A corn germplasm disease resistance identification and purification rejuvenation cro hosting method and system

Abstract

This invention relates to the interdisciplinary field of maize germplasm breeding and agricultural informatization, and in particular to a CRO-managed method for disease resistance identification and purification rejuvenation of maize germplasm. The technical solution includes: constructing a standardized identification process for the simultaneous artificial induction of four diseases: large leaf spot, small leaf spot, rust, and sheath blight; designing a four-stage, progressively gradual purification process at the seedling, jointing, pre-pollen shedding, and maturity stages; and supporting standardized plot isolation and intelligent soil homogenization management specifications. Simultaneously, a four-layer dedicated intelligent management system is built, achieving automatic data archiving throughout the entire process through on-site data collection via mobile field terminals, a traceability coding engine, and blockchain evidence storage. The annual coefficient of variation for multi-disease identification results is reduced to below 9%, and the purity of inbred lines remains stable at ≥99.8% after two consecutive generations of plant-row comparison. The entire solution is suitable for large-scale maize CRO germplasm management, disease resistance identification, purification rejuvenation, and standardized business in Hainan's Nanfan breeding area, meeting all the needs of breeding clients for delivery, Nanfan park supervision, and digital auditing and traceability of scientific research projects.

Description

Technical Field

[0001] This invention relates to the interdisciplinary field of maize germplasm breeding and agricultural informatization, and in particular to a CRO-managed method and system for the identification, purification and rejuvenation of maize germplasm disease resistance. Background Technology

[0002] Many breeding units in Hainan's Nanfan region have outsourced their maize inbred line disease resistance identification and germplasm purification services to third-party CROs. However, the existing technology suffers from multiple deficiencies and lacks a supporting intelligent management and digital system, as detailed below: Disease resistance identification relies on natural disease development in the field, making it impossible to simultaneously induce multiple diseases. There are no unified quantitative standards for spore concentration and inoculation timing, leading to significant errors in annual and personnel-based testing, and making identification data incomparable. There is no standardized program for the simultaneous artificial induction of four diseases: large leaf spot, small leaf spot, rust, and sheath blight. Germplasm removal is only performed once at maturity, resulting in the omission of many hybrid plants during the seedling, jointing, and pre-pollen shedding stages. There is no standardized four-level progressive removal process, leading to severe segregation of traits after continuous germplasm propagation and low self-pollination purity. The layout of experimental plots lacks unified quantitative parameters, and there are no standardized control rules for isolating isolation zones, protective rows, and soil homogenization. Differences in the microenvironment of each plot interfere with disease resistance identification results, and data from different batches cannot be compared horizontally. There are no unified standards for self-pollination bagging, pollination, and plant-row comparison operations, resulting in high risks of cross-pollination by foreign pollen and mechanical contamination, further reducing purification purity. The entire process relies on paper-based records, with disease surveys, images of cleaned plants, individual plant numbers, and seed information stored in a scattered manner. There is no unique traceability code binding, and there is a lack of mobile data collection, cloud storage, and automatic generation of audit reports, making it difficult for regulatory verification and customer acceptance.

[0003] In summary, existing technologies do not simultaneously integrate a unified solution for simultaneous artificial induction and identification of multiple diseases, four-level progressive purification, standardized plot management, and end-to-end digital intelligent management, and therefore cannot meet the standardized and traceable delivery requirements of large-scale maize CRO germplasm trusteeship in Hainan Nanfan.

[0004] In view of this, we propose a CRO-managed method and system for the identification, purification, and rejuvenation of maize germplasm disease resistance to solve existing problems. Summary of the Invention

[0005] The purpose of this invention is to provide a CRO-managed method and system for identifying, purifying, and rejuvenating maize germplasm to combat disease resistance, thereby addressing the problems mentioned in the background section.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a CRO-managed method for disease resistance identification, purification, and rejuvenation of maize germplasm, using a single maize germplasm as an independent CRO-managed unit and assigning a unique traceability code, comprising the following steps: S1. Standardized experimental plot planning and soil homogenization management: Each variety is set up as an independent experimental plot, with uniform plot specifications and biological replication; protective rows are arranged around the plots, isolation walkways are reserved between plots, and pollen isolation belts are set up around the experimental fields; soil fertility data of the plots are tested before sowing and entered into the intelligent management and control system, which automatically generates a zonal soil improvement plan and regulates the homogenization of soil fertility in the plots; plot layout maps are drawn and uploaded to the cloud to generate a unique traceability code for each plot. S2. Simultaneous Artificial Induction of Multiple Diseases for Disease Resistance Identification: A mixed suspension containing pathogens of multiple target diseases was prepared, and multiple diseases were simultaneously induced in corn at the appropriate time using foliar inoculation and soil irrigation. Disease-resistant and susceptible control varieties were simultaneously arranged in the test plots, and were sown at the same time as the germplasm to be tested, with unified water and fertilizer management. At the appropriate time, the occurrence of diseases was investigated plant by plant using a grading standard, disease images were collected on-site and disease level data were entered, and the system automatically calculated the disease index and associated it with the plot source code. Among them, the multiple target diseases include large leaf spot, small leaf spot, rust, and sheath blight. Large leaf spot and small leaf spot are distinguished by the diameter of the skin lesions. The skin lesion diameter is greater than the set threshold and the skin lesion diameter is less than or equal to the set threshold and the small leaf spot. S3. Multi-level gradient field screening: The screening is carried out in different gradients at multiple growth and development stages of maize. At each stage, different plants, segregating plants and degenerate plants are removed according to the corresponding screening criteria. At least 30 typical target plants are retained in each plot and labeled. The on-site images of the screening at each level are uploaded to the intelligent management and control system for archiving in real time. S4. Standardized single-plant isolation self-pollination purification: Selected typical single plants are isolated by bagging before the female ear silks emerge. Pollen from the male ear of the plant is collected for artificial self-pollination. After pollination, the isolation bag is replaced. Single ears are harvested, dried and threshed independently. The single plant number is entered and the system is associated with the corresponding plot traceability code. S5. Multi-generation row comparison and rejuvenation: The single ear seeds obtained in the previous step are planted independently as rows and biological replicates are set up. The traits are compared with standard germplasm at multiple growth and development stages. Rows that show trait segregation at any stage are eliminated. Stable rows are screened for multiple generations and mixed harvested to obtain high-purity core germplasm. Row comparison data and purity test data are entered into the system simultaneously. S6. Full-process intelligent traceability and archiving: The intelligent management and control system uses the community traceability code as the main key to automatically collect business data and image data of each link, and supports one-click export of multiple types of reports.

