Screening method for excellent cultivars of peony based on genetic information of flower pattern and color

Abstract

The application discloses a peony excellent variety screening method based on flower type and color genetic information, and belongs to the technical field of Chinese medicinal material breeding. The application effectively eliminates mixed individuals by systematically analyzing the genetic background of double-petal / single-petal and different flower colors through ITS, SSR, SNP, chloroplast whole genome and transcriptome multi-dimensional molecular markers. The application carries out multi-dimensional and three-dimensional evaluation of yield, medicinal material quality and efficacy of the selected single plant based on the genetic fingerprint as a screening premise. The application deeply fuses the flower color phenotype, molecular data and quality indexes by using high-order tensor decomposition and entropy weight-TOPSIS method, quantitatively evaluates the comprehensive closeness of each single plant, and accurately locates the excellent variety with high yield, high quality and strong efficacy. The application converts the traditional experience-dependent seed selection into quantitative screening based on multi-source data fusion, shortens the breeding cycle, realizes accurate identification at the individual level, and provides reliable technical support for the breeding of medicinal peony varieties.

Description

Technical Field

[0001] This invention belongs to the field of Chinese medicinal herb breeding technology, specifically involving a method for screening superior peony varieties based on genetic information of flower shape and color. Background Technology

[0002] White peony root is a processed product of the root of Paeonia lactiflora Pall, a plant in the Ranunculaceae family. It is a commonly used bulk Chinese medicinal herb with the effects of nourishing blood and astringing yin, soothing the liver and relieving pain, and calming liver yang. In recent years, due to the large market demand for white peony root and the rapid expansion of its cultivation area, the local seed source in Sichuan can no longer meet the production needs, and a large number of varieties have been introduced from other provinces such as Anhui and Zhejiang for cultivation. The local varieties in Sichuan are double-flowered, while the varieties in Anhui, Zhejiang and other places are mostly single-flowered, resulting in a mixed state of various flower types and colors of medicinal peonies cultivated in Sichuan Province, and the quality of the medicinal materials is uneven.

[0003] Existing research has made preliminary explorations into the differences in metabolites among different peony varieties, indicating that there are differences in chemical composition among them. However, there is currently insufficient systematic classification research on medicinal peonies with different flower shapes and colors, no variety evaluation system based on flower shape and color characteristics has been established, and there is a lack of screening methods that link the genetic information of flower shape and color with the quality, yield, and efficacy of medicinal materials.

[0004] Currently, the screening of superior varieties of Chinese medicinal materials mainly relies on traditional methods such as phenotypic observation and chemical component content determination, which has the following shortcomings: Lack of genetic information guidance: Existing screening methods are mostly based on single-index evaluation of phenotypic traits or chemical components, and do not incorporate genetic information such as the genetic stability and kinship of varieties into the evaluation system. It is difficult to reveal the essential reasons for the differences between varieties at the molecular level, and it is also impossible to provide genetic basis for subsequent molecular breeding and hybridization breeding.

[0005] Evaluation indicators are limited: Existing technologies usually only focus on one type of indicator of medicinal materials (chemical composition content or yield), without comprehensively considering multiple dimensions such as genetic stability, quality of medicinal materials, yield and efficacy, making it difficult to comprehensively evaluate the quality of varieties.

[0006] The relationship between flower shape and color and quality has not been established: Flower shape and color are the most intuitive and easily identifiable phenotypic characteristics of medicinal peony. However, current technology has not yet established a systematic relationship between flower shape and color and the genetic information, quality, yield and efficacy of medicinal materials. Therefore, it is impossible to use this simple characteristic of flower shape and color to guide the rapid screening of superior varieties.

[0007] To address the problems existing in the current technology, we propose a method for screening superior peony varieties based on genetic information of flower shape and color. Summary of the Invention

[0008] The purpose of this invention is to address the shortcomings of existing technologies by providing a method for screening superior peony varieties based on genetic information of flower shape and color.

[0009] This invention is implemented as follows: a method for screening superior peony varieties based on genetic information of flower shape and color, wherein the method for screening superior peony varieties based on genetic information of flower shape and color includes: S10. During the peak blooming period of peonies, peonies were classified into two types according to flower shape: double-petaled and single-petaled, and the flower shape was marked. The flower color was also classified according to petal color and marked. Leaf and petal tissues were collected from the marked peony plants of different flower shapes and colors as test samples. The flower colors are divided into: double-petaled white, double-petaled deep purple-pink, double-petaled light purple-pink, single-petaled deep purple-pink, single-petaled bright purple-red, single-petaled purple-pink, single-petaled light purple-pink, and single-petaled white. S20. Complete leaf and petal tissues of peonies with different flower shapes and colors were obtained, and genomic DNA and total RNA of peonies were extracted respectively. Based on the extracted genomic DNA and total RNA, multidimensional genetic information analysis of peonies was carried out to establish a genetic fingerprint map based on flower shape and color phenotype. S30, using genetic fingerprinting as a priori, conduct multi-dimensional quality evaluation on at least one group of target variety individual plants, wherein the multi-dimensional quality evaluation includes efficacy evaluation, yield evaluation, and medicinal material quality evaluation, to obtain multi-dimensional evaluation results of target variety individual plants; S40: Obtain multi-dimensional evaluation results of individual plants of the target variety. Combine flower shape and color, multi-dimensional evaluation results, and modal correlation matrix of genetic fingerprint. Construct a regression screening model based on the modal correlation matrix, calculate the comprehensive closeness score of the target variety, and screen out superior varieties.

