GS model-based early prediction method for seedlings of different leaf types of pinus massoniana and SNP and InDel loci
Through the SNP and InDel loci information based on the GS model, the breeding value is calculated using the rrBLUP model, the problem of early identification of different leaf types of seedlings of Matsuma Matsuma is solved, the quality and growth performance of seedlings are improved, and the important breeding technical foundation is provided.
Patent Information
- Application Number
- CN202411458887.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-18
- Publication Date
- 2025-01-14
AI Technical Summary
The prior art cannot effectively identify different leaf-type seedlings of Matsuma Matsuma in the early stage, resulting in waste of seedling resources and difficult to predict the differences in growth performance.
Using the GS model-based method, the breeding values were constructed using the screened SNP and InDel site information, the training and testing population was constructed through the 5-fold cross-validation method, and the breeding value was calculated using the rrBLUP model, and the seedlings were determined to be all newborn leaves or all times. Leaves.
Early prediction of different leaf types of pine pine seedlings were achieved, the quality and growth performance of the seedlings were improved, and the technical basis for breeding of good seedlings was provided.
Smart Images
Figure CN119314552A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of forest seedling cultivation, and particularly relates to a method for early prediction of Masson pine seedlings with different leaf types based on a GS model, and SNP and InDel sites. Background Art
[0002] Masson pine is the main pioneer tree species for afforestation in the southern mountainous areas of my country. It has the characteristics of strong adaptability, rapid growth and high productivity. Masson pine mainly reproduces by seeds, usually raising seedlings in spring, and afforestation between the end of December of the current year and February of the following year. After the seeds of Masson pine germinate, two leaf types, primary leaves and secondary leaves (needles), will appear successively. The primary leaves are strip-shaped, single leaves are born on the young stems, and are relatively slender; the secondary leaves, that is, the needles seen on adult trees, are usually 2-3 pieces in a bundle born on the young stems, and are thicker. Some seedlings will not grow complete secondary leaves until the end of August. Studies have shown that the time of the existence of the primary leaves of Masson pine seedlings and the time of the appearance of the secondary leaves are mainly controlled by genetics, and there are significant differences among provenances, families and individuals. When the Masson pine seedlings of the same year are planted in the Qianzhong area, two different leaf types of seedlings will appear: all primary leaves and all secondary needles.
[0003] Studies have shown that the morphological indicators of secondary leaves of Masson pine seedlings, such as leaf length, volume and surface area, are significantly larger than those of primary leaves. Seedlings with secondary leaves have a larger photosynthetic area, more dry matter accumulation, and stronger drought resistance under severe drought stress conditions. Therefore, the seedling quality of seedlings with all secondary leaves at the time of planting and the growth performance after afforestation are higher than those of seedlings with all primary leaves. Therefore, selecting and cultivating seedlings with all secondary leaves can help improve the growth, drought resistance and seedling quality of seedlings.
[0004] However, the identification of Masson pine seedlings with different leaf types can only be judged by the morphological characteristics of the seedling leaves in September and October. At this time, the seedlings have already taken shape, wasting a lot of seedling resources and making early selection impossible. At the same time, the complete genome data of Masson pine has not yet been published, which limits the development of molecular markers for seedling leaf type identification based on the genome. Summary of the invention
[0005] In view of the above-mentioned deficiencies in the prior art, the present invention provides a method for early prediction of seedlings with different leaf types of Masson pine based on the GS model, as well as SNP and InDel sites. The present invention obtains SNP and InDel marker sites that can be used to identify seedlings with different leaf types of Masson pine, as well as a method for early prediction of seedlings with different leaf types of Masson pine.