[0007] Furthermore, in step S1: the plot specifications are: row length 4 to 6m, row spacing 0.5 to 0.7m, 18 to 22 plants per row, with 3 biological replicates; the protection behavior consists of more than 2 rows of the same type of maize, the isolation walkway width is 1.0 to 1.5m, and the pollen isolation zone distance is 200 to 500m; the soil fertility homogenization standard is: the relative standard deviation (RSD) of soil fertility in the same plot is ≤10%.

[0008] Further, in step S2, the volume ratio of the four pathogenic spores in the mixed suspension is a:b:c:d, where 0.8≤a≤1.2, 0.8≤b≤1.2, 0.8≤c≤1.2, 0.8≤d≤1.2, and the total spore concentration is 0.5×10⁻⁶. 6 Up to 2×10 6 The concentration is 1 / mL, and 0.01% to 0.1% surfactant is added to the solvent.

[0009] Furthermore, in step S2: foliar inoculation is carried out by spraying during the corn jointing stage; soil irrigation is carried out along the root and stem base during the corn tasseling stage; the appropriate survey period is 5 to 10 days after corn tasseling; the grading standard is a grading standard of 0 to 9; the disease index calculation formula is: DI = Σ(number of diseased plants at each level × corresponding disease level) / (total number of surveyed plants × highest disease level) × 100.

[0010] Furthermore, in step S3, multiple growth and development stages include the seedling stage, jointing stage, pre-pollen shedding stage, and maturity stage. The screening criteria for each stage are as follows: during the 3 to 5 leaf stage of the seedling stage, seedlings with abnormal leaf color, leaf shape, or plant height deviating from the plot average by more than ±20% are removed; during the jointing stage, plants with segregated stem or leaf characteristics and degenerate plants are removed; during the pre-pollen shedding stage, plants with abnormal tassel branching characteristics or anther color are eliminated; and during the maturity stage, segregated plants are eliminated based on ear shape, grain color, and cob color.

[0011] Furthermore, in step S4: the bagging is done using parchment paper bags, 1-3 days before the female ear emerges silk, and the bag opening is fixed; pollination time is from 8:00 am to 12:00 pm daily; after pollination, a new isolation bag is replaced; each ear is harvested independently and threshed in separate bags.

[0012] Furthermore, in step S5: the planting specifications are the same as in step S1, with 18 to 22 plants per row, and 3 biological replicates are set; trait comparisons are conducted at four stages: seedling stage, jointing stage, pre-pollen shedding stage, and maturity stage; after two consecutive generations of screening to stabilize the plant spacing, the plants are harvested together; purity is detected using molecular marker detection methods; the purity of the core germplasm is ≥99.8%.

[0013] A CRO managed system for maize germplasm disease resistance identification, purification, and rejuvenation is disclosed. This system executes a CRO managed method for maize germplasm disease resistance identification, purification, and rejuvenation. The system comprises a data acquisition terminal layer, a data transmission layer, a cloud service layer, and a business output layer. The data acquisition terminal layer includes field data acquisition terminals, plot mapping terminals, and image acquisition terminals, supporting the reading of traceability codes and on-site recording of field operation data and real-time images. The data transmission layer includes a wireless communication module to achieve bidirectional data transmission between the acquisition terminals and the cloud server. The cloud service layer includes a plot planning management module, a disease resistance identification management module, a purification process control module, and a traceability control module. The system includes a connection engine, a data storage module, and a report output module; a community planning and management module receives soil testing data, outputs soil improvement plans, and generates community layouts and traceability codes; a disease resistance identification and management module has a built-in disease grading standard and disease index calculation engine, and stores disease data and images; a purification process control module pushes work reminders in stages and manages the comparison ledger of individual plant numbers and plant rows; a traceability association engine uses the traceability code as a unique index to associate the entire process of business data; a data storage module encrypts and stores business records; a report output module generates various types of output files; and a business output layer includes a management backend, a client query terminal, and a data export interface.

[0014] Furthermore, the cloud service layer also includes a work node push module, which is used to push reminders of work time nodes for disease investigation, weed removal and screening, and plant row comparison to the field collection terminal in real time; the data storage module uses blockchain encryption for storage, and writes the business records to the blockchain after hashing; the report output module outputs files including germplasm delivery identification reports, park supervision statistics ledgers, and project audit data tables.

[0015] Furthermore, the traceability code serves as the unique primary key for data throughout the entire process, and the associated data range includes plot layout drawings, soil improvement records, disease inoculation and grading data and images, images of impurity removal at each stage, single-plant pollination ledgers, plant-row comparison records, and purity test reports.

[0016] Compared with the prior art, the beneficial effects of the present invention are: This invention pioneers a system for the simultaneous artificial induction and identification of four diseases: large spot, small spot, rust, and sheath blight. It standardizes spore ratios, inoculation methods, control settings, and grading criteria, reducing the annual coefficient of variation of the disease index from over 30% in traditional natural disease to below 9%. Identification data from different batches and by different operators can be compared horizontally. A four-stage, progressively gradual purification mechanism is constructed, encompassing the seedling stage, jointing stage, pre-pollen shedding stage, and maturity stage. This multi-level elimination of heterogeneous plants and comparison with two consecutive generations of plant rows ensures that the inbred lines achieve a stable purity of over 99.8%, compared to a single purification at maturity. The purity is increased by 2.5 percentage points; standardized plot size and multiple pollen isolation are matched with a soil intelligent homogenization module, which uniformly and quantitatively manages the microenvironment of the plot, greatly reducing the interference of soil fertility and pollen drift on disease resistance identification results; it is equipped with a dedicated four-layer intelligent management and control system, which collects data in real time on the mobile terminal, automatically collects the whole link information in the cloud, and uses blockchain to store evidence to prevent tampering, completely replacing paper handwritten ledgers and reducing the manual workload of file organization by more than 70%; the whole process parameters are standardized and the operation is replicable, which is suitable for large-scale corn CRO germplasm trusteeship business. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the workflow of a CRO-managed method for identifying, purifying, and rejuvenating maize germplasm disease resistance according to the present invention. Detailed Implementation

[0018] The technical solution of the present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0019] In the description of this invention, the term "CRO" refers to a Contract Research Organization, which in this invention specifically refers to the managed operation model of a third-party service organization in the seed breeding industry that accepts commissions from breeding units to conduct maize germplasm disease resistance identification and purification rejuvenation services. The term "traceability code" refers to a unique identification code generated by the intelligent management system for each independent CRO-managed germplasm and its corresponding experimental plot, serving as the primary key for data association throughout the entire process. The term "biological replication" refers to an independent replicate experimental unit set up under the same germplasm treatment conditions, used to eliminate random errors and ensure the statistical reliability of the data.

