SNP markers associated with seed morphology in Tabebuia chrysantha and their application

Abstract

The present invention discloses SNP markers related to the seed morphology of Tabebuia chrysantha and its application. The SNP markers include SNP1, SNP4, SNP8, SNP18 and SNP24, and the SNP molecular markers are used to type the seed morphological traits of Tabebuia chrysantha; the seed morphological traits are at least one of winged length, winged width, seed length and seed width. The application is to use the SNP markers to type the seed morphological traits of Tabebuia chrysantha, or to predict the expression levels of seed morphology-related genes in Tabebuia chrysantha. The present invention mines SNP markers related to seed morphology from Tabebuia chrysantha genomic DNA through whole-genome association analysis technology, providing a scientific basis for the genetic improvement, classification and breeding of Tabebuia chrysantha.

Description

Technical Field

[0001] The present invention relates to the technical field of molecular markers of Tabebuia chrysantha, and in particular to SNP markers related to seed morphology of Tabebuia chrysantha and applications thereof. Background Art

[0002] The morphological differences between some species of Tabebuia are small, which can easily lead to species confusion, and its classification has always been controversial. Handroanthus chrysanthus ) There are few reports on genetic relationships and taxonomic studies, especially in the systematic organization of the germplasm resources of Tabebuia chrysantha and the disclosure of its genetic diversity.

[0003] Seeds are important reproductive organs of plants, and their morphological characteristics are of great significance in taxonomy. Although there have been many successful cases of using seed morphology for classification in plant taxonomy at home and abroad, there is still a lack of research on the classification of P. tazetta seed morphology.

[0004] Traditional plant classification methods based on seed morphology rely on morphological observation and measurement, making it difficult to achieve efficient and accurate classification and genetic analysis. Therefore, there is an urgent need to develop a classification method based on molecular markers that can link the seed morphology of Tabebuia chrysantha with SNP markers to achieve genotype-based differentiation of seed morphological traits. Summary of the Invention

[0005] To this end, the technical problem to be solved by the present invention is to provide a SNP marker related to the seed morphology of Tabebuia chrysantha and its application. Through the whole genome association analysis technology, SNP markers related to seed morphology were mined from the genomic DNA of Tabebuia chrysantha, realizing the genotype-based differentiation of seed morphological traits, providing a scientific basis for the genetic improvement, classification and breeding of Tabebuia chrysantha.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] SNP markers related to the seed morphology of the yellow bell tree, the SNP markers include SNP1, SNP4, SNP8, SNP18 and SNP24, and the SNP molecular markers are used to identify the seed morphology of the yellow bell tree. Handroanthus chrysanthus The seed morphological traits are classified; the seed morphological traits are at least one of winged length, winged width, seed length and seed width; wherein:

[0008] SNP1 is located at 3975 bp on chromosome NKXS01000048.1, the reference base is G, and the mutated base is T;

[0009] SNP4 is located at 4051 bp on chromosome NKXS01000048.1, the reference base is T, and the mutated base is C;

[0010] SNP8 is located at 3134 bp on chromosome NKXS01002837.1, with a reference base of T and a mutated base of G;

[0011] SNP18 is located at 11016 bp on chromosome NKXS01004877.1, with a reference base of C and a mutated base of T;

[0012] SNP24 is located at 92361 bp on chromosome NKXS01001449.1, with a reference base of G and a mutated base of C. The positions of each SNP on the chromosome were calculated based on the genome data of P. tazetta.

[0013] The application of SNP markers related to the seed morphology of Tabebuia chrysantha, wherein the application is to use the above-mentioned SNP markers to type the seed morphological traits of Tabebuia chrysantha; or, the application is to use the above-mentioned SNP markers to predict the expression levels of key genes related to seed morphology in Tabebuia chrysantha.

[0014] In the above application, when the SNP marker is used to type the seed morphological traits of Tabebuia chrysantha, the genotype of the Tabebuia chrysantha at the site where at least one SNP marker among SNP1, SNP4, SNP8, SNP18 and SNP24 is located is obtained from the genomic DNA of the Tabebuia chrysantha.

[0015] In the above application, when the SNP marker is used to type the seed length, the genotype of the tested Tabebuia chrysantha at the site where SNP4 or SNP8 is located is obtained from the genomic DNA of the tested Tabebuia chrysantha; when the genotype at the site where SNP4 is located is C, or when the genotype at the site where SNP8 is located is G, the seeds of the tested Tabebuia chrysantha are of long type; when the genotype at the site where SNP4 is located is T, or when the genotype at the site where SNP8 is located is T, the seeds of the tested Tabebuia chrysantha are of short type.

[0016] For a tested Tabebuia chrysantha, as long as one of the alleles at the SNP4 locus is the mutant type C, the genotype at the SNP4 locus is considered to be C, i.e., the mutant type. If all the alleles at the SNP4 locus in the tested Tabebuia chrysantha genome are the reference type T, the genotype at the SNP4 locus is considered to be T, i.e., the reference type. The same applies to other SNPs.