[0006] To achieve the above purpose, the technical solution adopted by the present invention to solve the technical problem is: One of the purposes of the present invention is to provide a method for early prediction of Masson pine seedlings with different leaf types based on a GS model, comprising the following steps: (1) The SNP loci and / InDel loci obtained by screening are used as genotype data; the leaf types of all primary leaves and all secondary leaves recorded are used as phenotypic data; (2) Construct training and testing groups based on the 5-fold cross-validation method; (3) In the training population, the genotype data and phenotype data are used to construct a GS model. Then, in the test population, the genotype data and the constructed GS model are used to calculate the breeding value, and the leaf type is determined based on the breeding value.
[0007] Furthermore, the SNP loci include the following locus information: c38467 / f1p3 / 833 transcript has A / C at position 350 bp, A / C at position 168 bp, A / G at position 204 bp, G / A at position 360 bp, C / A at position 330 bp, T / C at position 263 bp, C / G at position 180 bp, A / C at position 623 bp, T / A at position 319 bp, A / C at position 102 bp, and C / G at position 158 bp; The c63972 / f4p2 / 1410 transcript has A / C at position 1159 bp, T / C at position 316 bp, A / G at position 639 bp, T / C at position 1148 bp, and A / G at position 1129 bp; G / C at position 1930 bp in the c54674 / f2p10 / 2357 transcript; T / C at position 1602bp of c52469 / f2p2 / 1960 transcript.
[0008] Furthermore, the InDel site includes the following site information: T / TCCTCC at position 1938 bp, C / CGCCA at position 1896 bp, and T / TTCTGGGTTGTCACGGGCGTCTGC at position 1902 bp in the c54674 / f2p10 / 2357 transcript; T / TCAATA at position 587 bp in the c38467 / f1p3 / 833 transcript; The c73647 / f2p3 / 1321 transcript has GAA / G at position 267bp, T / TTC at position 426bp, GGCA / G at position 264bp, and AAGCAG / A at position 413bp; GT / G at position 1206 bp in the c56918 / f2p0 / 1626 transcript; A / ATACCCCCAAGGG at the 273bp position of the c37900 / f1p0 / 684 transcript; A / AG at the 53bp position of the c64882 / f1p0 / 490 transcript; G / GGAA at the 479bp position of the c74763 / f24p0 / 793 transcript; The c2493 / f4p0 / 479 transcript has GT / G at position 419bp and A / ATC at position 461bp.
[0009] Furthermore, the GS model includes GBLUP, rrBLUP, SVC and RFC models.
[0010] Furthermore, the GS model is a rrBLUP model.
[0011] Furthermore, the process of calculating breeding values recorded in step (3) was repeated 500 times to eliminate sampling errors, and the correlation between the GEBV of the test population and the actual observed values was used as an indicator to evaluate the accuracy of the model.
[0012] Furthermore, the leaf shape determination criteria are: if the breeding value is less than 0.5, the Masson pine seedling is a seedling with all primary leaves; if the breeding value is greater than 0.5, the Masson pine seedling is a seedling with all secondary leaves.
[0013] The second object of the present invention is to provide a SNP locus that can be used for early prediction of different leaf types of Masson pine seedlings, including: c38467 / f1p3 / 833 transcript has A / C at position 350 bp, A / C at position 168 bp, A / G at position 204 bp, G / A at position 360 bp, C / A at position 330 bp, T / C at position 263 bp, C / G at position 180 bp, A / C at position 623 bp, T / A at position 319 bp, A / C at position 102 bp, and C / G at position 158 bp; The c63972 / f4p2 / 1410 transcript has A / C at position 1159 bp, T / C at position 316 bp, A / G at position 639 bp, T / C at position 1148 bp, and A / G at position 1129 bp; G / C at position 1930 bp in the c54674 / f2p10 / 2357 transcript; T / C at position 1602bp of c52469 / f2p2 / 1960 transcript.