[0010] Preferably, the multidimensional genetic information analysis of peony includes: Based on peony genomic DNA, nuclear genome polymorphism analysis was performed on peonies with different flower types and colors to obtain genetic distances, SNP fingerprints, and chloroplast whole genome sequences of peonies with different flower types and colors, and to generate nuclear genome polymorphism analysis results. Based on total RNA from peonies, differential transcriptome expression analysis was performed on peonies with different flower shapes and colors to obtain genetic information differences. Based on the genetic information differences, differentially expressed genes related to flower shape development and structural genes and transcription factors related to flower color formation were screened. Genetic fingerprints based on flower pattern and color phenotypes were constructed by combining nuclear genome polymorphism analysis results and key expressed genes.

[0011] Preferably, the nuclear genome polymorphism analysis of peonies with different flower types and colors includes: The internal transcribed spacer region sequence of ribosomal DNA was amplified by PCR to obtain the ITS amplified sequence. The ITS amplified sequence was then subjected to Sanger sequencing. The sequencing results were compared with the GenBank database to calculate the genetic distance of peonies with different flower shapes and colors. Based on peony genomic DNA, peony-specific SSR primer pairs were designed and synthesized. PCR amplification was performed on peony genomic DNA of different flower types and colors. After separation by capillary electrophoresis, STR typing was performed on the PCR amplification products. Microsatellite fingerprints of each flower type and color were constructed using allele fragment length polymorphism data. The Nei genetic similarity coefficient method was used to calculate the genetic similarity coefficient. Combined with the genetic similarity coefficient, the genetic similarity coefficient matrix of different flower types and colors of peonies was calculated. The UPGMA method was used to construct a genetic clustering tree. Combining the genetic similarity coefficient matrix and the clustering results, flower type and color groupings with similar genetic backgrounds were identified, and the genetic boundaries of the flower type and color groupings were determined. Using the genetic boundary of flower shape and color type grouping as a prior constraint, similar groups and distant groups of different flower shape and color types are determined based on the genetic boundary. For similar groups, fixed differential SNPs within the group are screened, and for distant groups, conserved common SNPs within the group are screened. A pre-constructed paired-end sequencing library was used to mine candidate SNP sites based on simplified genome sequencing data, resulting in a set of targeted SNP markers. SNP genotyping was performed on the target SNP marker set, an SNP fingerprint map was constructed based on the SNP genotyping results, and the genetic distance matrix was calculated using the Nei genetic distance method. A chloroplast genome map was constructed based on genomic DNA, and structural variation features of chloroplast genomes of different flower types and colors were extracted by combining the chloroplast genome map.

[0012] Preferably, the differential transcriptome expression analysis of peonies with different flower types and colors includes: Total RNA was extracted from peony using a plant total RNA extraction kit. The total RNA was transcribed into cDNA. An Illumina standard cDNA library was constructed based on the cDNA. The Illumina standard cDNA library was sequenced to obtain high-quality sequencing reads. High-quality sequencing reads were aligned to the peony reference genome, and transcripts were assembled and their expression levels were quantified to obtain the reference gene expression level. Differential expression analysis was performed based on the reference gene expression level to obtain differential expression values. Differential expression genes were screened using differential expression thresholds, and functional annotation and pathway enrichment analysis were performed on the differentially expressed genes of peony plants. Based on functional annotation and pathway enrichment analysis of differentially expressed genes, differentially expressed genes associated with stamen petalization were screened, and structural genes and transcription factors associated with flower color formation were also screened to obtain the key expressed genes.

[0013] Preferably, differentially expressed genes related to flower shape development, including AGAMOUS and AGL9 genes, are screened based on the results of genetic information differences; structural genes related to flower color formation, including flavonol synthase and chalcon synthase genes (CHS); and transcription factors related to flower color formation, including R2R3-MYB transcription factor and SPL9 transcription factor.

[0014] Preferably, the construction of a genetic fingerprint based on flower pattern and color phenotype by combining nuclear genome polymorphism analysis results and key expressed genes includes: The results of nuclear genome polymorphism analysis were obtained, and the genetic distance, SNP fingerprints, and structural variation features of chloroplast genomes of different flower types and colors of peonies were identified. The qRT-PCR expression levels of key expressed genes were also obtained. The results of nuclear genome polymorphism analysis and the qRT-PCR expression levels of key expressed genes were standardized to obtain standardized multidimensional genetic data. A hierarchical weighted integration strategy is adopted to assign weights to the distribution of multi-dimensional genetic data, and then the weighted multi-dimensional genetic data are linearly fused according to the weights to obtain a comprehensive genetic distance matrix. Based on the comprehensive genetic distance matrix, hierarchical clustering was performed using the UPGMA method to construct genetic clustering trees for different flower patterns and colors. Molecular identifier combinations were labeled on the genetic clustering trees, and a genetic fingerprint map based on flower pattern and color phenotypes was output with flower pattern and color type as the index unit.