[0017] In the above application, when the SNP marker is used to type the seed width, the genotype of the site SNP18 of the tested Tabebuia chrysantha is obtained from the genomic DNA of the tested Tabebuia chrysantha; when the genotype of the site SNP18 is T, the seeds of the tested Tabebuia chrysantha are of wide type; when the genotype of the site SNP18 is C, the seeds of the tested Tabebuia chrysantha are of narrow type.

[0018] In the above application, when the SNP marker is used to type the wing length, the genotype of the yellow-flowered trumpet tree to be tested at the site where SNP24 is located is obtained from the genomic DNA of the yellow-flowered trumpet tree to be tested; when the genotype at the site where SNP24 is located is C, the seeds of the yellow-flowered trumpet tree to be tested are long-winged type, and when the genotype at the site where SNP24 is located is G, the seeds of the yellow-flowered trumpet tree to be tested are short-winged type.

[0019] In the above application, when the SNP marker is used to type the wing width, the genotype of the yellow-flowered Tabebuia to be tested at the site where SNP1 is located is obtained from the genomic DNA of the yellow-flowered Tabebuia to be tested; when the genotype at the site where SNP1 is located is T, the seeds of the yellow-flowered Tabebuia to be tested are wide-winged type, and when the genotype at the site where SNP1 is located is G, the seeds of the yellow-flowered Tabebuia to be tested are narrow-winged type.

[0020] In the above application, when the SNP marker is used to simultaneously genotype seed length, seed width, and wing width, the genotype of the tested Tabebuia chrysantha at the site of SNP1 or SNP4 is obtained from the genomic DNA of the tested Tabebuia chrysantha;

[0021] When the genotype of the SNP1 site is T, the seeds of the tested Tabebuia chrysantha are long, wide, and have wide wings. When the genotype of the SNP1 site is G, the seeds of the tested Tabebuia chrysantha are short, narrow, and have narrow wings.

[0022] Alternatively, when the genotype of the site where SNP4 is located is C, the seeds of the tested Tabebuia chrysantha are long, wide and have wide wings; when the genotype of the site where SNP4 is located is T, the seeds of the tested Tabebuia chrysantha are short, narrow and have narrow wings.

[0023] In the above application, the key gene related to seed morphology in Tabebuia chrysantha is CDL12_15522;

[0024] When predicting the expression level of the CDL12_15522 gene, the genotype of the tested Tabebuia chrysantha at the SNP8 site is obtained from the genomic DNA of the tested Tabebuia chrysantha; when the genotype of the SNP8 site is G, the CDL12_15522 gene is highly expressed; when the genotype of the SNP8 site is T, the CDL12_15522 gene is lowly expressed.

[0025] In the above application, when obtaining the genotype of the tested Tabebuia chrysantha at the site where any SNP marker among SNP1, SNP4, SNP8, SNP18 and SNP24 is located, PCR is performed using the genomic DNA of the tested Tabebuia chrysantha as the template DNA, and the PCR product is sequenced to obtain the genotype of the tested Tabebuia chrysantha at the site where the corresponding SNP marker is located.

[0026] The CDL12_15522 gene is located in the flanking region of SNP8, and the two are tightly linked. Experiments have shown that different genotypes at SNP8 result in different expression levels of the CDL12_15522 gene. This suggests that by measuring the expression level of the CDL12_15522 gene, seed morphology in Tabebuia chrysantha can be predicted.

[0027] The technical solution of the present invention achieves the following beneficial technical effects:

[0028] 1. This study, for the first time, identifies single-nucleotide polymorphism (SNP) markers associated with seed morphology in Tabebuia chrysantha, providing an important foundation for studying the molecular genetic mechanisms underlying seed morphology. These SNP markers facilitate rapid and accurate identification of seed morphological characteristics, which in turn provides a foundation for taxonomic research on the plant.

[0029] 2. The present invention also screened candidate genes related to the seed morphology of Tabebuia chrysantha. Among them, the CDL12_15522 gene is located in the flanking region of the SNP marker related to the seed morphology of Tabebuia chrysantha. qRT-PCR experiments showed that among Tabebuia chrysantha individuals with large differences in seed morphology, the expression levels of the candidate gene CDL12_15522 also differed greatly, that is, CDL12_15522 may have the function of regulating the seed morphology of Tabebuia chrysantha, providing new clues for the study of genes regulating the seed morphology of Tabebuia chrysantha. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 Frequency distribution diagram of seed length measurement results of 126 Tabebuia chrysantha individuals in an embodiment of the present invention;

[0031] Figure 2 Frequency distribution diagram of seed width measurement results of 126 Tabebuia chrysantha individuals in an embodiment of the present invention;

[0032] Figure 3 Frequency distribution diagram of the wing length measurement results of 126 individuals of Tabebuia chrysantha in the embodiment of the present invention;

[0033] Figure 4 Frequency distribution diagram of the wing width measurement results of 126 individuals of Tabebuia chrysantha in the embodiment of the present invention;

[0034] Figure 5 A graph showing the correlation analysis results among seed length, seed width, winged length, and winged width of Tabebuia vulgaris in an embodiment of the present invention;