[0020] This invention's experiments are uniformly conducted in the core area of ​​Hainan's maize breeding region, including but not limited to CRO-managed experimental bases in Sanya Yazhou Bay, Ledong, and Lingshui. The experimental period is from November to March of the following year, during the maize breeding season, matching Hainan's dry season climate conditions to ensure consistency in environmental factors such as temperature, light, and precipitation. Supporting hardware includes: industrial-grade dustproof and waterproof touchscreen tablets (with visibility capabilities in bright field conditions), GPS / GNSS high-precision plot mapping terminals, high-definition disease image capture terminals (including macro shooting capabilities), backpack-mounted manual / electric spraying and inoculation equipment, electronic weighing equipment, and conventional field breeding tools. The supporting software platform is the Nanfan Maize CRO Multi-Disease Identification and Multi-Level Purification and Rejuvenation Intelligent Management and Control System. This system is a four-layer cloud-native application with built-in standardized plot planning modules, multi-disease identification management modules, multi-level purification process control modules, traceability coding and association engines, blockchain encrypted evidence storage modules, and multi-type report output modules. The system pushes task reminders to operators via mini-programs / apps, and all data is transmitted back to the cloud server in real time via 4G / 5G networks.

[0021] This invention uses a single maize germplasm (including but not limited to inbred lines, recombinant inbred lines, backcross lines, etc.) as an independent CRO (Centralized Reproductive Technology) unit, with each germplasm assigned a unique traceability code. The germplasm types cover a variety of ecotypes such as sweet corn, waxy corn, popcorn, silage corn, and common corn.

[0022] The present invention will be further described below through specific embodiments and comparative examples.

[0023] Example 1: Intelligent managed identification and purification of sweet corn inbred lines throughout the entire process This embodiment uses a fresh sweet corn inbred line (number HZ-S-021) commissioned by a breeding unit as an example to fully demonstrate the entire process of the present invention.

[0024] like Figure 1 As shown, the workflow of a CRO-managed method for disease resistance identification and purification rejuvenation of maize germplasm includes: standardized plot planning and intelligent soil homogenization, simultaneous artificial induction of four diseases for disease resistance identification, four-level progressive gradient impurity removal screening, standardized single-plant isolation self-pollination, two-generation plant row comparison rejuvenation, full data upload to the intelligent management and control system, blockchain evidence storage and report output.

[0025] S1: Standardized Community Planning and Intelligent Homogenized Soil Management The commissioned germplasm was the sweet corn inbred line HZ-S-021, with each germplasm sample independently divided into dedicated experimental plots. Before sowing, mixed soil samples were collected from the experimental plots and sent to a qualified testing institution to determine basic soil fertility indicators, including but not limited to: pH value, organic matter content, available nitrogen, available phosphorus, and available potassium. In this example, the initial test results for the plots were: pH=6.2, organic matter 12.6 g / kg, and available potassium content differing by 17.8% between the highest and lowest values ​​within the plot. The calculated initial fertility RSD was 18.2%, exceeding the control standard.

[0026] Operators, equipped with GPS surveying terminals, collected latitude and longitude coordinates along the boundaries of the plots to be planned. They then uploaded the soil testing data, along with the plot boundary coordinates, to the intelligent management and control system cloud via a WeChat mini-program. The system's built-in soil homogenization algorithm module automatically calculated the soil test values ​​based on the preset target fertility standard range, outputting a zoned fertilization improvement plan: for low-fertility areas with a deficiency of available potassium, it was recommended to apply 80 kg of well-rotted organic fertilizer and 12 kg of potassium sulfate per acre; for other areas, differentiated fertilization prescriptions were generated based on the specific nutrient deficiency. After applying the base fertilizer according to the system's output improvement plan, operators used a checkerboard multi-point sampling method to retest the soil fertility of the plot. The retest result showed an RSD of 8.5%, meeting the homogenization control standard of RSD ≤ 10%.

[0027] The plots were divided according to uniform specifications: 5m row length, 0.6m row spacing, 20 plants per row, and 3 biological replicates per germplasm, meaning each germplasm contained 3 independent plots. A 1.2m wide isolation walkway was reserved between plots. Two rows of the same type of non-experimental maize were arranged around each plot as a protective row to eliminate the marginal effect on plant growth. A 300m wide continuous pollen isolation belt (which can be made of tall crops such as sorghum or sugarcane, or a buffer zone) was set up around the entire experimental field to prevent maize pollen from adjacent plots from drifting into the experimental area and causing genetic contamination. The GPS-mapped plot boundary coordinates, layout dimensions, and improvement records were all automatically uploaded to the cloud. The system automatically generated a unique traceability code YM-TB-202501-A03 for each managed unit based on preset coding rules (germplasm type code + year + entrusted batch + sequence number), and simultaneously output standardized plot CAD layout drawings, storing them in the data file under this code in the cloud.

[0028] It should be noted that the specific parameters in step S1 of this invention can be adjusted within a certain range based on factors such as actual plot conditions, germplasm type, and planting density requirements. In other feasible embodiments, the row length can be set to 4 to 6 m, the row spacing can be set to 0.5 to 0.7 m, the number of plants per row can be 18 to 22, the width of the isolation walkway can be 1.0 to 1.5 m, the pollen isolation strip distance can be 200 to 500 m, and the soil fertility homogenization standard uses RSD ≤ 10% as the baseline threshold. Conventional substitutions of these parameters do not deviate from the core control concept of this invention.

[0029] S2: Simultaneous Artificial Induction of Multiple Diseases for Disease Resistance Identification This embodiment employs a simultaneous artificial induction scheme for four target diseases: large leaf spot, small leaf spot, rust, and sheath blight.

[0030] Inoculum preparation: During the disease outbreak season, morphologically identified typical diseased plant samples were collected from the field. Spores of the pathogens *Exserohilum turcicum*, *Cochliobolus heterostrophus*, *Puccinia sorghi*, and *Rhizoctonia solani* were selected and purified and propagated under laboratory conditions. Spore concentration was then determined using a hemocytometer. The four spores were mixed in a volume ratio of 1:1:1:1 and diluted with deionized water to a total spore concentration of 1×10⁻⁶. 6 The concentration of spores per mL is increased, and 0.05% (volume fraction) of Tween-20 is added to the solvent as a surfactant to enhance the adhesion of the suspension to the leaf surface. The mixed suspension should be prepared and used immediately, and inoculation should be completed within 2 hours to ensure spore viability.

[0031] Inoculation procedure: During the corn jointing stage (around the time when the 7th to 8th leaves are fully expanded), choose a sunny and windless evening or a cloudy day. Use a backpack manual sprayer to evenly spray the prepared mixed spore suspension onto both sides of the leaves of all plants to be tested, ensuring that the leaf surface is evenly covered with fine droplets without forming runoff. During the corn tasseling stage (around the time when the 12th to 13th leaves have expanded and the tassels have not yet emerged), pour the inoculum solution into the soil along the base of the plant's rootstock to induce sheath blight by utilizing soil moisture.