[0015] Preferably, the multi-dimensional quality evaluation includes efficacy evaluation, yield evaluation, and medicinal material quality evaluation, to obtain multi-dimensional evaluation results for a single plant of the target variety; Methods for multi-dimensional quality evaluation of at least one group of target variety individual plants, including: Based on the genetic fingerprint profile, target variety plants with flower shape and color characteristics are selected from flower shape and color types that are determined to have stable and consistent genetic background. The representative flower shape and color characteristics include: the flower shape is double or single, and the plant is in normal growth status and free from pests and diseases. The number of target variety plants selected for each flower shape and color type is not less than 20, and the genetic fingerprint profile index number of each plant is recorded. The yield of a single plant of the target variety is evaluated to obtain the yield evaluation value of a single plant of the target variety. The quality of medicinal materials was evaluated on individual plants of the target variety to obtain a comprehensive quality evaluation value. When evaluating the quality of medicinal materials on individual plants of the target variety, hyperspectral analysis, UPLC-MS chemical composition identification, DESI-MSI mass spectrometry imaging analysis, UPLC multi-component content determination, and qNMR total glycoside content determination were used. After normalizing the measurement results, the correlation degree between the individual plants of the target variety and the ideal high-quality medicinal material reference sequence was calculated using the grey relational analysis method, and the correlation degree was used as the comprehensive quality evaluation value of the medicinal materials. The expression levels of key genes in individual plants of the target variety were detected, and the contents of pentagalloyl glucose, paeoniflorin, and hydroxypaeoniflorin were detected by UPLC-MS chemical composition identification method to obtain the total amount of medicinal components. The expression levels of key genes and the total amount of medicinal components were positively normalized, and the entropy weight method was used to determine the weights of the expression levels of key genes and the total amount of medicinal components to obtain the comprehensive value of medicinal components. The yield evaluation value, comprehensive quality evaluation value, and comprehensive efficacy component value of the target variety single plant are integrated into a multi-dimensional quality evaluation result, and stored in association based on the genetic fingerprint index number of the target variety single plant.

[0016] Preferably, when evaluating the yield of a single plant of the target variety, the product evaluation includes aboveground trait indicators and underground yield indicators. By combining the aboveground trait indicators and underground yield indicators, the weight of each indicator is determined using the entropy weight method to obtain the yield evaluation value of a single plant of the target variety. The aboveground trait indicators include stem height, number of leaves, number of flowers, and number of fruits of single-petaled flower type. The underground yield indicators include fresh and dry weight of roots, number of taproots, root width, number of buds, length of the longest taproot, and diameter of the thickest taproot.

[0017] Preferably, the calculation of the comprehensive similarity score of the target variety includes: Using the flower shape and color of individual plants of the target variety as the row dimension, the multi-dimensional evaluation results of individual plants of the target variety as the column dimension, and the molecular identifier combination of the genetic fingerprint as the depth dimension, a third-order data tensor representing the excellence of individual plants of the target variety is constructed. The third-order data tensor was decomposed using high-order singular value decomposition (HOSVD) to obtain the core tensor and three sets of modal correlation matrices: flower type-evaluation index matrix, flower type-genetic information matrix, and evaluation index-genetic information matrix. Factor matrices corresponding to the modal correlation matrices were extracted, and the factor matrices were solved using the HOSVD algorithm. The top r principal components with a cumulative variance contribution rate ≥ 85% were retained. Based on the element values ​​of the core tensor, the synergistic correlation patterns among the three modalities of flower type, evaluation index, and genetic information were identified. When the element value of the core tensor was greater than the preset significance threshold (0.2), it was determined that there was a significant synergistic correlation between flower type, evaluation index, and genetic information, and the significant correlation triplet was recorded. An ideal high-quality variety reference sequence was pre-constructed. The entropy weight-TOPSIS method was used to calculate the comprehensive closeness score between the target variety and the ideal reference sequence. The comprehensive closeness score between the target variety and the ideal reference sequence was used as the dependent variable, and the principal component scores of flower type and color type, the principal component scores of evaluation indicators extracted from the factor matrix, and the quantitative coding values ​​of molecular identifier dimension were used as independent variables to construct a multivariate nonlinear regression screening model. in, , representing the comprehensive similarity score and the basic predicted value of the comprehensive similarity between the i-th target variety single plant and the ideal reference sequence, respectively. This represents the regression coefficient, the m-th original independent variable (the principal component score of flower pattern and color type, the principal component score of the evaluation index extracted from the factor matrix, and the quantitative encoding value of the molecular identifier dimension). Represents the random error term. Represents a nonlinear transformation function; The regression coefficients were estimated using the least squares method, and a significance test was performed. When the adjusted coefficient of determination of the model was ≥0.80 and the significance test was <0.01, the regression screening model was deemed to be a good fit. Substitute the multi-dimensional evaluation results of the target varieties into the fitted regression screening model to calculate the predicted comprehensive similarity score. Sort the varieties by comprehensive similarity score from high to low and select the top 3 flower type and color type as candidates for superior varieties. For all target variety individual plants within the candidate types of superior varieties, sort them by comprehensive similarity score and select the top 30% of individuals as core breeding materials to complete the screening of superior varieties.

[0018] Preferably, when pre-constructing the reference sequence for ideal high-quality varieties, for each flower type and color type, significantly associated individuals in the significantly associated triplet within that flower type and color type whose yield evaluation value, comprehensive medicinal material quality evaluation value, and comprehensive medicinal component value are all in the top 10 percentile are selected, and the mean of the multi-dimensional evaluation results of the significantly associated individuals is used as the ideal reference value for the flower type and color type. At the same time, the common molecular marker features of the flower type and color type are extracted as ideal genetic markers, and the ideal high-quality variety reference sequence is constructed by combining the ideal reference value and the ideal genetic markers.

