combination with preset permeability weights and preset damage degree weights, construct a seed treatment effectiveness score, and determine the best pre-treatment scheme according to the seed treatment effectiveness score.

[0014] Preferably, the permeability weights and the damage degree weights are set in response to strategic emphasis of a breeding task.

[0015] In the case of processing rare breeding materials, the damage degree weights are increased to prioritize the preservation of embryo viability.

[0016] In the case of performing large-scale preliminary screening, the permeability weights are increased to pursue high conversion efficiency.

[0017] Preferably, the genetic value evaluation module comprises:

[0018] a score determination unit configured to identify an allelic genotype of a preset target trait-associated molecular marker and convert it into a molecular marker site score;

[0019] an index construction unit configured to perform weighted operation on the molecular marker site score in combination with a trait relative importance weight determined by the market demand data to construct the breeding value index.

[0020] Preferably, the genetic value evaluation module further comprises:

[0021] an elite individual screening unit configured to preset a selection threshold and compare the breeding value index with the selection threshold;

[0022] When the breeding value index is greater than the selection threshold, the corresponding individual is marked as the elite individual.

[0023] When the breeding value index is not greater than the selection threshold, the corresponding individual is marked as an eliminated individual.

[0024] Preferably, the decision feedback module is specifically configured to integrate the germplasm information of the excellent single plant, the experienced optimal pretreatment scheme, the genotype data and the calculated breeding value index into the successful breeding event record.

[0025] Preferably, the decision feedback module comprises:

[0026] A database management unit is configured to receive the successful breeding event record, establish a traceable association between the germplasm identity, the treatment scheme, the genotype, the breeding value index and the final field phenotype data, and form a historical database.

[0027] A decision support unit is configured to analyze the associated data in the historical database, recommend a parent combination, an optimal pretreatment scheme and a trait relative importance weight allocation scheme for a new breeding target, and generate the breeding strategy.

[0028] Preferably, the decision support unit learns the breeding cases in the historical database by training a machine learning model, and realizes the prediction and recommendation functions for a new breeding project.

[0029] In view of the problem of low breeding efficiency of Euryale ferox due to hard seed coat in the prior art, the present application provides a systematic solution, which has the following beneficial effects:

[0030] 1. The present application solves the core contradiction between enhancing the permeability of seed coat and maintaining embryo viability by establishing a quantitative evaluation model. The system can flexibly adjust the evaluation weight according to the emphasis of the breeding task, such as prioritizing the preservation of embryo viability when processing rare materials, or pursuing high efficiency when large-scale screening, so as to objectively and scientifically determine the optimal pretreatment scheme, and overcome the key physical obstacles in breeding operation.

[0031] 2. The present application improves the breeding selection from traditional qualitative judgment to quantitative evaluation. By constructing a comprehensive breeding value index, it can combine market demand to perform weighted evaluation on multiple target traits, comprehensively measure the genetic value of individuals, and ensure that the selected offspring not only aggregates multiple excellent genes, but also has a comprehensive trait combination that best meets the market and breeding target, avoiding the one-sidedness of single trait selection.

[0032] 3. The present application integrates each successful breeding practice, including germplasm source, treatment method, genotype, value index and final field performance data, into a complete and traceable event record and stores it in the database. This design systematizes the dispersed breeding information and forms a data-driven closed loop, providing a data basis for subsequent intelligent analysis and system self-optimization.

[0033] 4. The application uses machine learning technology to deeply analyze the historical database, autonomously learns the complex rules from germplasm resources to the final agronomic traits, and when a new breeding project starts, the system can generate a data-driven breeding strategy, intelligently recommend parent combinations, the best treatment scheme and trait weight, and improve the scientificity and predictability of breeding decisions. BRIEF DESCRIPTION OF DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0035] Figure 1 The logical block diagram of the system of the present application is shown in Figure 1.

[0036] Figure 2 The logical block diagram of the processing scheme optimization module of the present application is shown in Figure 2.

[0037] Figure 3 The logical block diagram of the genetic value evaluation module of the present application is shown in Figure 3.