[0014] The third object of the present invention is to provide an InDel site that can be used for early prediction of different leaf types of Masson pine seedlings, including: T / TCCTCC at position 1938 bp, C / CGCCA at position 1896 bp, and T / TTCTGGGTTGTCACGGGCGTCTGC at position 1902 bp in the c54674 / f2p10 / 2357 transcript; T / TCAATA at position 587 bp in the c38467 / f1p3 / 833 transcript; The c73647 / f2p3 / 1321 transcript has GAA / G at position 267bp, T / TTC at position 426bp, GGCA / G at position 264bp, and AAGCAG / A at position 413bp; GT / G at position 1206 bp in the c56918 / f2p0 / 1626 transcript; A / ATACCCCCAAGGG at the 273bp position of the c37900 / f1p0 / 684 transcript; A / AG at the 53bp position of the c64882 / f1p0 / 490 transcript; G / GGAA at the 479bp position of the c74763 / f24p0 / 793 transcript; The c2493 / f4p0 / 479 transcript has GT / G at position 419bp and A / ATC at position 461bp.
[0015] The fourth object of the present invention is to provide a method for developing SNP and InDel markers based on Masson pine transcriptome sequencing, the process of which is as follows: (1) Select multiple Masson pine seeds from different sources and without direct genetic relationship, disinfect and germinate them. After the seeds germinate, transplant the germinated seedlings into non-woven nutrient bags. When the seedlings grow to the point where the seedlings with all primary leaves and all secondary leaves can be clearly distinguished, record the leaf shape information; seedlings with all primary leaves are marked as 0, and seedlings with all secondary leaves are marked as 1.
[0016] (2) RNA was extracted from stem and leaf tissues of seedlings with all primary leaves and all secondary leaves in different months. The extracted RNA was mixed and sequenced using PacBio Sequel for the third-generation full-length transcriptome. The third-generation transcriptome data was quality controlled and filtered after being downloaded from the machine. The transcript sequence was processed for redundancy using CD-HIF software to obtain the full-length transcript Unigene. The obtained full-length transcript was used as the reference sequence for GWAS and GS analysis.
[0017] (3) The same number of stem and leaf tissues of seedlings with all primary leaves and all secondary leaves were selected for RNA extraction, and the extracted RNA was subjected to high-throughput second-generation sequencing using the Illumina NovaSeq 6000 platform. After removing reads with adapters, reads containing N (N indicates that the base information cannot be determined), and low-quality reads, the data was processed to obtain high-quality clean data.
[0018] (4) Using GATK4 software, align the clean data obtained in step (3) to the full-length transcript obtained in step (2) to obtain a VCF file storing SNP and InDel genotyping; and use plink and VCFtools software to filter low-quality SNPs and InDels, retaining high-quality SNP and InDel site information for subsequent analysis.
[0019] (5) Use plink and Admixture software to perform principal component and kinship analysis on the high-quality SNPs and InDels described in step (4) to obtain the eigenvector matrix and kinship coefficient matrix of all individuals to control the false positives in GWAS analysis caused by unreasonable population structure.
[0020] (6) The leaf type information of the Masson pine seedlings recorded in step (1) is used as the phenotypic data; the high-quality SNP and InDel site information described in step (4) is used as the genotypic data; when the phenotypic data and genotypic data are missing, the missing values are filled; the phenotypic data and genotypic data are subjected to GWAS analysis using the assoc model of the plink software, and the eigenvector matrix and kinship coefficient matrix described in step (5) are added as covariates to the model analysis using the "--covar" parameter to estimate the significance of each SNP and InDel p Value and phenotypic explanation rate; the obtained association results p SNP loci <0.01 / total number of SNPs and p InDel sites with a number <1 / total number of InDels were considered key sites.
[0021] Beneficial effects of the present invention: The present invention develops SNP and InDel markers that can be used for early identification of leaf types of Masson pine seedlings based on third-generation and second-generation transcriptome data, which can provide an important technical and material basis for the breeding of improved Masson pine seedlings. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] Figure 1 The phenotypes of Masson pine seedlings with all primary leaves and all secondary leaves; Figure 2 The Manhattan plot obtained by performing genome-wide association analysis based on SNP sites in Example 2; Figure 3 This is the Manhattan plot obtained by performing genome-wide association analysis based on InDel sites in Example 2. DETAILED DESCRIPTION
[0023] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.