[0032] Control settings: Within the same experimental plot, under the conditions of simultaneous sowing and unified water and fertilizer management with the germplasm to be tested, highly resistant control varieties (standard varieties with known high resistance levels against all four target diseases) and highly susceptible control varieties (standard varieties with known high susceptibility levels against all four target diseases) confirmed by national or provincial regional trials are simultaneously deployed to ensure the effectiveness of disease identification—only when the disease index of the highly susceptible control variety reaches the preset effective threshold is the batch identification data considered valid.

[0033] Disease Investigation: On the 7th day (±1 day) after corn tasseling, operators entered each experimental plot using an industrial touch-screen tablet. The tablet scanned the plot's traceability code, and the system automatically identified the current managed unit. Operators visually assessed each surviving plant in the plot using the internationally recognized corn disease grading standard from 0 to 9 (Grade 0: immune, no lesions; Grades 1-3: mildly infected; Grades 4-6: moderately infected; Grades 7-9: severely infected to highly infected). During the investigation, typical symptoms of each diseased plant were captured on-site using a high-definition camera, and the images were automatically associated with the current plant number and corresponding disease grade. All grading data was entered plant-by-plant using the tablet's built-in spreadsheet interface.

[0034] The system's backend has a built-in automatic disease index calculation engine that calculates the disease index for each disease in real time according to the preset formula DI=Σ(number of diseased plants at each level × corresponding disease level) / (total number of plants surveyed × highest disease level)×100. All graded data, disease index calculation results, and on-site disease images are automatically bound to the traceability code YM-TB-202501-A03 and uniformly archived and stored. The measurement results for this year in this example are: Large leaf spot DI=12.6, Small leaf spot DI=10.8, Rust DI=14.2, and Sheath blight DI=9.7.

[0035] It should be noted that the parameters in step S2 of this invention can be adjusted within a certain range: the volume ratio of the four spores can be equivalently replaced within the range of 0.8 to 1.2: 0.8 to 1.2: 0.8 to 1.2: 0.8 to 1.2, and the total spore concentration can be 0.5 × 10⁻⁶. 6 Up to 2×10 6 The surfactant can be selected within the range of cells / mL, and any one of Tween-20, Tween-80, Triton X-100 or other agriculturally acceptable surfactants can be used, with a concentration range of 0.01% to 0.1% (volume fraction); the disease investigation time point can be selected within 5 to 10 days after corn tasseling.

[0036] S3: Multi-level, stepwise gradient field impurity removal and screening The intelligent control system automatically calculates four time nodes—seedling stage, jointing stage, pre-pollen shedding stage, and maturity stage—based on the sowing date entered by the operator in S1 and combined with the accumulated temperature model of corn development stages. 48 hours before each node is reached, a reminder for impurity removal is sent to the operator via a mini-program. The specific impurity removal operation in this embodiment is as follows: Seedling stage (3 to 5 leaf stage): Operators inspect seedlings row by row in each plot, using the standard traits of the target germplasm as a reference. Seedlings with leaf color deviating more than 1 / 3 from the target (e.g., yellowish-green or light green when the target is dark green), significantly different leaf shape (e.g., long, thin, or twisted leaves when the target is short and wide), or exceeding the average height of all seedlings in the plot by ±20% are removed. Simultaneously, a panoramic image of the entire plot during the seedling stage is taken and uploaded to the cloud, recording the number of removed plants.

[0037] During the jointing stage (approximately the 7th to 10th leaf unfolding stage): Focus on identifying individual plants whose stem thickness differs significantly from the target germplasm (stems are noticeably too thin or too thick) and whose leaf morphology (leaf width, leaf length, leaf angle, etc.) deviates from the target population traits. Simultaneously remove significantly weak, obviously degenerated, or suspected severely diseased individual plants. Mark the removal points and take photographs as evidence.

[0038] Early pollen shedding stage (the stage when the male spike is about to emerge but has not yet shed pollen): Each plant is inspected for the number and morphology of male spike branches, and the color of the anthers (the anther color can be light yellow, pink, purple, etc., depending on the germplasm; the standard anther color of the target germplasm should be used as a reference). Hybrid plants with abnormal male spike branching traits or anther colors deviating from the typical color of the target germplasm are discarded. The significance of removing hybrids at this stage is particularly crucial—removing hybrid plants before they release pollen effectively blocks genetic contamination of the target plant by exogenous pollen. This is a quality control advantage that a single screening at maturity cannot provide.

[0039] Maturity period (after the ears are fully mature and before harvest): Based on ear shape (cylindrical, conical, intermediate, etc.), kernel color (yellow, white, red, purple, etc.), and cob color (white cob, red cob, pink cob, etc.), the final round of screening is completed, and all segregators that are inconsistent with the ear traits of the target germplasm are eliminated.

[0040] After the four-stage tiered screening process described above, 35 typical target plants are retained in each plot. Operators affix plastic numbered tags to the retained plants starting from the seedling stage (candidate plants are initially marked at the seedling stage, and the tags are dynamically adjusted based on the screening results at subsequent stages). On-site images, removal quantities, and tag numbers for each stage of impurity removal are all uploaded in real-time to the intelligent management system via tablet / mini-program and stored and archived using the traceability code as the primary key. In other embodiments of this invention, the number of typical target plants retained in each plot can be set to any integer of no less than 30 plants, depending on the germplasm type and the client's requirements.

[0041] It should be noted that the four stages of "seedling stage—jointing stage—pre-pollen shedding stage—maturity stage" in step S3 of this invention constitute a complete multi-level gradient impurity removal process. However, the judgment thresholds and specific screening traits for each stage can be adaptively adjusted according to the biological characteristics of different maize germplasm types. For example, for silage maize, additional trait survey items such as plant height, stem diameter, and leaf greenness can be added; for waxy maize, auxiliary judgment indicators for grain waxiness can be added. These adaptive adjustments do not change the core multi-level gradient impurity removal mechanism of this invention and are all equivalent alternatives to this invention.

[0042] S4: Standardized single-plant isolation self-pollination purification operation After the system removes impurities in the early stage of pollen shedding, it automatically predicts the silking time of the female ears of the selected individual plants based on weather forecasts and variety maturity data, and sends a reminder to the operators to put on bags for isolation 2 to 3 days in advance.

[0043] Bagging Isolation: One to two days before the selected typical single plant's female ear is about to silk (the timing of bagging is judged by the exposure of 1 to 2 cm of silks at the tip of the female ear), a 20cm × 12cm parchment paper bag is used to cover the female ear from bottom to top. The opening of the bag is secured and locked at the ear stalk with a paperclip to prevent foreign pollen from entering through the gaps in the bag opening. The parchment paper bag has good air permeability, waterproofness, and light transmission, which can ensure the gas exchange and light conditions required for the normal growth and development of the female ear, while effectively blocking foreign pollen.