[0038] Figure 4 The logical block diagram of the decision feedback module of the present application is shown in Figure 4. DETAILED DESCRIPTION

[0039] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0040] Embodiment one

[0041] Please refer to Figure 1 The present application provides a gordon euryale breeding database construction system based on big data collection, which comprises:

[0042] A data collection module is used to collect germplasm resource data and market demand data, and to collect final agronomic trait performance data.

[0043] A processing scheme optimization module is used to receive germplasm resource data and determine the best pretreatment scheme, which aims to maximize seed shell permeability and minimize embryo damage.

[0044] A genetic value evaluation module is configured to genotype seeds treated according to the optimal pretreatment scheme, construct a breeding value index, and screen out excellent single plants based on the breeding value index;

[0045] A decision feedback module is configured to store relevant data of the excellent single plants in a successful breeding event record, and generate a breeding strategy based on historical successful breeding event records to guide new breeding practices, and then respond to the final agronomic trait performance data;

[0046] The embodiment of the present application provides a gordon euryale breeding database construction system based on big data collection, which aims to systematically solve the core problem of low application efficiency of modern breeding technology caused by the hard physical barrier of gordon euryale seed coat; the system is driven by a closed-loop feedback data flow, and the data collection module systematically collects various types of basic data required for breeding projects, including germplasm resource data of specific gordon euryale varieties and market demand data defining the breeding direction; after receiving the specific germplasm data, the processing scheme optimization module accurately calculates the optimal pretreatment scheme aimed at maximizing the permeability of the seed coat and minimizing the embryo damage; the seeds treated by this scheme are subjected to high-throughput genotyping and screening by the genetic value evaluation module, and the excellent single plants with the highest genetic potential are identified according to the quantitative breeding value index; the decision feedback module solidifies the complete data files of these excellent single plants into a successful breeding event record for structured storage, and forms a new breeding strategy by analyzing a large number of historical records to guide subsequent breeding practices;

[0047] A key operating mechanism is that the data collection module also collects the final agronomic trait performance data of the excellent lines that have entered the field planting, and the collected data serves as an objective standard for verifying and correcting the prediction performance of the entire system, thereby constructing a data-driven closed-loop breeding optimization system that can continuously learn and optimize itself.

[0048] Embodiment two

[0049] Please refer to Figure 2 The processing scheme optimization module comprises:

[0050] An index acquisition unit is configured to acquire the original permeability index and the original damage degree index through a pretreatment experiment;

[0051] A score construction unit is configured to convert the original permeability index into a permeability score and convert the original damage degree index into a damage degree score;

[0052] A scheme determination unit is configured to combine a preset permeability weight and a preset damage degree weight, perform weighted operation on the permeability score and the damage degree score, construct a seed treatment effectiveness score, and determine the optimal pretreatment scheme according to the seed treatment effectiveness score;

[0053] Setting of permeability weight and injury degree weight, in response to strategic focus of breeding task;

[0054] Wherein, in the treatment of rare breeding materials, the injury degree weight is increased to prioritize the preservation of embryo viability;

[0055] In the large-scale preliminary screening, the permeability weight is increased to pursue high transformation efficiency;

[0056] The processing scheme optimization module in the embodiment of the application introduces a quantitative evaluation model to realize its core function, that is, to solve the core contradiction between enhancing seed coat permeability and maintaining embryo viability; The running logic of the module is started by the index acquisition unit, which acquires two key raw measurement data through rigorous experimental design for each pre-treatment scheme to be evaluated: one is the raw permeability index reflecting the treatment effect, and the other is the raw injury degree index evaluating the negative impact; After obtaining the raw data, the score construction unit performs standardization processing to ensure its comparability and calculability; For the raw permeability index, a linear normalization method is adopted to convert it into a dimensionless permeability score; For the raw injury degree index which is a ratio value itself, it is directly adopted as the injury degree score;

[0057] The core step of operation is completed by the scheme determination unit, which performs a key weighted operation to construct the seed treatment effectiveness score; The construction of this quantitative model is rooted in the multi-objective decision-making theory, aiming to integrate the two mutually restrictive objectives into a single, sortable comprehensive index; Its calculation follows the following formula:

[0058]

[0059] The logical derivation basis of this formula is that the positive effect of permeability gain and the negative effect of embryo injury need to be quantitatively weighed, so as to provide a clear and objective decision-making basis for complex biological experimental results;