[0019] Compared with the prior art, the embodiments of this application have the following main advantages: This invention systematically analyzes the genetic background of double / single petals and different flower colors using multidimensional molecular markers such as ITS, SSR, SNP, chloroplast whole genome, and transcriptome, effectively eliminating mixed individuals. Furthermore, using genetic fingerprinting as a screening prerequisite, it conducts multidimensional and comprehensive evaluations of yield, medicinal quality, and efficacy for selected individual plants. By utilizing high-order tensor decomposition and entropy-weighted TOPSIS methods, it deeply integrates flower color phenotype, molecular data, and quality indicators to quantitatively assess the overall similarity of each individual plant, thereby accurately identifying high-yielding, high-quality, and highly effective superior varieties. This transforms traditional experience-based selection into quantitative screening based on multi-source data fusion, shortening the breeding cycle and achieving precise identification at the individual level, providing reliable technical support for the breeding of superior medicinal peony varieties.

[0020] In this embodiment of the invention, by integrating multidimensional molecular marker technologies such as ITS sequence analysis, SSR microsatellite markers, SNP targeted screening, chloroplast whole genome sequencing, and transcriptome differential expression analysis, the genetic background of double / single-petaled and different colored medicinal peonies was analyzed, and a comprehensive genetic fingerprint map including nuclear gene sequence polymorphism, chloroplast genome structural variation, and functional gene expression differences was constructed. The genetic fingerprint map uses Bootstrap resampling to assess branch support rate, and a threshold of ≥90% is used to determine independent varieties or lines with stable and consistent genetic backgrounds. This effectively identifies and excludes individuals with mixed germplasm, thereby solving the technical bottleneck of traditional phenotypic observation being unable to distinguish between genetic stability and hybridization. This provides a reliable molecular-level a priori basis for subsequent precise screening guided by genetic information.

[0021] This invention uses genetic fingerprinting as a screening prerequisite constraint to simultaneously conduct three-dimensional evaluations of target varieties with verified genetic backgrounds, including yield evaluation, medicinal quality evaluation, and efficacy evaluation. Yield evaluation covers 10 indicators of aboveground agronomic traits and underground root yield, objectively weighted using entropy weighting. Medicinal quality evaluation integrates hyperspectral analysis, UPLC-MS chemical composition identification, DESI-MSI mass spectrometry imaging, UPLC multi-component content determination, and qNMR total glycoside content determination, quantifying the correlation using grey relational analysis. Efficacy evaluation integrates key gene expression levels and core efficacy component content, comprehensively using entropy weighting. This multi-dimensional evaluation system overcomes the limitations of existing single-indicator evaluations, achieving full-chain quality control from genetic stability determination, phenotypic yield measurement, qualitative and quantitative chemical composition analysis, and efficacy material basis analysis, ensuring that selected individual plants possess comprehensive traits of high yield, high quality, and high efficacy.

[0022] This invention constructs a third-order data tensor from flower shape and color phenotypes, multi-dimensional evaluation results, and molecular identifiers from genetic fingerprinting. It employs High-Order Singular Value Decomposition (HOSVD) to extract modal association matrices and identify significant co-associated triples. Then, it combines the entropy-weighted TOPSIS method to calculate a comprehensive proximity score, constructing a multivariate nonlinear regression screening model with principal component scores of flower shape and color, principal component scores of evaluation indicators, and quantitative coding values ​​of molecular identifiers as independent variables. This regression screening model transforms traditional experience-based selection into quantitative screening based on multi-source data fusion. It can overcome the limitations of single flower shape and color types in all target varieties, enabling precise cross-type individual-level screening. The selected core breeding materials cover multiple flower shape and color types, each with its own differentiated advantages, shortening the breeding cycle and providing diversified high-quality germplasm resources for subsequent targeted breeding and hybridization for different application needs. Attached Figure Description

[0023] Figure 1 A schematic diagram of the standard phenotypic phenotype of peony flower type and color is shown in an embodiment of the present invention.

[0024] Figure 2 This invention illustrates a schematic diagram of peony plant markings for different flower types and colors in an embodiment of the invention.

[0025] Figure 3 The SNP fingerprint spectrum based on fixed differential SNP sites in similar grouping is shown in an embodiment of the present invention.

[0026] Figure 4 A schematic diagram illustrating the implementation process of a method for screening superior peony varieties based on genetic information of flower shape and color is shown. Detailed Implementation

[0027] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used herein in the specification of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and foregoing drawings of this application are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or foregoing drawings of this application are used to distinguish different objects, not to describe a particular order.

[0028] This invention addresses the industry pain points of mixed varieties and inconsistent quality of white peony root, as well as the lack of a systematic screening method based on flower shape and color. Targeted research and development was conducted to address these issues. During the research process, it was discovered that existing technologies can only solve single problems and cannot simultaneously address the determination of genetic background stability, the correlation between flower shape and color phenotypes and molecular genetic information, or the integration of multi-dimensional quality evaluation and efficacy verification.

[0029] Based on existing technologies, this invention innovatively proposes an overall technical solution for screening superior peony varieties based on flower shape and color genetic information. By organically combining flower shape and color phenotypic classification, multi-dimensional genetic information analysis, genetic fingerprint map construction, multi-dimensional quality evaluation, three-dimensional correlation matrix and regression screening model, it achieves multiple technical effects such as rapid identification of varieties with stable genetic background, establishment of correlation between flower shape and color and quality, and comprehensive screening of multiple indicators.