[0035] Figure 6 The optimal K value analysis results of a group of 126 Tabebuia chrysantha in the embodiment of the present invention;

[0036] Figure 7 A histogram of the genetic composition of samples consisting of 126 Tabebuia chrysantha in an embodiment of the present invention;

[0037] Figure 8 A neighbor-joining phylogenetic tree of 126 individuals of Tabebuia chrysantha in the embodiment of the present invention;

[0038] Figure 9 Principal component analysis distribution diagram of 126 Tabebuia chrysantha individuals in the embodiment of the present invention;

[0039] Figure 10 QQ plots of four models of seed length of 126 Tabebuia chrysantha individuals in the embodiment of the present invention;

[0040] Figure 11 QQ plots of four models of seed width of 126 Tabebuia chrysantha individuals in the embodiment of the present invention;

[0041] Figure 12 QQ plots of four models of wing length of 126 Tabebuia chrysantha individuals in the embodiment of the present invention;

[0042] Figure 13 QQ plots of four models of wing width of 126 Tabebuia chrysantha individuals in the embodiment of the present invention;

[0043] Figure 14 Manhattan plot of the seed length GWAS analysis results of 126 Tabebuia chrysantha individuals in the embodiment of the present invention;

[0044] Figure 15 The results of the GWAS analysis of seed width of 126 individuals of Tabebuia chrysantha in the embodiment of the present invention;

[0045] Figure 16 The GWAS analysis results of wing length of 126 individuals of Tabebuia chrysantha in the embodiment of the present invention;

[0046] Figure 17 The results of GWAS analysis of wing width of 126 individuals of Tabebuia chrysantha in the embodiment of the present invention;

[0047] Figure 18QQ plot of the seed length MLM (QK) model of 126 Tabebuia chrysantha individuals in the embodiment of the present invention;

[0048] Figure 19 QQ plot of the seed width MLM (QK) model of 126 Tabebuia chrysantha individuals in the embodiment of the present invention;

[0049] Figure 20 QQ plot of the MLM (QK) model of wing length of 126 individuals of Tabebuia chrysantha in the embodiment of the present invention;

[0050] Figure 21 QQ plot of the wing width MLM (QK) model of 126 Tabebuia chrysantha individuals in the embodiment of the present invention;

[0051] Figure 22 Venn diagram between SNPs controlling different seed morphological traits in the embodiment of the present invention;

[0052] Figure 23 Venn diagram of candidate genes within the flanking regions of each SNP in the embodiments of the present invention;

[0053] Figure 24 A comparison chart of the expression levels of nine key candidate genes in four Tabebuia chrysantha individuals, F1, F4, M1, and M6, in the examples of the present invention. DETAILED DESCRIPTION

[0054] 1. Materials and Methods

[0055] 1.1 Experimental Materials

[0056] A total of 126 accessions of Tabebuia chrysantha were collected from 12 regions in Guangdong Province, including Zhanjiang, Maoming, Yangjiang, Zhaoqing, and Jiangmen, as the associated populations. Detailed information on the associated populations is shown in Table 1.

[0057] Table 1 Geographic information of 126 Tabebuia chrysantha samples

[0058]

[0059] 1.2 Phenotypic data determination

[0060] At seed maturity, 200 plump, healthy seeds were collected from each of the 126 accessions (126 Tabebuia chrysantha plants) described above. Six seeds were randomly selected from these 200 seeds for seed morphological characterization. Tabebuia chrysantha seeds have thin, filmy wings surrounding them. Four parameters were measured using a vernier caliper: wing length (WL, corresponding to long / short wing genotypes), wing width (WW, corresponding to wide / narrow wing genotypes), seed length (NWL, the length of the seed without wings, corresponding to long / short genotypes), and seed width (NWW, the width of the seed without wings, corresponding to wide / narrow genotypes). Three independent measurements were performed for each seed, and the average of the measurements was used as the final data. R software (version 3.5.0) was used to perform relevant statistical analysis on the measured phenotypic data, and further analysis of variance and correlation analysis were performed on the phenotypic data.

[0061] 1.3 Population structure and kinship analysis

[0062] Since the whole genome of the yellow-flowered trumpet tree has not been sequenced, the only trumpet tree species is the red-flowered trumpet tree ( Handroanthus impetiginosus ) underwent whole-genome sequencing. Therefore, molecular characterizations (SNPs, transcriptome, etc.) of Tabebuia species are based on the Tabebuia rubra genome. 126 Tabebuia rubra individuals were genotyped using GBS (Genotyping-by-Sequencing). Specifically, high-throughput sequencing was performed on 126 Tabebuia rubra individuals. The resulting raw image data files were converted into raw sequenced sequences through base calling analysis. After quality control of the raw sequences and filtering of low-quality sequences, high-quality sequences were obtained. These high-quality sequences were then aligned with the published Tabebuia rubra reference genome to identify genome-wide SNP markers.