[0024] Example 1 Development of SNP and InDel polymorphic sites in Masson pine seedlings based on transcriptome sequencing: 1. Experimental materials and transcriptome sequencing methods The seeds used in this experiment were from the 1.5-generation seed garden in the National Masson Pine Seed Base of Ma'anshan Forest Farm in Duyun City, Guizhou Province. The selected seeds were sterilized and placed in an artificial incubator for germination. The conditions were as follows: temperature 25 °C / 20 °C (12 h day / 12 h night), relative humidity 65%, and light radiation 560 μmol m -2 s -1 About 4 weeks after emergence, the seedlings were transplanted into non-woven seedling bags with a substrate mixture ratio of topsoil under the Masson pine forest: nutrient soil: vermiculite (2:1:1, v / v / v, pH 5.3). Subsequently, the container seedlings were evenly placed in the outdoor open-air seedbed of the nursery of the College of Forestry of Guizhou University, ensuring that all seedlings were under the same environmental conditions.
[0025] The experiment was conducted in the nursery of the Forestry College of Guizhou University from March 2018 to December 2022. Three experiments (EXP1, EXP2 and EXP3) were carried out successively, and a total of 112 samples (56 seedlings with all primary leaves and 56 seedlings with all secondary leaves) were selected for the experiment. Among them, EXP1 randomly selected 18 seedlings with all primary leaves and 18 seedlings with secondary leaves that grew well from August to December 2018 for relevant experiments. Total RNA was extracted from the stem and leaf tissues of Masson pine seedlings, and the RNA purity (OD260 / 280 and OD260 / 230 ratios) was tested to see if it met the standards. Then, the RNA of the samples that passed the quality inspection was used to construct the cDNA library. The cDNA library was sequenced by high-throughput second-generation sequencing using the Illumina NovaSeq 6000 platform. After removing reads with adapters, reads containing N (N indicates that the base information cannot be determined), and low-quality reads, the data was processed to obtain high-quality clean data. A total of 32 samples of second-generation sequencing libraries were created in the experiment. In addition, the stem and leaf tissues of primary leaf seedlings and secondary leaf seedlings were selected for RNA extraction in mid-August, September, October, and November, respectively. Mixed samples of stem and leaf tissues of seedlings in the four periods were collected and three-generation full-length transcriptome sequencing was performed using PacBio Sequel. After correction and de-redundancy analysis of the three-generation full-length transcripts, a total of 71,563 transcript sequences were obtained; EXP2 and EXP3 were carried out in September 2021 and September-December 2022, respectively. The selection of all primary leaf seedlings and all secondary leaf seedlings and the second-generation transcriptome sequencing methods were the same as EXP1, generating 18 and 62 samples of second-generation sequencing libraries, respectively.
[0026] 2. SNP and InDel site development and phenotypic and genotypic data preparation The recorded leaf morphological information of Masson pine was used as phenotypic data for analysis. Seedlings with all primary leaves were recorded as 0, and seedlings with all secondary leaves were recorded as 1 (see Figure 1 ).
[0027] Using GATK4 software, 112 second-generation transcriptome sequencing libraries were aligned to 71,563 transcript sequences obtained by third-generation full-length transcriptome sequencing to generate VCF files storing site information. After merging the 112 VCF files, plink and VCFtools software were used to filter genotypes according to the criteria of sequencing depth greater than 6×, minimum allele frequency greater than 0.05, and missing rate less than 40%, and a total of 82,085 and 10,443 SNP and InDel polymorphic sites were obtained. The "--012" parameter of plink software was used to convert the VCF containing 82,085 and 10,443 SNP and inDe sites into a matrix containing only numbers 0, 1, and 2 as genotype data. Among them, 0 represents non-mutation genotype; 1 represents heterozygous mutation genotype; 2 represents homozygous mutation genotype.