[0030] This scheme not only solves the problems of existing technologies lacking genetic information guidance, having single evaluation indicators, and failing to establish a correlation between flower shape and color and quality, but also achieves the following overall technical effects: Using the intuitive phenotypic characteristic of flower shape and color as a priori, a genetic fingerprint based on the flower shape and color phenotype is constructed through the synergy of nuclear genome polymorphism analysis (ITS sequence, microsatellite markers, SNP markers, chloroplast genome) and transcriptome differential expression analysis (flower shape development gene AG / AGL9, flower color formation structure gene CHS, and transcription factor R2R3-MYB / SPL9). Then, using the genetic fingerprint as a priori constraint, yield, medicinal quality, and efficacy of individual plants of the target variety are evaluated. Finally, through the analysis of flower shape and color, multi-dimensional evaluation results, third-order data tensors of the genetic fingerprint, higher-order singular value decomposition, and entropy-weighted TOPSIS, a comprehensive proximity score is calculated to screen out superior varieties that combine high yield, high content of medicinal components, and strong blood-tonifying effects. The technical effect of this invention is the result of the synergistic effect of flower type and color phenotypic classification, multi-dimensional genetic information analysis, medicinal material quality evaluation, efficacy verification and tensor decomposition regression screening technology scheme. It is not a conventional combination of common knowledge in the field and cannot be expected by those skilled in the art based on existing technology.

[0031] This invention systematically analyzes the genetic background of double / single petals and different flower colors using multidimensional molecular markers such as ITS, SSR, SNP, chloroplast whole genome, and transcriptome, effectively eliminating mixed individuals. Furthermore, using genetic fingerprinting as a screening prerequisite, it conducts multidimensional and comprehensive evaluations of yield, medicinal quality, and efficacy for selected individual plants. By utilizing high-order tensor decomposition and entropy-weighted TOPSIS methods, it deeply integrates flower color phenotype, molecular data, and quality indicators to quantitatively assess the overall similarity of each individual plant, thereby accurately identifying high-yielding, high-quality, and highly effective superior varieties. This transforms traditional experience-based selection into quantitative screening based on multi-source data fusion, shortening the breeding cycle and achieving precise identification at the individual level, providing reliable technical support for the breeding of superior medicinal peony varieties.

[0032] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0033] Example 1: Peony flower type and color phenotypic marking and sample collection In this embodiment of the invention, during the peak flowering season of peonies in April and May, peony plant samples of different flower types and colors were collected at a peony cultivation base in Qu County, Sichuan Province (107°09′E, 30°56′N, altitude 642m). Cultivated medicinal peonies were classified into two types according to flower type: double-petaled and single-petaled, and further classified into eight flower color types as shown in Table 1. Figure 1 A schematic diagram of the standard phenotypic phenotype of peony flower type and color is shown in an embodiment of the present invention.

[0034] Table 1 For each flower type and color, 40 plants in normal growth condition and free from pests and diseases were selected and individually marked. Figure 2 The illustration shows a schematic diagram of peony plant marking with different flower types and colors in an embodiment of the present invention. Leaves (for DNA extraction) and petals and petaloid stamen tissues (for RNA extraction) were collected respectively, flash-frozen in liquid nitrogen and stored at -80°C for later use.

[0035] Example 2: Multidimensional genetic information analysis and construction of genetic fingerprinting In this embodiment of the invention, during the multidimensional genetic information analysis and genetic fingerprinting, complete leaf and petal tissues of peonies with different flower shapes and colors are obtained, and genomic DNA and total RNA of peonies are extracted respectively. Based on the extracted genomic DNA and total RNA, multidimensional genetic information analysis of peonies is performed to establish a genetic fingerprinting based on flower shape and color phenotypes. The genetic fingerprinting is of independent varieties / strains with stable genetic backgrounds, and excludes genetic fingerprinting of hybrid individuals.

[0036] Specifically, the multidimensional genetic information analysis of peony includes: S101, based on peony genomic DNA, performs nuclear genome polymorphism analysis on peonies with different flower types and colors, obtains genetic distance, SNP fingerprint map, chloroplast whole genome sequence of peonies with different flower types and colors, and generates nuclear genome polymorphism analysis results; Genomic DNA was extracted from leaves using a modified CTAB method. Integrity was assessed by 1% agarose gel electrophoresis, and concentration and A260 / A280 (1.8–2.0) and A260 / A230 (≥2.0) ratios were determined using a micro spectrophotometer. Total RNA was extracted from petals and petaloid stamens using a plant total RNA extraction kit. 28S / 18S bands were detected by 1% agarose gel electrophoresis, and A260 / A280 and A260 / A230 ratios were determined using a micro spectrophotometer. DNA that passed quality control was used in subsequent experiments.

[0037] S102, based on total RNA of peony, transcriptome differential expression analysis was performed on peonies with different flower shapes and colors to obtain the results of genetic information differences. Based on the results of genetic information differences, differentially expressed genes related to flower shape development and structural genes and transcription factors related to flower color formation were screened. S103, combining the results of nuclear genome polymorphism analysis and key expressed genes, constructs a genetic fingerprint map based on flower shape and color phenotype. The genetic fingerprint map is used to characterize the polymorphic characteristics of different flower shapes and colors of peonies at the nuclear gene sequence level, the structural variation characteristics at the chloroplast genome level, and the difference characteristics at the functional gene expression level. Based on this, it is determined whether each flower shape and color type is an independent variety or strain with a stable and consistent genetic background, and hybrid individuals are excluded.

[0038] In this embodiment of the invention, the nuclear genome polymorphism analysis of peonies with different flower types and colors includes: S1011, ITS sequence analysis: The internal transcribed spacer region sequence of ribosomal DNA was amplified by PCR to obtain the ITS amplified sequence. The ITS amplified sequence was then subjected to Sanger sequencing. The sequencing results were compared with the GenBank database to calculate the genetic distance of peonies with different flower shapes and colors. Among them, when using PCR to amplify the internal transcribed spacer region sequence of ribosomal DNA, the internal transcribed spacer region sequence includes the full-length sequences of ITS1, 5.8S rRNA and ITS2, with the ITS1-5.8S-ITS2 region of ribosomal DNA as the target. In this embodiment, universal primers ITS1 (5'-TCCGTAGGTGAACCTGCGG-3') and ITS4 (5'-TCCTCCGCTTATTGATATGC-3') were used for PCR amplification to obtain the ITS amplified sequence.