[0044] Artificial self-pollination: From 9:00 AM to 11:00 AM daily after the female ear is isolated by bagging (during this period, pollen shedding from the male ear is most active and pollen viability is highest), the operator removes the male ear from the plant and gently shakes the pollen into a pre-prepared clean petri dish or parchment paper bag. Immediately, using a brush or cotton swab, the pollen is evenly applied to the surface of the female ear's silks, ensuring full contact between the silks and pollen. A new parchment paper bag is then placed back over the female ear for secondary isolation to prevent re-contamination by subsequent foreign pollen. The same female ear is pollinated 2 to 3 times consecutively during the silking period (once daily) to improve the seed setting rate.

[0045] Harvesting and Threshing: Approximately 40 to 50 days after pollination, once the ears are fully mature, harvest them individually, placing each ear in a mesh bag and labeling it with a single plant number tag. Dry them in baskets until the moisture content drops to the safe storage standard (approximately 13% to 14%). Threshing is done by ear, with each ear's kernels individually bagged and labeled to prevent mechanical mixing. Operators enter the individual plant number and its corresponding seed bag number into the mini-program. The system automatically associates each individual plant number and seed bag information with the corresponding local traceability code YM-TB-202501-A03.

[0046] It should be noted that, in other embodiments of the present invention, the bagging operation can be carried out 1 to 3 days before the female ear emerges from silking, and the pollination time can be slightly adjusted between 8:00 am and 12:00 pm according to specific climatic conditions. The isolation bags can also be made of materials with similar air-permeable and barrier functions, such as sulfuric acid paper bags. These adjustments do not depart from the protection scope of the present invention.

[0047] S5: Multi-generational line comparison rejuvenation and purification The following year, during the next breeding season, the 35 single-ear seeds obtained from S4 in the previous year (from 35 selected typical target plants) were planted independently in separate rows, with each row maintaining the same specifications as S1 (5m row length, 0.6m row spacing, 20 plants per row). Three biological replicates were set up for each single-ear seed source. Clear inter-row separation markers (such as row signs of different colors) were set up between each row to prevent confusion during harvest.

[0048] The system automatically retrieves the S3 multi-level impurity removal files, S2 disease identification results, and standard trait description data of the target germplasm stored from the previous year as electronic standard references for the row, which can be accessed by operators during field comparisons. Operators conduct comprehensive trait surveys of all rows at four stages: seedling stage, jointing stage, pre-pollen shedding stage, and maturity stage. If any row shows any trait segregation at any stage (including but not limited to: plant height, leaf color, leaf shape, stem thickness, tassel branching pattern, anther color, pollen shedding period, ear type, grain color, cob color, etc., showing identifiable differences from the standard inbred line), the entire row is immediately eliminated (the whole row is removed and not retained as purification material). The system simultaneously records the eliminated rows and preserves the on-site comparison images.

[0049] Two consecutive generations were screened using the same standards and methods, comparing plant rows. After two generations of screening, the plants that consistently exhibited stable and uniform traits, perfectly matching the standard germplasm characteristics, were harvested, mixed, and threshed to serve as high-purity core inbred line germplasm. Samples were sent to a qualified testing institution, where the purity of the final core germplasm was determined using the SSR (simple repeat sequence) molecular marker detection method. The test results in this example show that the final purity of the HZ-S-021 sweet corn inbred line was 99.85%, significantly better than the CRO high-purity inbred delivery standard.

[0050] It should be noted that, in other embodiments of the present invention, the number of generations for comparison can be two or more (e.g., three or four generations), and the purity detection method can be any molecular marker detection method known in the art, such as SSR marker, SNP chip, InDel marker, etc. The core germplasm purity threshold is set at ≥99.8% as the lower limit.

[0051] S6: End-to-end intelligent traceability archiving and blockchain evidence storage The intelligent control system uses the YM-TB-202501-A03 traceability code as the primary key to automatically connect all electronic data generated throughout the entire process of this embodiment, including: (1) Community planning data: GPS survey drawings, community CAD layout map, and comparison table of soil improvement test data before and after improvement; (2) Disease identification data: original data of plant-by-plant grading of four diseases, calculation results of disease index, and on-site symptom images of various diseases; (3) Multi-level impurity removal data: timestamps of impurity removal operations in four stages, number of plants removed, panoramic / close-up images of each stage, list of retained individual plant numbers and images of labels; (4) Self-pollination data: the date of bagging, pollination, harvest, and ear grain characteristics of each selected single plant; (5) Plant row comparison data: trait survey records of each stage of plant row comparison between two generations, images of eliminated plant rows, and images of mixed harvesting of retained plant rows; (6) Purity test data: SSR molecular detection original electrophoresis image or genotyping data, final purity test report.

[0052] The system generates a Merkle tree for all the aforementioned business datasets, performs a SHA-256 hash operation on the root node to generate a unique digital fingerprint, and writes this fingerprint, along with a timestamp, into the consortium blockchain network to form an immutable evidence. Clients, regulators, or auditors can verify the authenticity of key business data under any traceability code item at any time through the verification portal provided by the system.

[0053] The one-click export function supports the simultaneous generation of three types of standardized documents: (1) CRO customer delivery appraisal report - including disease identification results, purity certificate and core pedigree information; (2) Nanfan Park supervision and statistics ledger - summarizing the quarterly management business statistics according to the park's required format; (3) Scientific research project audit electronic case file - a complete data package containing all original images, original records and evidence certificates. No manual paper materials are required throughout the process, which greatly improves delivery efficiency and data integrity.

[0054] Example 2: Batch intelligent simultaneous identification and purification of 12 samples of silage corn using CRO This embodiment verifies the ability of the present invention to be applied on a large scale in the scenario of batch germplasm management.

[0055] The experimental batch included 12 silage maize inbred lines (numbered SQ-01 to SQ-12), from 6 different breeding units. The system operation procedure is as follows: Batch Import: In the PC management backend of the intelligent control system, operators import basic information (germplasm name, source unit, target trait description, seed quantity, etc.) for 12 silage corn samples at once. The system automatically generates personalized soil homogenization fertilization plans and standardized plot layout maps based on the biological characteristics of each germplasm sample, and generates independent traceability codes for each sample under unified coding rules (SQ-01 to SQ-12 corresponding to exclusive codes). A total of 36 independent experimental plots are allocated to the 12 germplasms (each sample has 3 replicates).

[0056] Uniform inoculation: In field management, operators prepared a mixed spore suspension of the four diseases (spore volume ratio 1:1:1:1, total spore concentration 1×10⁻⁶) following a procedure identical to that in Example 1. 6 Foliar spraying at the jointing stage and soil irrigation at the tasseling stage were carried out sequentially in 36 plots of 12 germplasms using a backpack sprayer (containing 0.05% Tween-20 per mL).