[0028] Example 2 Genome-wide association analysis identifies significant loci associated with leaf shape of Masson pine seedlings: 1. Significant SNP loci associated with leaf shape of Masson pine seedlings Based on the 82,085 SNP sites obtained by filtering and the phenotypic values of 0 (seedlings with all primary leaves) and 1 (seedlings with all secondary leaves), the "--assoc" parameter of the plink software, i.e., the joint variance analysis model, was used for association analysis. In addition, the "--covar" parameter was used to add the first three items of the principal component analysis of the genotype data as covariates to the model analysis; the "--adjust" parameter was used to adjust the association results. p The significance threshold p The value was set to 0.01 / total number of SNPs (1.22E-07), and a total of 18 SNP loci associated with leaf type of Masson pine seedlings were detected (see Figure 2 The phenotypic explanation rate ranged from 27.97 to 34.49% (Table 1).
[0029] Table 1 Genome-wide association analysis detected significant SNP loci associated with leaf shape of Masson pine seedlings
[0030] 2. Significant InDel loci related to leaf shape of Masson pine seedlings Based on the 10443 InDel sites obtained by filtering and the phenotypic values of 0 (seedlings with all primary leaves) and 1 (seedlings with all secondary leaves), the “--assoc” parameter of the plink software, i.e., the joint variance analysis model, was used for association analysis. In addition, the “--covar” parameter was used to add the first three items of the principal component analysis of the genotype data as covariates to the model analysis; the “--adjust” parameter was used to adjust the association results. p The significance threshold pThe value was set to 1 / the total number of InDels (9.98E-05), and a total of 14 InDel loci related to the leaf type of Masson pine seedlings were detected (see Figure 3 The phenotypic explanation rate ranged from 15.18% to 31.30% (Table 2).
[0031] Table 2 Significant InDel loci associated with leaf shape of Masson pine seedlings detected by genome-wide association analysis
[0032] Example 3 Selection of the optimal whole genome prediction model: 1. Selection of whole genome selection model and evaluation indicators GBLUP and rrBLUP based on G matrix, as well as support vector machine (SVC) and random forest (RFC) for diquartile traits were selected. GBLUP and rrBLUP were implemented by R packages "BGLR" and "rrBLUP", respectively, and SVC and RFC were implemented by R packages "randomForest" and "e1071", respectively. A 5-fold cross-validation method was used, with 90 (80%) Masson pine seedlings as the training population and 22 (20%) Masson pine seedlings as the test population. The selection model was established using genotype data and phenotypic data in the training population, and the genomic estimated breeding value (GEBV) was estimated using genotype data and prediction models in the test population. This was repeated 500 times to eliminate sampling errors, and the GEBV of the test population was correlated with the actual observed value (R 2 ) is used as an indicator to evaluate the accuracy of the model.
[0033] 2. Genome-wide selection analysis based on SNP loci The 18 key SNPs obtained in Example 2 were subjected to GS modeling analysis, and the accuracy of the model was between 0.634 and 0.679. Among them, the accuracy of the rrBLUP model was the highest, reaching 0.679. p The top 25, 50, 100, 200, 500, 1000, 2000, 5000 and 10000 SNP loci were selected in descending order and the above four GS models were used for modeling analysis. As shown in Table 3, the prediction accuracy (R 2) was significantly higher than the RFC and SVC models based on machine learning methods. As the marker density gradually increased from 25 to 10,000, the accuracy of the four models showed a trend of first increasing and then decreasing. Among them, the accuracy of the GBLUP, rrBLUP, RFC and SVC models ranged from 0.765 to 0.946, 0.705 to 0.956, 0.612 to 0.819 and 0.666 to 0.953, respectively, reaching the maximum value when the SNP density was 1,000, 5,000, 2,000 and 2,000, respectively. The rrBLUP model had the highest prediction accuracy at 5,000 SNPs, reaching 0.956.