[0039] PCR reaction mixture (25 μL): 12.5 μL 2×Taq PCR Master Mix, 1.0 μL each of forward and reverse primers (10 μmol / L), 1.0 μL template DNA (approximately 50 ng), and 9.5 μL ddH2O. Amplification program: 94℃ pre-denaturation for 5 min; 94℃ denaturation for 30 s, 55℃ annealing for 30 s, 72℃ extension for 1 min, 35 cycles; final extension at 72℃ for 10 min.

[0040] After 1% agarose gel electrophoresis, the amplified ITS sequences were sent to the Beijing Nuosai Genome Research Center for Sanger bidirectional sequencing. The sequencing results were assembled using BioEdit v7.2.3 software and then BLAST-aligned with ITS sequences of Paeonia species in the GenBank database. The genetic distances between different flower types and colors were calculated using the Kimura 2-parameter model in MEGA-X v10.2.4 software, and a Neighbor-Joining phylogenetic tree (Bootstrap 1000 replicates) was constructed. When calculating the genetic distances between different flower types and colors of peonies, the ITS sequence lengths for the eight flower types ranged from 642 to 658 bp. The genetic distance between PP-1 and PP-3 types was the smallest (0.012), while the genetic distance between PP-5 and PP-8 types was the largest (0.089). The phylogenetic tree shows that double-flowered types (PP-1, PP-2, PP-3) cluster in one branch, while single-flowered types (PP-4, PP-5, PP-6, PP-7, PP-8) cluster in another branch, supporting the rationality of flower type as a primary classification criterion.

[0041] S1012, Preliminary Exploration of Microsatellite Marker Analysis and Population Genetic Structure: Based on peony genomic DNA, peony-specific SSR primer pairs were designed and synthesized. PCR amplification was performed on peony genomic DNA of different flower types and colors. After separation by capillary electrophoresis, STR typing was performed on the PCR amplification products. Microsatellite fingerprints of each flower type and color were constructed using allele fragment length polymorphism data. The Nei genetic similarity coefficient method was used to calculate the genetic similarity coefficient. Combined with the genetic similarity coefficient, the genetic similarity coefficient matrix of different flower types and colors of peonies was calculated. The UPGMA method was used to construct a genetic clustering tree. Combining the genetic similarity coefficient matrix and the clustering results, flower type and color groupings with similar genetic backgrounds were identified, and the genetic boundaries of flower type and color groupings were determined. In this embodiment of the invention, the primer sequences of the peony-specific SSR primer pairs designed and synthesized are shown in Table 2.

[0042] Table 2 When performing PCR amplification on peony genomic DNA of different flower types and colors, a 20 μL reaction system was used. Allele frequencies, observed heterozygosity (Ho), expected heterozygosity (He), and polymorphism information content (PIC) were calculated using Cervus v3.0.7 software. The genetic similarity coefficient matrix was calculated using Ntsys-pc v2.10e software based on the Nei genetic similarity coefficient method, and a genetic clustering tree was constructed using the UPGMA method. In this embodiment, 12 pairs of SSR primers detected a total of 76 alleles, with an average of 4.35 alleles and an average PIC value of 0.682. PP-1 and PP-3 had the highest genetic similarity coefficient (0.91) and were identified as closely related; PP-5 and PP-8 had the lowest genetic similarity coefficient (0.41) and were identified as distantly related. The UPGMA clustering tree divided the eight types into two major clusters at a genetic similarity coefficient of 0.65: double-flowered clusters and single-flowered clusters, consistent with the ITS analysis results.

[0043] S1013, SNP marker-targeted screening and fingerprinting: Using the genetic boundary of flower shape and color type grouping as a priori constraint, similar and distant groups of different flower shape and color types are determined based on the genetic boundary. For similar groups, fixed differential SNPs are screened within the group, and for distant groups, conserved common SNPs are screened within the group. Specifically, when determining similar and distant groups of different flower shape and color types based on the genetic boundary, peony flower shape and color types with a genetic similarity coefficient ≥0.8 are judged as similar groups, and peony flower shape and color types with a genetic similarity coefficient ≤0.55 are judged as distant groups; when screening for fixed differential SNPs within the group, the fixation index is ≥0.65. S1014 involves the pre-construction of a paired-end sequencing library. Candidate SNP sites are mined from simplified genome sequencing data to obtain a set of targeted SNP markers. The selection criteria for the targeted SNP marker set are: overall polymorphism information content ≥ 0.3, minor allele frequency ≥ 0.05, and linkage disequilibrium coefficient ≤ 0.75. Specifically, a pre-constructed paired-end sequencing library with a 350bp insert is used, and sequencing is performed on an Illumina NovaSeq 6000 platform using the GBS (Genotyping-by-Sequencing) simplified genome sequencing strategy. After quality control using Trimmomatic v0.39, SNP calling is performed using the TASSEL-GBS v5.2.64 pipeline.

[0044] S1015 involves SNP genotyping validation of the targeted SNP marker set, constructing an SNP fingerprint map based on the SNP genotyping results, and calculating the genetic distance matrix using the Nei genetic distance method. For SNP genotyping validation of the targeted SNP marker set, KASP (Kompetitive Allele Specific PCR) technology is used. Based on the KASP genotyping results, an SNP fingerprint map is constructed, and the genetic distance matrix is ​​calculated using the Nei genetic distance method.