[0060] Wherein, represents the permeability score of a specific pre-treatment scheme The final comprehensive effectiveness score calculated is a dimensionless value, which is directly used for the ranking of different schemes;

[0061] is a parameter set defining a pre-treatment scheme, such as the combination of chemical reagent concentration and treatment time;

[0062] is the raw permeability index According to the formula The converted permeability score, which is derived from the experimental measurement value of the index acquisition unit, such as the tracer dye absorbance value measured using a spectrophotometer, and are the maximum and minimum values of the original permeability measurement observed in all experimental schemes in this round, respectively;

[0063] is the damage degree score equal to the original damage degree index which is also derived from the experimental measurement value, such as the embryo mortality rate calculated through the germination experiment, which itself is a dimensionless ratio between 0 and 1;

[0064] and are crucial weight coefficients representing the relative importance of the permeability score and the damage degree score, respectively, both of which are dimensionless parameters and satisfy The value setting logic responds to the specific breeding task strategy focus and is set by technical personnel with relevant field knowledge according to the breeding goal;

[0065] The application of this formula is reflected in that the scheme determination unit can dynamically adjust the weight according to the nature of the breeding task, so that technical personnel in this field can implement it; for example, when dealing with a non-replaceable rare Euryale ferox germplasm resource, to maximize its genetic value, the damage degree weight is set at a higher level, such as 0.7, at which time the calculation result of the formula will preferentially select the mild treatment scheme that has the least impact on embryo viability; conversely, when performing large-scale preliminary screening, the goal is to obtain the maximum number of transformants within a limited time, the permeability weight is raised to a higher value, such as 0.7, so as to select the treatment scheme with the highest transformation efficiency; by calculating the score of each scheme, the module can output the best pretreatment scheme with the highest score Through this data-driven quantitative decision-making paradigm, the success rate and overall efficiency of subsequent breeding operations are substantially guaranteed.

[0066] Example Three

[0067] Please refer to Figure 3 the genetic value assessment module, which includes:

[0068] The score determination unit is used to identify the allele genotype of the pre-set target trait-associated molecular marker and convert it into a molecular marker site score;

[0069] The index construction unit is used to combine the trait relative importance weight determined by market demand data to weight the molecular marker site score and construct a breeding value index;

[0070] The genetic value evaluation module further comprises:

[0071] The superior plant screening unit is configured to preset a selection threshold and compare the breeding value index with the selection threshold;

[0072] When the breeding value index is greater than the selection threshold, the corresponding individual is marked as a superior individual;

[0073] When the breeding value index is not greater than the selection threshold, the corresponding individual is marked as a rejected individual;

[0074] The decision feedback module is specifically configured to integrate the germplasm information of the superior individual, the optimal pretreatment scheme experienced, the genotype data and the calculated breeding value index into a successful breeding event record;

[0075] The genetic value evaluation module in the embodiment of the application is positioned at the core function of quickly and accurately evaluating the genetic potential of a large number of offspring individuals after successfully breaking through the shell barrier; the evaluation process is started by the score determination unit, which accurately identifies the allelic genotype of each individual at these key sites by analyzing the genotyping data generated by high-throughput sequencing according to the preset target trait-associated molecular marker set, and converts the genotype information into a quantifiable and dimensionless molecular marker site score without ambiguity according to the preset genetic effect model;

[0076] The index construction unit integrates information from multiple molecular marker sites into a comprehensive breeding value index based on the classical quantitative genetics selection index theory; the theoretical basis of the index model aims to overcome the trade-off caused by single trait selection and achieve comprehensive evaluation of the overall genetic advantage of individuals, and the construction process follows the following formula:

[0077]

[0078] The logical basis of this formula is that a mathematical tool that can integrate information from multiple genetic sites and make comprehensive decisions in combination with market economic value is needed, so as to provide a single and clear screening index for breeders to speed up the selection process;

[0079] Wherein, represents the comprehensive breeding value index of the individual , which is a dimensionless evaluation value;

[0080] is the unique identifier of the current individual plant being evaluated;

[0081] is the total number of target trait-associated molecular markers contained in the evaluation system, and the source of this parameter is the authoritative research results published in the prior art or the set of markers verified in the prior exploration of the system to be strongly associated with the target trait, ensuring the biological significance of the selected markers, and each molecular marker in the evaluation system corresponds to a specific target trait;