[0034] Table 3 GS prediction accuracy of different models and SNP densities
[0035] 3. Genome-wide selection analysis based on InDel sites The 14 key InDels obtained in Example 2 were subjected to GS modeling analysis, and the accuracy of the model was between 0.604 and 0.816. Among them, the accuracy of the rrBLUP model was the highest, reaching 0.816. p The top 25, 50, 100, 200, 500, 1000, 2000, 5000 and 10000 InDel sites were selected in order of value from small to large, and the above four GS models were used for modeling analysis. As shown in Table 4, as the marker density gradually increased from 25 to 10000, the accuracy of the GBLUP, rrBLUP and SVC models showed a trend of first increasing and then decreasing, while the accuracy of the RFC model showed a trend of gradually decreasing. The accuracy of the GBLUP, rrBLUP, RFC and SVC models ranged from 0.813 to 0.915, 0.782 to 0.944, 0.749 to 0.897 and 0.641 to 0.937, respectively, and reached the maximum value when the InDel density was 500, 5000, 25 and 2000, respectively. The rrBLUP model had the highest prediction accuracy at 5000 SNPs, which could reach 0.944. In summary, when the number of marker loci is less than or equal to 100, the accuracy of the GS analysis model using InDel loci is higher than that of SNP loci. On the contrary, when the number of marker loci is greater than or equal to 200, the accuracy of the GS analysis model using SNP loci is higher than that of InDel loci. In summary, the rrBLUP model is the optimal model for predicting the leaf type of Masson pine seedlings. The rrBLUP model and the set of 5000 SNP loci are the optimal system for predicting the leaf type of Masson pine seedlings.
[0036] Table 4 GS prediction accuracy of different models and InDel densities
[0037] Example 4 Prediction of leaf shape of Masson pine seedlings based on the optimal whole genome selection system: According to the optimal prediction system (rrBLUP-5000) selected in Example 3, the GEBV of 22 Masson pine seedlings in 5 different test populations was predicted using a 5-fold crossover method. For the GBLUP and rrBLUP models, Masson pine seedlings with a GEBV value less than 0.5 were recorded as 0 (i.e., all primary leaf seedlings), and Masson pine seedlings with a GEBV value greater than 0.5 were recorded as 1 (i.e., all secondary leaf seedlings).
[0038] RFC and SVC models can be used for association analysis of binary traits (either 0 or 1). In the prediction results, a GEBV value of 0 means a seedling with all primary leaves, and a GEBV value of 1 means a seedling with all secondary leaves. The cor() function was used to perform correlation analysis on the GEBV values of the 22 seedlings in the prediction population and the true breeding values. The correlation between the rrBLUP model and the GS model of 5000 SNPs was 0.956 ( p <0.0001), which can be used to predict the leaf shape of Masson pine seedlings.
[0039] Finally, it should be noted that the above specific implementation methods are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to examples, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.
Claims
1. A method for early prediction of different leaf types of Masson pine seedlings based on the GS model, characterized in that: The following steps are involved: (1) The SNP loci and / InDel loci obtained by screening are used as genotype data; the leaf types of all primary leaves and all secondary leaves recorded are used as phenotypic data; (2) Construct training and testing groups based on the 5-fold cross-validation method; (3) In the training population, the genotype data and phenotype data are used to construct a GS model. Then, in the test population, the genotype data and the constructed GS model are used to calculate the breeding value, and the leaf type is determined based on the breeding value.
2. The method according to claim 1, characterized in that The SNP site includes the following site information: c38467 / f1p3 / 833 transcript has A / C at position 350 bp, A / C at position 168 bp, A / G at position 204 bp, G / A at position 360 bp, C / A at position 330 bp, T / C at position 263 bp, C / G at position 180 bp, A / C at position 623 bp, T / A at position 319 bp, A / C at position 102 bp, and C / G at position 158 bp; The c63972 / f4p2 / 1410 transcript has A / C at position 1159 bp, T / C at position 316 bp, A / G at position 639 bp, T / C at position 1148 bp, and A / G at position 1129 bp; G / C at position 1930 bp in the c54674 / f2p10 / 2357 transcript; T / C at position 1602bp of c52469 / f2p2 / 1960 transcript.