[0045] In this embodiment of the invention, a total of 151 high-quality SNP loci were screened, including 40 SNPs with fixed differences in closely related groups, 34 SNPs with conservative commonality in distantly related groups, and the remaining 77 SNPs that are population-wide common. Figure 3 This invention illustrates an SNP fingerprint map based on closely grouped, fixed-difference SNP loci in an embodiment of the invention. The SNP fingerprint map is constructed with 40 fixed-difference SNP markers (SNP-FD-PP-001 to SNP-FD-PP-040) as rows and eight flower type and color types (PP-1 double white, PP-2 double deep purple-pink, PP-3 double light purple-pink, PP-4 single deep purple-pink, PP-5 single bright purple-red, PP-6 single purple-pink, PP-7 single light purple-pink, PP-8 single white) as columns. The green, purple, yellow, blue, and red colors represent different homozygous alleles, while white represents heterozygous genotypes or missing data. The SNP fingerprint clustering characteristics of each flower type and color can be clearly seen from the figure. This map intuitively shows the fixed differential SNP distribution patterns between double-flowered types (PP-1 to PP-3) and single-flowered types (PP-4 to PP-8), as well as within similar groups (such as PP-1 and PP-2), thus providing core molecular markers for subsequent construction of genetic fingerprint maps and determination of genetic background stability.

[0046] S1016, Chloroplast Whole Genome Sequencing and Structural Variation Analysis: A chloroplast genome map was constructed based on genomic DNA. Structural variation features of chloroplast genomes with different flower types and colors were extracted from the chloroplast genome map. Specifically, after extracting genomic DNA, a 350bp insert library was constructed. Paired-end sequencing was performed using the Illumina HiSeq X Ten platform. After quality control using the NGSQC Toolkit, the raw data was assembled using GetOrganelle software. The complete chloroplast genome sequence was obtained after Bandage visualization correction and Bowtie2 alignment verification. Gene annotation was performed using PGA software, boundary correction was performed using Geneious software, and the chloroplast genome map was drawn using Organellar Genome DRAW software. Boundary variations of large single-copy regions, small single-copy regions, and inverted repeat regions were analyzed using IRscope software. The distribution and density of simple repeat sequences were analyzed using MISA software. Codon... W software analysis revealed codon usage biases, comparing differences in IR boundary contraction / expansion, gene loss / pseudogenization, and selection pressure in chloroplast genomes across different flower types and colors. This clarified the structural variation characteristics at the chloroplast genome level. The chloroplast genomes of the eight flower types ranged in size from 152348 to 152897 bp, all exhibiting typical tetrad structures. The IR boundary of the PP-1 type expanded, while pseudogenization of the ycf1 gene was detected in the PP-5 type. These structural variation characteristics serve as type-specific markers at the chloroplast level.

[0047] In this embodiment of the invention, the differential transcriptome expression analysis of peonies with different flower types and colors includes: S1021, total RNA was extracted from peony using a plant total RNA extraction kit. The total RNA was transcribed into cDNA, and an Illumina standard cDNA library was constructed based on the cDNA. The Illumina standard cDNA library was then sequenced to obtain high-quality reads. Specifically, during plant total RNA extraction, total RNA was extracted from peony petals and petaloid stamens of different flower types and colors using a plant total RNA extraction kit. The integrity and concentration of the 28S / 18S bands were detected by 1% agarose gel electrophoresis, and then analyzed using a micro spectrophotometer. The A260 / A280 ratio and A260 / A230 ratio were measured, with quality control standards of 1.8-2.0 and ≥2.0, respectively. For sequencing Illumina standard cDNA libraries to obtain high-quality reads, quality-controlled total RNA was used as a template. mRNA was enriched using Oligo(dT) magnetic beads, followed by fragmentation, reverse transcription to synthesize the first strand of cDNA, synthesis of the second strand of cDNA, end repair, A-tailing, ligation of sequencing adapters, and PCR amplification to construct an Illumina standard cDNA library. After the library was tested for insert size and concentration using an Agilent 2100 Bioanalyzer, paired-end 150bp high-throughput sequencing was performed using the Illumina NovaSeq platform. The raw sequencing data underwent FastQC quality control to remove adapter sequences and low-quality reads, resulting in high-quality clean reads. In S1022, high-quality sequencing reads were aligned to the peony reference genome, and transcript assembly and expression quantification were performed to obtain the reference gene expression level. Differential expression analysis was conducted based on the reference gene expression level to obtain differential expression values. Differentially expressed genes were screened using differential expression thresholds, and functional annotation and pathway enrichment analysis were performed on the differentially expressed genes of peony plants. Specifically, HISAT2 v2.2.1 was used to align clean reads to the peony reference genome, and StringTiev 2.1.4 was used for transcript assembly and expression quantification (FPKM value). S1023, based on functional annotation and pathway enrichment analysis of differentially expressed genes, differentially expressed genes associated with stamen petalization were screened, as well as structural genes and transcription factors associated with flower color formation, to obtain key expression genes after screening. The expression levels of key expression genes screened by transcriptome were verified by qRT-PCR technology. The peony Actin gene was used as an internal reference gene, and the SYBR Green method was used for quantitative fluorescence. Three technical replicates were set for each sample to verify the reliability of key expression genes, and the peony plants corresponding to the key expression genes were labeled.