[0057] Batch Disease Survey: Seven days after tasseling, two operators, each carrying an industrial tablet, entered the field to survey plots with different traceability codes. The tablets automatically identified the plot affiliation by scanning the codes, entered disease level data from 0 to 9 for each plant, and took images. The system backend aggregated the disease index of 12 germplasm samples in real time, generating 12 independent disease identification data files in batches.

[0058] Differentiated impurity removal management: Based on the biological characteristics of each germplasm and the unified four-stage push operation, the system adds three additional survey and input items to the electronic survey forms for the pre-pollen shedding and maturity stages, namely plant height (accurate to cm), stem diameter (accurate to mm), and leaf greenness (using a visual evaluation scoring standard of 1 to 9 levels), to meet the special attention to biomass-related traits in silage corn breeding.

[0059] The steps of single-plant self-pollination, ear row harvesting, and plant-row comparison were performed in batches according to the standardized process in Example 1. After the plant-row comparison screening of two generations was completed the following year, the system automatically split and generated 12 complete purification pedigree archive data packages according to 12 independent traceability codes, and delivered them to the corresponding 6 breeding units.

[0060] Experimental results: The average purity of the 12 silage corn inbred lines after two generations of purification was 99.82% (range 99.78% to 99.89%); the average annual coefficient of variation of the disease index for multiple disease identification was 6.2%; the labor time expenditure in the file sorting process was reduced by 72% compared with the traditional paper management model, and the overall operational efficiency of large-scale outsourced operations was greatly improved.

[0061] Example 3: Standardized Management and Control of Spatially Isolated Land Parcels This embodiment verifies the applicability of the present invention in complex plots with insufficient natural isolation conditions.

[0062] A CRO-managed experimental plot is located in an area surrounded by conventional farmland. According to GPS measurements, there is a commercial corn production field 180m to its north, which does not meet the standard requirement of a 300m natural pollen buffer zone.

[0063] System Response: During the planning phase of the S1 plot, operators entered information about the presence of commercial cornfields on the north side of the plot into the system as a note. The system's built-in pollen drift risk assessment module automatically calculated and assessed the risk based on historical wind direction statistics (northeast winds prevail during the Hainan breeding season), generating the following remedial isolation plan and simultaneously writing it into the source tracing file: (1) On the threatened north windward side, plan to plant an 18m wide sorghum tall stalk barrier belt (sorghum plants can reach 2.5 to 3.0m in height, significantly higher than corn plants, which can effectively intercept and dilute the airflow carrying foreign corn pollen). The barrier belt layout drawing is incorporated into the community planning CAD drawing. After the barrier belt is planted, take photos, archive them, and upload them to the cloud. (2) The system will increase the frequency of inspections in the early stage of pollen shedding in the S3 phase from once every 3 days to once a day, and push inspection check-in reminders to operators via the mini-program at 7:00 am every day to ensure that any hybrid plants are discovered and removed as soon as possible before their male ears shed pollen; (3) During the S4 self-pollination stage, the system mandates that all selected typical single plants be covered with double-layered parchment bags for the female ears (i.e., a new bag of the same size is placed over the first parchment bag, and the openings of the two bags are fixed respectively) to block the entry of foreign pollen to the greatest extent possible with double barriers.

[0064] Results: After harvesting, representative samples were sent for SSR molecular purity testing. The purity was 99.81%, which was not substantially different from that under standard isolation conditions (99.85% in Example 1). This example fully demonstrates that the intelligent isolation adjustment scheme of this invention can effectively ensure the purity of germplasm purification under extreme conditions where the natural isolation distance is severely insufficient (only 180m), and has broad adaptability to complex site scenarios.

[0065] Comparison Example 1: Traditional single-stage impurity removal during the maturity phase (no multi-level gradient process, no intelligent control system) This comparative example is used to verify the purification and gain effect of the multi-level gradient impurity removal process and the supporting intelligent control system in step S3 of the present invention.

[0066] Experimental conditions control: A comparative experiment was conducted on the same plot of land, using the same sweet corn inbred line HZ-S-021 as in Example 1, with the same plot size (row length 5m, row spacing 0.6m, 20 plants per row, 3 replicates). The differences are: (1) The impurity removal process only retains the single screening based on ear characteristics at maturity, omitting the three rounds of screening at the seedling stage, jointing stage, and pollen shedding stage; (2) The entire process adopts traditional paper handwritten ledger records, without using an intelligent management and control system, without mobile terminal on-site collection, without unique traceability code binding, without cloud archiving and blockchain evidence storage.

[0067] Experimental results: After two consecutive generations of plant comparison, the purity of the HZ-S-021 inbred line was only 97.3%, 2.5 percentage points lower than that of Example 1. Disease identification relied on natural disease development instead of artificial inoculation. The annual coefficient of variation for large leaf spot disease over the two years was 31.5%, indicating severe data fluctuations and making cross-sectional comparisons impossible. Paper records were partially damaged by moisture in the high temperature and humidity environment of Hainan fields. The manual labor time for post-harvest record processing was approximately three times that of Example 1 (approximately 120 hours / record vs. 40 hours / record), and it was impossible to automatically split the germplasm into independent pedigree files and electronic audit files.

[0068] This comparative example demonstrates that without the multi-level gradient impurity removal process and digital intelligent control system of this invention, the purity of germplasm purification, the comparability of identification data, and the efficiency of record management are all significantly lower than those of the solution of this invention.

[0069] Comparative Example 2: Disease resistance identification of natural disease onset (without a multi-disease simultaneous artificial induction system) This comparative example is used to verify the identification stability gain effect of the multi-disease synchronous artificial induction identification system in step S2 of the present invention.

[0070] Experimental conditions were controlled as follows: A comparative experiment was conducted in the same plot of land, using the same sweet corn inbred line HZ-S-021 as in Example 1, and with the same plot size. The difference was that no mixed spore suspension was prepared for artificial inoculation; the disease development relied entirely on natural pathogen sources in the field environment. The remaining procedures (multi-level gradient weeding, single-plant self-pollination, comparison of two generations of plants and rows, etc.) were consistent with Example 1.

[0071] Results of repeated experiments over two consecutive years: Under the natural disease pattern, the annual coefficient of variation for the disease index of large leaf spot was 31.5%, small leaf spot 28.7%, and rust 35.2%; while the annual coefficients of variation for the three diseases corresponding to the artificial induction scheme of this invention in Example 1 were 7.8%, 6.5%, and 8.9%, respectively. Due to the interannual fluctuations in climatic conditions and field inoculum levels, the identification data for natural disease cannot be compared horizontally between different years and batches, and does not meet the standardized acceptance requirements under the CRO managed model.

[0072] This comparative example demonstrates that the four-disease synchronous artificial induction system of the present invention can reduce the annual fluctuation of disease resistance identification data to 1 / 4 to 1 / 3 of that of the natural disease pattern, significantly improving the repeatability and comparability of identification results.