[0063] Population structure analysis was performed using Admixture software (version 1.3) based on genome-wide SNP markers filtered for linkage disequilibrium (LD). To determine the optimal number of subpopulations (K value), cluster analysis was performed assuming a range of 1 to 9 subpopulations. The K value that minimized the cross-validation error (CV error) was used as the optimal number of subpopulations. Subsequently, Admixture software was run again. Based on the selected optimal K value and the SNP markers filtered for linkage disequilibrium (LD), Admixture calculated the genetic composition of each sample and determined the proportion of genetic material inherited from each subpopulation (i.e., the proportion of genetic material inherited by the sample from each subpopulation).

[0064] Admixture software will output a population structure matrix (Q matrix). Each row of the Q matrix represents a sample, each column represents a subpopulation, and the elements in the matrix (i.e., Q values) represent the proportion of genetic material inherited by the sample from the corresponding subpopulation.

[0065] Finally, pophelper software (version 2.2.7) was used to draw a histogram of the genetic composition of each sample in each subpopulation based on the output results of Admixture.

[0066] Using the selected SNP markers (i.e., the genome-wide SNP markers described above), a phylogenetic tree based on the neighbor-joining (NJ) method was constructed using MEGA-X software (model: p-distance; bootstrap: 500). Furthermore, principal component analysis (PCA) was performed using GCTA software (version 1.93.2) to obtain the variance explained by each principal component (PC) and the sample score matrix for each PC. Finally, kinship analysis was performed using TASSEL 5.0 software to generate a kinship matrix between each sample.

[0067] 1.4 Genome-wide association analysis

[0068] Using the genome-wide SNP molecular markers obtained after alignment with the reference genome of the red bell tree, SNP sites with a minimum allele frequency (MAF) ≥ 0.05 were used for genome-wide association analysis (GWAS analysis). GEMMA software (version 0.98.1) was used for GWAS analysis. Common models used in GWAS analysis include simple generalized linear models (GLM), generalized linear models with Q matrix as covariates [GLM(Q)], mixed linear models with K matrix as covariates [MLM(K)], and mixed linear models with Q matrix and K matrix as covariates [MLM(QK)]. In this embodiment, the population structure matrix corresponding to the optimal K value obtained by Admixture software analysis in "1.3 Population Structure and Kinship Analysis" was used as the Q matrix, and the kinship matrix between samples was used as the K matrix. GWAS analysis was performed on various sub-morphological related traits using the above four models, and the distribution of actual P values ​​under different models was compared with the distribution of theoretical P values ​​by QQ plot to determine the optimal model.

[0069] After determining the optimal model, the Bonferroni multiple test correction method was used to determine the significance threshold of the P value, and the significantly associated regions were screened out. The SNPs with the strongest association signals (i.e., topassociated SNPs) were further screened from the significantly associated regions for subsequent association site analysis.

[0070] 1.5 Analysis of candidate genes at related loci

[0071] After performing genome-wide association analysis, the published Tabebuia serrata genome database (https: / / www.ncbi.nlm.nih.gov / datasets / taxonomy / 429701 / ) was used in combination with BLAST software for comparison to screen SNP markers that were significantly associated with seed morphological traits. -10 Under the conditions of , significantly associated SNP markers were mapped against the Tabebuia chinensis genome. Specifically, the significantly associated SNP markers were located on the Tabebuia chinensis genome, and related genes were searched within the 50 kb upstream and downstream flanking regions to identify candidate genes. GO and KEGG analyses were performed on the candidate genes. Functional annotation was performed for SNPs and other variants within the coding regions of the candidate genes.

[0072] 1.6 qRT-PCR validation

[0073] Leaves from four individuals (F1, F4, M1, and M6) with significant differences in seed morphology phenotypes were selected (F and M represent location numbers, and numbers represent individual numbers). RNA was extracted using an RNA extraction kit (TaKaRa MiniBEST Universal RNA Extraction Kit), and RNA was transcribed into cDNA using a reverse transcription kit (TUREscript 1st Stand cDNA SYNTHESIS Kit). Primers were designed for the candidate genes mentioned above and qRT-PCR was performed using an ABI 7500 real-time quantitative PCR instrument. Three technical replicates were set for each sample (i.e., each individual). 2 −ΔΔCt Methods The relative expression levels of candidate genes were calculated. 18S gene was used as the internal reference gene in qRT-PCR detection.

[0074] 2. Results and Analysis

[0075] 2.1 Sequencing quality

[0076] High-throughput sequencing of 126 individuals of the plant family Tabebuia chrysantha was performed using GBS technology. The raw image data files were converted into raw sequence sequences through base calling analysis. After quality control of the raw sequences and filtering of low-quality sequences, the number of high-quality sequences ranged from 4,677,442 to 2,378,244. After quality control, the Q30 values ​​of the high-quality sequences ranged from 87.90% to 94.13%, the Q20 values ​​ranged from 95.01% to 98.08%, and the average GC content was 38.19%. The alignment rates of the population samples to the Tabebuia chrysantha reference genome ranged from 81.04% to 95.11%, with an average alignment rate of 88.65%. These results indicate that the samples used in this experiment have high similarity to the reference genome (Tabebuia chrysantha genome) and that the sequencing quality is high. After filtering out low-quality sequences (locus sites), 131,559 high-quality SNPs were retained for subsequent analysis.