[0082] is the individual The score at the th molecular marker locus, which is derived from the analysis results of the genotyping experiment data by the score determination unit, for example, for a SNP site, the genotypes of advantageous homozygous, heterozygous, and disadvantageous homozygous are respectively assigned scores of 2, 1, and 0;

[0083] is the relative importance weight of the trait corresponding to the th molecular marker, which is a dimensionless parameter, and its value is directly responsive to the market demand data collected by the data acquisition module and the expected economic benefits of the breeding project. To ensure that the breeding value indices calculated for different individuals or different breeding projects are comparable, the sum of the weights of all traits should satisfy the constraint condition ;

[0084] The application of this formula is to calculate the for each individual, and the index construction unit completes the quantitative ranking of its genetic potential; after the intervention of the elite screening unit, a clear selection threshold is preset, and the setting logic of this threshold is based on the selection rate set by the breeder according to the breeding resources and targets, so as to ensure that it can be implemented by those skilled in the art; when the value of a certain individual is higher than this threshold, the individual is marked as an excellent single plant and obtains the qualification to enter the next round of cultivation, otherwise it is marked as an eliminated individual; this series of calculation and comparison converts complex genotype data into clear breeding decisions; after this screening process, the decision feedback module integrates all relevant information of these screened excellent single plants, including their original germplasm information, the best pretreatment scheme they have experienced , complete genotype data, and the calculated value, to generate a structured and uniquely identified successful breeding event record; the generation of this record file not only marks the complete closure of the breeding selection process from data input to value output, but also reserves high-quality data units for the decision support function of the system.

[0085] Example Four

[0086] Please refer to Figure 4 , the decision feedback module comprises:

[0087] a database management unit configured to receive successful breeding event records and establish traceable correlations between germplasm identity, treatment scheme, genotype, breeding value index and final field phenotype data to form a historical database;

[0088] a decision support unit configured to analyze the correlation data in the historical database and recommend parent combinations, optimal pretreatment schemes and trait relative importance weight distribution schemes for new breeding objectives to generate breeding strategies;

[0089] the decision support unit learns breeding cases in the historical database by training a machine learning model to achieve prediction and recommendation functions for new breeding projects;

[0090] The decision feedback module in the embodiment of the present application is the core decision and feedback component of the system, which converts historical breeding data into accurate guidance for future breeding activities; the workflow of the module is that the database management unit, as the data warehouse center of the system, is responsible for receiving and deeply integrating each successful breeding event record generated by the previous modules; the core function of the unit is to establish strong logical correlations between data, strictly bind germplasm identity ID, optimal treatment scheme ID (and its score), offspring genotype ID (and its score) and final field phenotype data collected by the data collection module, thereby constructing a historical database with rich content and complete traceability;

[0091] Relying on the structured data foundation, the decision support unit can exert its powerful analysis and prediction functions; the unit is embedded with a machine learning model, which is trained by using the massive breeding cases accumulated in the historical database to learn and master the complex rules hidden behind the data through deep learning, i.e., the nonlinear mapping relationship between parent combinations, treatment methods, genotypes and final agronomic traits;

[0092] The technical effect produced by this design in the application level is remarkable; when a breeder sets a new breeding objective and inputs the germplasm information of the parent to be selected, the decision support unit can call its prediction algorithm to generate a highly customized breeding strategy; based on the trained model, the unit can predict the probability of producing high offspring from different parent hybrid combinations, thereby recommending the optimal parent combination; at the same time, it can match or directly predict the most suitable seed coat pretreatment scheme for the new parent from the historical database according to the germplasm characteristics of the new parent ; moreover, it can recommend a set of optimal trait relative importance weights for the model according to the new breeding objective and the latest market demand data The allocation scheme; in this way, the decision feedback module changes the past breeding experience from a fuzzy, artificial-dependent paradigm into a computable, predictable, and action-guiding intelligent decision; the output breeding strategy directly guides the new round of breeding practice, and the new data generated by the practice will enter the system through the data collection module for further training and optimization of the machine learning model, thereby constructing a self-adaptive breeding system capable of realizing continuous performance optimization through data iteration.