[0073] Comparative Example 3: Manual paper records only (without a four-tier intelligent management and control system) This comparative example is used to verify the data management efficiency improvement effect of the four-layer architecture intelligent control system of the present invention.

[0074] Experimental conditions were controlled using the same field process parameters as in Example 1 (simultaneous manual inoculation of four diseases, multi-level gradient weeding, standardized single-plant self-pollination, and comparison of two generations of plants and rows, etc.). However, no intelligent management system was used for data collection, storage, and management throughout the process. All soil testing data, disease grading survey forms, weeding records, single-plant number tags, pollination ledgers, and plant-row comparison records were recorded and archived using traditional paper handwriting or printing methods. There was no unique traceability code or systematic binding of data at each stage, no cloud storage, no blockchain evidence, and no automatic report export function.

[0075] Comparative results: The total manual time spent on organizing, classifying, archiving, and verifying the entire set of identification and purification data increased by approximately 70% compared to Example 1 (approximately 180 man-hours vs. approximately 40 man-hours); paper documents showed some damage, blurred handwriting, and partial missing parts due to frequent handling and the hot and humid environment of Hainan; due to the lack of a systematic association with unique codes, when it is necessary to retrieve the complete pedigree file of a certain germplasm, it is necessary to manually search through multiple scattered record books and photo folders, which takes about 3 days per document; it is impossible to provide tamper-proof evidence that meets audit requirements, and it does not meet the hard requirements of Nanfan government supervision and digital audit of scientific research projects for the authenticity and completeness of data.

[0076] This comparative example demonstrates that the four-layer architecture intelligent management and control system of this invention achieves an order-of-magnitude improvement in data management efficiency through real-time data collection on mobile devices, automatic aggregation in the cloud, and blockchain-based evidence storage, and provides technical support for regulatory auditing that cannot be achieved by traditional paper-based methods.

[0077] Example 4: Verification of method repeatability within a multi-parameter range Using the same sweet corn inbred line germplasm as in Example 1, the following alternative parameters were selected to perform the entire process on adjacent plots at the same experimental site: Adjusted plot specifications: row length 4.5m, row spacing 0.65m, 19 plants per row, isolation walkway width 1.0m, pollen isolation strip 200m (subject to actual terrain limitations). The system automatically generates an adaptive plot layout drawing based on the plot boundaries.

[0078] Inoculation parameters were adjusted as follows: the volume ratio of the four spore types was adjusted to 1.1:0.9:1.0:1.0 (fluctuations within the range of 0.8 to 1.2 are allowed), and the total spore concentration was 1.5 × 10⁻⁶. 6 The inoculation rate was 1 / mL, and the Tween-20 concentration was 0.03% (volume fraction). The inoculation time for foliar spraying at the jointing stage and soil irrigation at the tasseling stage remained unchanged.

[0079] Adjustment of disease investigation time point: Disease investigation shall be carried out on the 9th day after male extraction (the 7th day in Example 1, within the tolerance range).

[0080] Fine-tuning of the impurity removal threshold: The standard for removing abnormal seedlings during the seedling stage is that the plant height deviates from the average value of the plot by ±22% (±20% in Example 1), and 32 typical single plants are retained (≥30 plants).

[0081] Independent verification results: DI=12.9 for large spot disease, DI=11.2 for small spot disease, DI=14.0 for rust disease, and DI=9.5 for sheath blight. The annual coefficient of variation of the disease index of the four diseases is ≤8.5%. The purity of SSR molecules after comparison of two generations of plants and rows is 99.82%.

[0082] Based on the above embodiments, the complete operation flow of the method of the present invention can be summarized as follows: Before the start of the annual breeding season in southern China, operators create new CRO-managed projects through the PC backend of the intelligent management system, entering the basic germplasm information (germplasm name, source, type, quantity, etc.) provided by the client. Based on the GPS mapping data and soil testing data of the plot, the system automatically generates standardized experimental plot planning maps and soil homogenization improvement plans.

[0083] After sowing, the system automatically calculates key developmental stages such as seedling stage, jointing stage, pre-pollen shedding stage, maturity stage, tasseling stage, and silking stage based on the sowing date and variety maturity information. It also sets reminders for tasks such as disease investigation, weed removal at each stage, bagging and isolation, and artificial pollination. At that time, the system will push to-do notifications to the field operators via the mini-program / APP.

[0084] When operators arrive at each work node, they bring an industrial tablet / mobile phone into the corresponding community, scan the code to read the community traceability code and then carry out the work. All on-site data (graded records, images, individual plant numbers, etc.) are collected, uploaded and bound to the data file under the code in the cloud in real time.

[0085] When comparing plant rows across different years, the system automatically retrieves all historical records of the corresponding germplasm stored in the previous year as a comparison reference, assisting operators in determining phenotypic consistency. After all plant row comparisons are completed, the system generates a final purification pedigree summary table.

[0086] After the entire process is completed, the system automatically collects all data using the traceability code as the primary key. After being stored on the blockchain, standardized reports are exported with one click and delivered to customers or submitted to regulatory authorities. The data does not require manual reprocessing throughout the entire process.

[0087] The above specific embodiments are merely several preferred embodiments of the present invention. Based on the technical solutions of the present invention and the relevant teachings of the above embodiments, those skilled in the art can make various alternative improvements and combinations to the above specific embodiments.

Claims

1. A CRO-managed method for identifying, purifying, and rejuvenating maize germplasm against disease, characterized in that, Using a single maize germplasm as an independent CRO hosting unit and assigning it a unique traceability code, the process includes the following steps: S1. Standardized experimental plot planning and soil homogenization management: Each variety is set up as an independent experimental plot, with uniform plot specifications and biological replication; protective rows are arranged around the plots, isolation walkways are reserved between plots, and pollen isolation belts are set up around the experimental fields; soil fertility data of the plots are tested before sowing and entered into the intelligent management and control system, which automatically generates a zonal soil improvement plan and regulates the homogenization of soil fertility in the plots; plot layout maps are drawn and uploaded to the cloud to generate a unique traceability code for each plot. S2. Simultaneous Artificial Induction of Multiple Diseases for Disease Resistance Identification: A mixed suspension containing pathogens of multiple target diseases was prepared, and multiple diseases were simultaneously induced in corn at the appropriate time using foliar inoculation and soil irrigation. Disease-resistant and susceptible control varieties were simultaneously arranged in the test plots, and were sown at the same time as the germplasm to be tested, with unified water and fertilizer management. At the appropriate time, the occurrence of diseases was investigated plant by plant using a grading standard, disease images were collected on-site and disease level data were entered, and the system automatically calculated the disease index and associated it with the plot source code. Among them, the multiple target diseases include large leaf spot, small leaf spot, rust, and sheath blight. Large leaf spot and small leaf spot are distinguished by the diameter of the skin lesions. The skin lesion diameter is greater than the set threshold and the skin lesion diameter is less than or equal to the set threshold and the small leaf spot. S3. Multi-level gradient field screening: The screening is carried out in different gradients at multiple growth and development stages of maize. At each stage, different plants, segregating plants and degenerate plants are removed according to the corresponding screening criteria. At least 30 typical target plants are retained in each plot and labeled. The on-site images of the screening at each level are uploaded to the intelligent management and control system for archiving in real time. S4. Standardized single-plant isolation self-pollination purification: Selected typical single plants are isolated by bagging before the female ear silks emerge. Pollen from the male ear of the plant is collected for artificial self-pollination. After pollination, the isolation bag is replaced. Single ears are harvested, dried and threshed independently. The single plant number is entered and the system is associated with the corresponding plot traceability code. S5. Multi-generation row comparison and rejuvenation: The single ear seeds obtained in the previous step are planted independently as rows and biological replicates are set up. The traits are compared with standard germplasm at multiple growth and development stages. Rows that show trait segregation at any stage are eliminated. Stable rows are screened for multiple generations and mixed harvested to obtain high-purity core germplasm. Row comparison data and purity test data are entered into the system simultaneously. S6. Full-process intelligent traceability and archiving: The intelligent management and control system uses the community traceability code as the main key to automatically collect business data and image data of each link, and supports one-click export of multiple types of reports.