[0077] 2.2 Phenotypic analysis

[0078] Descriptive statistical analysis was performed on four seed morphology-related traits in the associated group of 126 individuals of P. fulva (see Table 2). The analysis results showed that the various traits showed extremely significant differences among different individuals, and the variation was small. Specifically, the variation range of seed length was 8.14 mm to 15.98 mm, with an average of 11.96 mm and a coefficient of variation of 7.13%; the variation range of seed width was 5.51 mm to 10.00 mm, with an average of 7.77 mm and a coefficient of variation of 8.94%; the variation range of wing length was 20.10 mm to 31.66 mm, with an average of 26.52 mm and a coefficient of variation of 10.94%; the variation range of wing width was 6.60 mm to 11.11 mm, with an average of 8.86 mm and a coefficient of variation of 9.41%. Further analysis found that there was an extremely significant correlation between the four traits NWL, NWW, WL and WW (see Figure 5 ).like Figure 1 、 Figure 2 、 Figure 3 and Figure 4 These are the frequency distribution diagrams of the four traits NWL, NWW, WL and WW. The unit of the horizontal axis in the diagram is mm. It can be seen from the trait frequency distribution diagram that these four seed morphological traits show a continuous normal distribution, which indicates that these traits are quantitative traits controlled by micro-effect polygenes and are suitable for whole-genome association analysis.

[0079] Table 2 Statistics and variance analysis of seed morphological traits of 126 Tabebuia chrysantha plants

[0080]

[0081] 2.3 Population structure and kinship

[0082] Before conducting the whole-genome association analysis, the population structure and phylogenetic relationships among the population materials were analyzed using 131,559 SNP markers that densely covered the whole genome of P. tazetta (i.e., the SNP markers finally screened in “2.1 Sequencing Quality”). Figure 6 This is the result of population structure analysis. As can be seen from the figure, when K=2, the ΔK value is the smallest, so the 126 samples are finally divided into two subgroups. Among them, one subgroup contains 113 samples, and the other smaller subgroup contains only 13 samples. Figure 7 The results of the neighbor-joining evolutionary tree construction are consistent with the population structure analysis, and the 126 samples are divided into two groups (see Figure 8 The results of principal component analysis (PCA) showed that the first two principal components explained 10.03% and 2.43% of the genetic variance, and the 126 samples of Tabebuia chrysantha were divided into two subgroups (see Figure 9 ), reflecting a certain degree of genetic differentiation between the two populations. Therefore, in the subsequent association analysis between seed morphological traits and SNP markers, the Q matrix generated when K = 2 was used as population structure information.

[0083] In addition, Table 3 shows the results of kinship analysis.

[0084] Table 3 Statistical analysis of kinship among samples

[0085]

[0086] As shown in the table, among the 126 Tabebuia chrysantha individuals, 80.30% had relatedness coefficients ranging from 0 to 0.2, indicating that the vast majority of individuals were distantly related. 17.45% had relatedness coefficients ranging from -0.6 to -0.4, indicating that some individuals were weakly related. Furthermore, 2.22% had relatedness coefficients greater than 0.6 or less than -0.6, indicating that a small number of individuals were highly similar and closely related. Overall, among the 126 Tabebuia chrysantha individuals, very few showed high similarity, while the majority were distantly related or even nonexistent. This result meets the requirements for conducting genome-wide association studies.

[0087] 2.4 Selection of association analysis model

[0088] In this example, four models, namely GLM, GLM(Q), MLM(K) and MLM(QK), were used to perform genome-wide association analysis on seed morphology-related traits. Figure 10 、 Figure 11 、 Figure 12 and Figure 13 The following are QQ plots for the four traits NWL, NWW, WL, and WW. Analysis of the QQ plots for each trait reveals that the GLM and MLM(K) models have poor control of false positives for NWL, NWW, WL, and WW. However, the GLM(Q) and MLM(QK) models, after incorporating population structure and inter-accession kinship, offer better control of false positives, although the control is slightly more stringent for some traits. Further comparison revealed that the -log(P) value distribution of the MLM(QK) model was closest to the predicted value (appearing closest to the dashed line in the figure). Therefore, the MLM(QK) model was selected as the final model for the genome-wide association analysis, used to locate significant association loci and conduct subsequent analyses.

[0089] 2.5 Association analysis between SNP markers and phenotypic traits

[0090] P≤5.0×10⁻ 5 As the significant association threshold, GWAS analysis was performed on the phenotypes (observed values) of each trait in 126 Tabebuia chrysantha materials and SNP markers. Figure 14 、 Figure 15 、 Figure 16 and Figure 17 These are Manhattan plots of the GWAS analysis results for the four traits NWL, NWW, WL, and WW. The red dotted line in the figure represents the significance threshold, and the blue dotted line represents the recommended significance level. Figure 18 、 Figure 19 、 Figure 20 and Figure 21 These are the QQplot diagrams of the MLM (QK) model of the GWAS analysis of the four traits NWL, NWW, WL and WW. The red oblique line in the figure represents the diagonal line with expected -log10 (p) as the horizontal and vertical coordinates. When the point deviates upward from the red diagonal line, it indicates that there is a non-random correlation between the site and the trait. Points that exceed a given threshold in the non-random correlation (Manhattan plot) can be considered as significantly correlated sites.