[0093] The above is only a preferred embodiment of the present application, and does not limit the present application in other forms. Any person skilled in the art can use the disclosed technical content to make changes or modifications into equivalent embodiments applied to other fields, but any simple modification, equivalent change and modification made to the above embodiments without departing from the technical solution content of the present application, according to the technical essence of the present application, still belongs to the protection scope of the technical solution of the present application.

Claims

1. A system for constructing a gordon euryale breeding database based on big data collection, characterized in that, The application comprises: a data collection module for collecting germplasm resource data and market demand data, and collecting final agronomic performance data; a treatment scheme optimization module for receiving the germplasm resource data, determining an optimal pretreatment scheme, and the optimal pretreatment scheme aims to maximize seed shell permeability and minimize embryo damage; a genetic value evaluation module for genotyping seeds treated according to the optimal pretreatment scheme, constructing a breeding value index, and screening excellent single plants based on the breeding value index; a decision feedback module for storing relevant data of the excellent single plants in a successful breeding event record, generating a breeding strategy based on historical successful breeding event records to guide new breeding practices, and responding to the final agronomic performance data; The treatment scheme optimization module comprises: an index acquisition unit for acquiring original permeability indexes and original damage degree indexes through pretreatment experiments; a score construction unit for converting the original permeability indexes into permeability scores and converting the original damage degree indexes into damage degree scores; a scheme determination unit for combining a preset permeability weight and a preset damage degree weight, performing weighted operation on the permeability scores and the damage degree scores, constructing a seed treatment effectiveness score, and determining the optimal pretreatment scheme according to the seed treatment effectiveness score; The calculation of the treatment effectiveness score follows the formula: S eff (p) = w P ·P score (p) - w D ·D score (p) where S eff (p) represents the final overall effectiveness score computed for a specific pretreatment protocol p; p is a set of parameters defining one pretreatment protocol; P score (p) is the permeability score converted from the original permeability indicator P raw (p) by the formula P score (p) = (P raw (p) - P min ) / (P max - P min ) where P max and P min are the maximum and minimum values of the observed permeability measurements in all experimental protocols in this round; D score (p) is the damage score; w P and w D represent the relative importance of the permeability score and the damage score, respectively. The setting of the permeability weight and the damage degree weight responds to the strategic emphasis of the breeding task; wherein, when processing rare breeding materials, the damage degree weight is increased to prioritize embryo viability; when performing large-scale preliminary screening, the permeability weight is increased to pursue high conversion efficiency; The genetic value evaluation module comprises: a score determination unit for identifying the allelic genotype of a preset target trait-associated molecular marker and converting it into a molecular marker site score; an index construction unit for combining the relative importance weight of the trait determined by the market demand data to weight the molecular marker site score and construct the breeding value index; The calculation of the breeding value index follows the formula: wherein IBV s represents the comprehensive breeding value index of the individual s; s is the unique identification of the individual plant currently being evaluated; n is the total number of target trait-associated molecular markers contained in the evaluation system; M i,s is the score of the individual s at the i-th molecular marker locus; w i is the relative importance weight of the trait corresponding to the i-th molecular marker.

2. The system of claim 1, wherein: The genetic value evaluation module further comprises: a superior plant screening unit for presetting a selection threshold and comparing the breeding value index with the selection threshold; wherein, when the breeding value index is greater than the selection threshold, the corresponding individual is marked as the excellent single plant; when the breeding value index is not greater than the selection threshold, the corresponding individual is marked as an eliminated individual.

3. The system of claim 2, wherein: The decision feedback module is specifically configured to integrate the germplasm information of the excellent single plants, the optimal pretreatment scheme experienced, the genotype data, and the calculated breeding value index into the successful breeding event record.

4. The system of claim 1, wherein: The decision feedback module comprises: a database management unit for receiving the successful breeding event record, establishing traceable associations between germplasm identity, treatment scheme, genotype, breeding value index, and final field phenotype data, and forming a historical database; A decision support unit is configured to analyze the associated data in the historical database, recommend parent combinations, optimal pretreatment schemes and trait relative importance weight distribution schemes for new breeding objectives, and generate the breeding strategy.

5. The system of claim 4, wherein: The decision support unit learns the breeding cases in the historical database by training a machine learning model, and realizes the prediction and recommendation functions for new breeding projects.

Images