3. The method according to claim 1, characterized in that: The InDel site includes the following site information: T / TCCTCC at position 1938 bp, C / CGCCA at position 1896 bp, and T / TTCTGGGTTGTCACGGGCGTCTGC at position 1902 bp in the c54674 / f2p10 / 2357 transcript; T / TCAATA at position 587 bp in the c38467 / f1p3 / 833 transcript; The c73647 / f2p3 / 1321 transcript has GAA / G at position 267bp, T / TTC at position 426bp, GGCA / G at position 264bp, and AAGCAG / A at position 413bp; GT / G at position 1206 bp in the c56918 / f2p0 / 1626 transcript; A / ATACCCCCAAGGG at position 273 bp in the c37900 / f1p0 / 684 transcript; A / AG at position 53 bp in the c64882 / f1p0 / 490 transcript; G / GGAA at position 479 bp in the c74763 / f24p0 / 793 transcript; The c2493 / f4p0 / 479 transcript has GT / G at position 419bp and A / ATC at position 461bp.
4. The method according to claim 1, characterized in that GS models include GBLUP, rrBLUP, SVC and RFC models.
5. The method according to claim 4, characterized in that The GS model is the rrBLUP model.
6. The method according to any one of claims 1 to 5, characterized in that The process of calculating breeding values described in step (3) was repeated 500 times to eliminate sampling errors, and the correlation between the GEBV of the test population and the actual observed values was used as an indicator to evaluate the accuracy of the model.
7. The method according to claim 1, characterized in that The leaf type determination criteria are: if the breeding value is less than 0.5, then the Masson pine seedling is a seedling with all primary leaves; if the breeding value is greater than 0.5, then the Masson pine seedling is a seedling with all secondary leaves.
8. A SNP locus that can be used for early prediction of different leaf types of Masson pine seedlings, characterized in that: include: c38467 / f1p3 / 833 transcript has A / C at position 350 bp, A / C at position 168 bp, A / G at position 204 bp, G / A at position 360 bp, C / A at position 330 bp, T / C at position 263 bp, C / G at position 180 bp, A / C at position 623 bp, T / A at position 319 bp, A / C at position 102 bp, and C / G at position 158 bp; The c63972 / f4p2 / 1410 transcript has A / C at position 1159 bp, T / C at position 316 bp, A / G at position 639 bp, T / C at position 1148 bp, and A / G at position 1129 bp; G / C at position 1930 bp in the c54674 / f2p10 / 2357 transcript; T / C at position 1602bp of c52469 / f2p2 / 1960 transcript.
9. An InDel site that can be used for early prediction of different leaf types of Masson pine seedlings, characterized in that: include: T / TCCTCC at position 1938 bp, C / CGCCA at position 1896 bp, and T / TTCTGGGTTGTCACGGGCGTCTGC at position 1902 bp in the c54674 / f2p10 / 2357 transcript; T / TCAATA at position 587 bp in the c38467 / f1p3 / 833 transcript; The c73647 / f2p3 / 1321 transcript has GAA / G at position 267bp, T / TTC at position 426bp, GGCA / G at position 264bp, and AAGCAG / A at position 413bp; GT / G at position 1206 bp in the c56918 / f2p0 / 1626 transcript; A / ATACCCCCAAGGG at the 273bp position of the c37900 / f1p0 / 684 transcript; A / AG at the 53bp position of the c64882 / f1p0 / 490 transcript; G / GGAA at the 479bp position of the c74763 / f24p0 / 793 transcript; The c2493 / f4p0 / 479 transcript has GT / G at position 419bp and A / ATC at position 461bp.