[0091] As shown in Tables 4-1 to 4-3, a total of 29 significant SNPs were detected in the association analysis with four seed morphology-related traits (including 9 pleiotropic SNPs. "-" in the table indicates that the sequencing results had many noise peaks, making it difficult to accurately determine the base at that site). The phenotypic variation (R²) explained by a single associated marker site ranged from 6.61% to 95.05%. Among them, 14 sites were significantly associated with seed length (NWL); 16 sites were significantly associated with seed width (NWW); 2 sites were significantly associated with wing length (WL); and 9 sites were significantly associated with wing width (WW). Among them:

[0092] The most significantly associated SNP with NWL was NKXS01000048.1__4051, located at 4051 bp of NKXS01000048.1, with a P value of 8.66E-08, explaining 31.92% of the phenotypic variation. The reference base at NKXS01000048.1__4051 was T, and the mutant base was C.

[0093] The most significantly associated SNP with NWW was NKXS01004877.1__11016, located at bp 11016 of NKXS01004877.1, with a P-value of 7.12E-06, explaining 10.64% of the phenotypic variation. The reference base at NKXS01004877.1__11016 was C, and the mutant base was T.

[0094] The most significantly associated SNP with WL was NKXS01001449.1__92361, located at 92,361 bp of NKXS01001449.1, with a P value of 1.77E-05, explaining 26.95% of the phenotypic variation. The reference base at NKXS01001449.1__92361 was G, and the mutant base was C.

[0095] The most significantly associated SNP with WW was NKXS01000048.1__3975, located at bp 3975 of NKXS01000048.1, with a P-value of 1.93E-06, explaining 29.38% of the phenotypic variation. The reference base at NKXS01000048.1__3975 was G, and the mutant base was T.

[0096] By comparing the associated SNPs of four seed morphology-related traits, 9 pleiotropic SNPs were found. Among them, 3 pleiotropic SNPs were found between NWL, NWW and WW; 4 pleiotropic SNPs were found between NWL and NWW; 1 pleiotropic SNP was found between NWL and WW; and 1 pleiotropic SNP was found between NWW and WW ( Figure 22 After Bonferroni multiple-test correction, two SNPs were found to be significantly associated with NWL: NKXS01000048.1__4051 and NKXS01002837.1__3134. These two loci showed strong association signals with NWL and may be the main loci regulating this trait.

[0097] Table 4-1 SNP sites significantly associated with seed morphological traits NWL detected in association analysis

[0098]

[0099] Table 4-2 SNP sites significantly associated with seed morphological traits NWW detected in association analysis

[0100]

[0101] Table 4-3 SNP sites significantly associated with seed morphological traits WL and WW detected in association analysis

[0102]

[0103] SNP1 and SNP4 are located on the same chromosome, with a physical distance of 76 base pairs between them. We further analyzed the relationship between the genotypes at SNP1 and SNP4 and the NWL, NWW, and WW measurements for selected individuals. The results showed that the mean NWL, NWW, and WW values ​​for mutant individuals (mutant at both SNP1 and SNP4) were 14.36 mm, 9.21 mm, and 10.15 mm, respectively. In contrast, the mean NWL, NWW, and WW values ​​for wild-type individuals (reference type at both SNP1 and SNP4) were 11.39 mm, 7.57 mm, and 8.47 mm, respectively. In other words, individuals with mutant SNP1 and SNP4 had greater mean NWL, NWW, and WW values ​​than wild-type individuals. The relationships between the genotypes at SNP1 and SNP4 and the WW measurements, as well as the relationships between the genotypes at SNP4 and the NWL measurements, were similar to those described above and are not further detailed here.

[0104] The relationship between the genotype at SNP8 and NWL measurements for some individuals was statistically analyzed. The statistical results showed that the mean NWL for individuals with the mutant SNP8 was 14.88 mm, while the mean NWL for individuals with the wild-type SNP8 was 11.39 mm. In other words, individuals with the mutant SNP8 had a greater NWL than those with the wild-type SNP8.

[0105] Similar results were found at SNP18 and SNP24. The mean NWW value for individuals with the SNP18 mutation was 9.34 mm, while the mean NWW value for individuals with the wild type was 7.57 mm. The mean WL value for individuals with the SNP24 mutation was 29.82 mm, while the mean WL value for individuals with the wild type was 26.66 mm.

[0106] The above results show that the genotypes at the above-mentioned SNP1, SNP4, SNP8, SNP18 and SNP24 sites can be used to accurately type the seeds of Tabebuia jacaranda.