2. The CRO-managed method for disease resistance identification, purification, and rejuvenation of maize germplasm according to claim 1, characterized in that, In step S1: the plot specifications are: row length 4 to 6m, row spacing 0.5 to 0.7m, 18 to 22 plants per row, with 3 biological replicates; the protection row consists of more than 2 rows of the same type of maize, the isolation walkway width is 1.0 to 1.5m, and the pollen isolation zone distance is 200 to 500m; the soil fertility homogenization standard is that the relative standard deviation (RSD) of soil fertility in the same plot is ≤10%.

3. The CRO-managed method for disease resistance identification, purification, and rejuvenation of maize germplasm according to claim 1, characterized in that: In step S2, the volume ratio of the four pathogenic spores in the mixed suspension is a:b:c:d, where 0.8 ≤ a ≤ 1.2, 0.8 ≤ b ≤ 1.2, 0.8 ≤ c ≤ 1.2, and 0.8 ≤ d ≤ 1.2, and the total spore concentration is 0.5 × 10⁻⁶. 6 Up to 2×10 6 The concentration is 1 / mL, and 0.01% to 0.1% surfactant is added to the solvent.

4. The CRO-managed method for disease resistance identification, purification, and rejuvenation of maize germplasm according to claim 1, characterized in that, In step S2: foliar inoculation is carried out by spraying during the corn jointing stage; soil irrigation is carried out along the root and stem base during the corn tasseling stage; the appropriate survey period is 5 to 10 days after corn tasseling; the grading standard is a grading standard of 0 to 9; the disease index calculation formula is: DI = Σ(number of diseased plants at each level × corresponding disease level) / (total number of surveyed plants × highest disease level) × 100.

5. The CRO-managed method for disease resistance identification, purification, and rejuvenation of maize germplasm according to claim 1, characterized in that, Step S3 involves multiple growth and development stages, including the seedling stage, jointing stage, pre-pollen shedding stage, and maturity stage. The screening criteria for each stage are as follows: during the 3 to 5 leaf stage of the seedling stage, seedlings with abnormal leaf color, leaf shape, or plant height deviating from the plot average by more than ±20% are removed; during the jointing stage, plants with segregated stem or leaf characteristics and degenerate plants are removed; during the pre-pollen shedding stage, plants with abnormal tassel branching characteristics or anther color are eliminated; and during the maturity stage, segregated plants are eliminated based on ear shape, grain color, and cob color.

6. The CRO-managed method for disease resistance identification, purification, and rejuvenation of maize germplasm according to claim 1, characterized in that, In step S4: bagging is done using parchment paper bags 1-3 days before the female ear emerges silks, and the bag opening is fixed; pollination time is from 8:00 am to 12:00 pm daily; after pollination, replace with new isolation bags; each ear is harvested independently and threshed in separate bags.

7. The CRO-managed method for disease resistance identification, purification, and rejuvenation of maize germplasm according to claim 1, characterized in that, In step S5: the planting specifications are the same as in step S1, with 18 to 22 plants per row, and 3 biological replicates are set up; trait comparisons are carried out at four stages: seedling stage, jointing stage, pre-pollen shedding stage, and maturity stage; after two consecutive generations of screening to stabilize the plant spacing, the plants are harvested together. Purity was determined using molecular marker detection methods; the purity of the core germplasm was ≥99.8%.

8. A CRO managed system for identifying, purifying, and rejuvenating maize germplasm against disease, characterized in that: The CRO-hosted method for identifying and purifying maize germplasm disease resistance as described in any one of claims 1-7 comprises a data acquisition terminal layer, a data transmission layer, a cloud service layer, and a business output layer. The data acquisition terminal layer includes field data acquisition terminals, plot mapping terminals, and image acquisition terminals, supporting the reading of traceability codes and on-site recording of field operation data and real-time images. The data transmission layer includes a wireless communication module to enable bidirectional data transmission between the acquisition terminals and the cloud server. The cloud service layer includes a plot planning and management module, a disease resistance identification management module, a purification process control module, a traceability association engine, a data storage module, and a report output module. The plot planning and management module receives soil testing data, outputs soil improvement plans, and generates plot layouts and traceability codes. The disease resistance identification management module has a built-in disease grading standard and disease index calculation engine, storing disease data and images. The purification process control module pushes operation reminders in stages and manages single-plant numbers and row-plant comparison ledgers. The traceability association engine uses traceability codes as unique indexes to associate data across the entire process. The data storage module encrypts and stores business records. The report output module generates various types of output files. The business output layer includes a management backend, a customer query terminal, and a data export interface.

9. The CRO managed system for disease resistance identification, purification, and rejuvenation of maize germplasm according to claim 8, characterized in that: The cloud service layer also includes a task node push module, which pushes task time node reminders for disease investigation, weed removal and screening, and plant row comparison to the field collection terminal in real time; the data storage module uses blockchain encryption for storage, and writes business records into the blockchain after hashing; the report output module outputs files including germplasm delivery identification reports, park supervision statistics ledgers, and project audit data tables.

10. A CRO managed system for identifying, purifying, and rejuvenating maize germplasm disease resistance according to claim 8, characterized in that: The traceability code serves as a unique primary key throughout the entire process, and the associated data includes plot layout drawings, soil improvement records, disease inoculation and grading data and images, images of impurity removal at each stage, single-plant pollination ledgers, plant-row comparison records, and purity test reports.

Images