[0107] 2.6 Candidate gene location and GO function analysis

[0108] Through genome-wide association studies (GWAS) of four seed morphological traits of Tabebuia chinensis, combined with published Tabebuia chinensis genome sequencing results, single-nucleotide polymorphism (SNP) markers significantly associated with seed morphological traits were mapped to the Tabebuia chinensis genome. Within a linkage disequilibrium (LD) decay distance of 50 kb (R²=0.1), a search for associated genes was conducted for the 29 SNPs significantly associated with seed morphology listed in Tables 4-1 to 4-3. A total of 68 associated genes (referred to as candidate genes) were identified, of which 29 candidate genes have functional annotations, as shown in Tables 5-1 to 5-3.

[0109] Table 5-1 Candidate genes corresponding to SNP sites significantly associated with NWL

[0110]

[0111] Table 5-2 Candidate genes corresponding to SNP sites significantly associated with NWW

[0112]

[0113] Table 5-3 Candidate genes corresponding to SNP sites significantly associated with WL and WW

[0114]

[0115] As shown in Tables 5-1 to 5-3, 31 candidate genes were identified among the 14 significant SNPs associated with seed length (NWL), of which 9 genes had functional annotations. Among the 16 significant SNPs associated with seed width (NWW), 28 candidate genes were identified, of which 18 genes had relatively clear functional annotations. Among the two significant SNPs associated with wing length (WL), 7 candidate genes were identified, of which 3 genes had functional annotations. Among the nine significant SNPs associated with wing width (WW), 6 candidate genes were identified, of which 3 genes had functional annotations.

[0116] In addition, one common candidate gene was located in each of NWL, NWW, and WW; two common candidate genes were located in each of NWL and NWW ( Figure 23 ).

[0117] In this example, genome-wide association analysis was used to discover for the first time a group of candidate genes associated with the seed morphology of Tabebuia chrysantha, providing a basis for further research on the molecular regulatory mechanism of the seed morphology of Tabebuia chrysantha.

[0118] 2.7 qRT-PCR verification results

[0119] The measurement results of seed morphological traits of four individuals, F1, F4, M1 and M6, and the genotypes of the loci where each SNP marker is located are shown in Table 6.

[0120] Table 6 Seed morphology and SNP types of F1, F4, M1 and M6

[0121]

[0122] To further verify the expression of candidate genes in P. tiliaceae with different seed morphological phenotypes, nine key candidate genes (six functional genes and three genes without functional annotation) were selected, including CDL12_00443, CDL12_15521, CDL12_15522, CDL12_15254, CDL12_04662, CDL12_04666, CDL12_11100, CDL12_11105, and CDL12_03077. The expression levels of each gene were measured by qRT-PCR using the primers shown in Table 7.

[0123] Among these nine key candidate genes, CDL12_00443 is located in the flanking regions of both NKXS01000048.1__3975 (SNP1) and NKXS01000048.1__4051 (SNP4), CDL12_15521 and CDL12_15522 are located in the flanking regions of NKXS01002837.1__3134 (SNP8), and CDL12_15254 is located in the flanking region of NKXS01002782.1__29157 (SNP24). These SNPs are highly likely to regulate the expression of genes in their respective flanking regions. SNP1, SNP4, SNP8, and SNP24 are all SNP sites closely associated with seed morphology in P. truncatum. Each SNP site is closely linked to the genes in its flanking region. The purpose of qRT-PCR is to verify whether the genotype at these SNP sites is closely related to the expression levels of the genes in its flanking region.

[0124] qRT-PCR results showed that the expression levels of different candidate genes showed expression differences in different phenotypes, among which the expression level of candidate gene CDL12_15522 showed significant differences in seeds of different morphologies (see Figure 24 , P<0.05). The seed morphology of P. tazetta can also be predicted based on the expression level of the candidate gene CDL12_15522. This candidate gene may have the function of regulating the seed morphology of P. tazetta.

[0125] Table 7 Primer information for each gene

[0126]

[0127] 3. Results and Discussion

[0128] Seed morphology is a relatively stable genetic characteristic of plants and is of great value in plant classification and genetics research. Seed morphology plays an important role in classification at the genus, species, and even below the species level.

[0129] According to the statistical and analytical results of seed morphology-related traits of 126 individuals of Tabebuia chrysantha from 12 regions, the seed morphology differences among the 126 individuals were extremely significant, and the coefficient of variation of each shape was relatively small. Therefore, seed morphology can be used as a relatively stable morphological characteristic and an important basis for the taxonomy of Tabebuia chrysantha.

[0130] Under natural conditions, due to the influence of genetic factors and environmental conditions, the four morphological indicators of Tabebuia chrysantha seeds—wing length (WL), wing width (WW), seed length (NWL), and seed width (NWW)—show significant variation among individual plants. Correlation analysis revealed a highly significant positive correlation between these four morphological indicators, indicating that these traits influence each other and vary synergistically, reflecting close consistency in genetic and environmental adaptation.

[0131] When using seed morphology as a basis for genus or species classification, it is necessary to accurately distinguish individuals with relative traits. However, most shapes related to seed morphology are quantitative traits, making it difficult to accurately distinguish individuals with relative traits through visual observation or conventional measurement methods alone. Therefore, developing molecular markers related to seed morphology, linking seed morphological traits with the genotype of the molecular marker, and distinguishing individuals with relative traits through genotype can better apply seed morphological traits to classification.

[0132] In the above examples, genome-wide association analysis of four seed morphological traits (NWL, NWW, WL, and WW) was performed using 131,559 SNP markers based on the MLM (QK) model. Twenty-nine SNPs were detected that were significantly associated with seed morphological traits. Further analysis revealed that nine SNPs exhibited pleiotropic effects, with three SNPs—NKXS01000048.1__3975, NKXS01000048.1__4022, and NKXS01000048.1__4051—being simultaneously detected in NWL, NWW, and WW.

[0133] After Bonferroni multiple test correction, it was found that the two SNPs NKXS01000048.1__4051 and NKXS01002837.1__3134 had strong association signals with NWL, and these sites may be the main effect sites regulating the NWL trait.

[0134] A total of 68 candidate genes were detected within the 50-bp region upstream and downstream of the 29 SNPs, 29 of which had functional annotations. The study found that CDL12_00443, CDL12_15521, and CDL12_15522 exhibited pleiotropic effects. The CDL12_00443 gene was detected in all three traits, NWL, NWW, and WW, while the CDL12_15521 and CDL12_15522 genes were detected in both NWL and NWW.

Claims

1. Application of SNP markers related to seed morphology of Tabebuia chrysantha, characterized in that: When the SNP4 or SNP8 marker is used to type the seed length of the yellow-flowered Tabebuia, the genotype of the yellow-flowered Tabebuia to be tested at the SNP4 or SNP8 site is obtained from the genomic DNA of the yellow-flowered Tabebuia to be tested, SNP4 is located at 4051bp on the NKXS01000048.1 chromosome, and SNP8 is located at 3134bp on the NKXS01002837.1 chromosome; when the genotype of the site where SNP4 is located is C, or when the genotype of the site where SNP8 is located is G, the seeds of the yellow-flowered Tabebuia to be tested are long type; when the genotype of the site where SNP4 is located is T, or when the genotype of the site where SNP8 is located is T, the seeds of the yellow-flowered Tabebuia to be tested are short type.

2. Application of SNP markers related to seed morphology of Tabebuia chrysantha, characterized in that: When the SNP18 marker is used to type the seed width of the yellow-flowered Tabebuia, the genotype of the SNP18 site of the yellow-flowered Tabebuia to be tested is obtained from the genomic DNA of the yellow-flowered Tabebuia to be tested. SNP18 is located at 11016bp on chromosome NKXS01004877.1; when the genotype of the SNP18 site is T, the seeds of the yellow-flowered Tabebuia to be tested are of a wide type; when the genotype of the SNP18 site is C, the seeds of the yellow-flowered Tabebuia to be tested are of a narrow type.

3. Application of SNP markers related to seed morphology of Tabebuia chrysantha, characterized in that: When the SNP24 marker is used to type the wing length, the genotype of the yellow-flowered Tabebuia to be tested at the SNP24 site is obtained from the genomic DNA of the yellow-flowered Tabebuia to be tested. SNP24 is located at 92361bp on chromosome NKXS01001449.

1. When the genotype of the SNP24 site is C, the seeds of the yellow-flowered Tabebuia to be tested are long-winged type, and when the genotype of the SNP24 site is G, the seeds of the yellow-flowered Tabebuia to be tested are short-winged type.

4. Application of SNP markers related to seed morphology of Tabebuia chrysantha, characterized in that: When the SNP1 marker is used to type the wing width, the genotype of the yellow-flowered Tabebuia to be tested at the SNP1 site is obtained from the genomic DNA of the yellow-flowered Tabebuia to be tested, and SNP1 is located at 3975bp on chromosome NKXS01000048.1; when the genotype at the SNP1 site is T, the seeds of the yellow-flowered Tabebuia to be tested are wide-winged type, and when the genotype at the SNP1 site is G, the seeds of the yellow-flowered Tabebuia to be tested are narrow-winged type.

5. Application of SNP markers related to seed morphology of Tabebuia chrysantha, characterized in that: When the SNP marker is used to simultaneously genotype seed length, seed width, and wing width, the genotype of the tested Tabebuia chrysantha at the sites of SNP1 and SNP4 is obtained from the genomic DNA of the tested Tabebuia chrysantha. SNP1 is located at 3975 bp on chromosome NKXS01000048.1, and SNP4 is located at 4051 bp on chromosome NKXS01000048.

1. When the genotype of the site where SNP1 is located is T and the genotype of the site where SNP4 is located is C, the seeds of the tested Tabebuia chrysantha are long, wide and have wide wings; when the genotype of the site where SNP1 is located is G and the genotype of the site where SNP4 is located is T, the seeds of the tested Tabebuia chrysantha are short, narrow and have narrow wings.

Images