[0048] Among them, differentially expressed genes related to flower shape development were screened based on the results of genetic information differences, including AGAMOUS and AGL9 genes; structural genes related to flower color formation included flavonol synthase and chalcon synthase genes (CHS); and transcription factors related to flower color formation included R2R3-MYB transcription factor and SPL9 transcription factor.

[0049] In this embodiment of the invention, when verifying the expression levels of key expression genes screened from the transcriptome using qRT-PCR technology, the SYBR Green method was employed. The reaction system consisted of 20 μL: 10 μL of 2×SYBR Green PCR Master Mix, 0.8 μL each of forward and reverse primers (10 μmol / L), 1.0 μL of cDNA template, and 7.4 μL of ddH2O. The reaction program was: 95℃ pre-denaturation for 30 s; 95℃ denaturation for 5 s, 60℃ annealing for 34 s, for 40 cycles. Three technical replicates were performed for each sample.

[0050] In this embodiment of the invention, when performing transcriptome differential expression analysis on peonies of different flower types and colors, a total of 2847 differentially expressed genes were screened. Among them, the AG gene, which is related to stamen petalization, was significantly downregulated in double-flowered types, showing a significant difference compared to single-flowered types. The expression level of the R2R3-MYB transcription factor was significantly higher in purple-flowered types than in white-flowered types, consistent with the trend of anthocyanin accumulation.

[0051] In this embodiment of the invention, the construction of a genetic fingerprint based on flower pattern and color phenotype by combining nuclear genome polymorphism analysis results and key expressed genes includes: S1031. Obtain the results of nuclear genome polymorphism analysis, identify the genetic distance, SNP fingerprint patterns, and structural variation features of chloroplast genomes for different flower types and colors of peonies in the nuclear genome polymorphism analysis results, and obtain the qRT-PCR expression level data of key expression genes. Standardize the nuclear genome polymorphism analysis results and the qRT-PCR expression level data of key expression genes. In the standardization process, range normalization is performed on the genetic distance, chloroplast genome structural variation features, genetic distance matrix and similarity coefficient matrix in SNP fingerprint patterns for different flower types and colors of peonies, and Z-score standardization is performed on the qRT-PCR expression level data to obtain standardized multidimensional genetic data. S1032 employs a hierarchical weighted integration strategy to assign weights to the distribution of multi-dimensional genetic data. The weighted multi-dimensional genetic data are then linearly fused according to their weights to obtain a comprehensive genetic distance matrix. Specifically, when assigning weights to the multi-dimensional genetic data distribution, the microsatellite Nei genetic similarity coefficient matrix is ​​used as a benchmark. The Mantel correlation coefficient r between the genetic distance matrix of the SNP fingerprint and the microsatellite matrix is ​​calculated to be 0.91. Weighting coefficients are set as follows: W1 = 0.60 (nuclear genome polymorphism), W2 = 0.25 (chloroplast structural variation), and W3 = 0.15 (functional gene expression). S1033, based on the comprehensive genetic distance matrix, hierarchical clustering using the UPGMA method was employed to construct genetic clustering trees for different flower patterns and colors. Molecular identifier combinations were labeled on the genetic clustering trees, and a genetic fingerprint map based on flower pattern and color phenotypes was output, with flower pattern and color type as the index unit. During the construction of the genetic clustering trees for different flower patterns and colors, 1000 Bootstrap resampling iterations were used to evaluate branch support. When all individuals of the same flower pattern and color type clustered in the same branch and the Bootstrap support was ≥90%, that type was determined to be a unique individual with a stable and consistent genetic background. When a variety or strain is established, and individuals of the same type are scattered across ≥2 branches, or individuals of different types are clustered together in the same branch with a Bootstrap support rate ≥45%, it is determined that there is a mixed provenance, and mixed individuals are marked and excluded. Molecular identifier combinations include ITS haplotype numbers, microsatellite core allele combinations, SNP characteristic allele combinations (including fixed differential SNPs in closely related groups and conserved common SNPs in distant groups), chloroplast IR boundary variation pattern codes, AG / AGL9 expression level, CHS expression level, R2R3-MYB expression level, and SPL9 expression level.

[0052] In this embodiment of the invention, by integrating multidimensional molecular marker technologies such as ITS sequence analysis, SSR microsatellite markers, SNP targeted screening, chloroplast whole genome sequencing, and transcriptome differential expression analysis, the genetic background of double / single-petaled and different colored medicinal peonies was analyzed, and a comprehensive genetic fingerprint map including nuclear gene sequence polymorphism, chloroplast genome structural variation, and functional gene expression differences was constructed. The genetic fingerprint map uses Bootstrap resampling to assess branch support rate, and a threshold of ≥90% is used to determine independent varieties or lines with stable and consistent genetic backgrounds. This effectively identifies and excludes individuals with mixed germplasm, thereby solving the technical bottleneck of traditional phenotypic observation being unable to distinguish between genetic stability and hybridization. This provides a reliable molecular-level a priori basis for subsequent precise screening guided by genetic information.

[0053] Example 3: Multidimensional Quality Evaluation In this embodiment of the invention, during multi-dimensional quality evaluation, the genetic fingerprint spectrum is used as a priori to perform multi-dimensional quality evaluation on at least one group of target variety individual plants. The multi-dimensional quality evaluation includes efficacy evaluation, yield evaluation, and medicinal material quality evaluation to obtain the multi-dimensional evaluation results of the target variety individual plants. Specifically, methods for multi-dimensional quality evaluation of at least one group of target variety individual plants include: S201, using genetic fingerprinting as a priori basis, select target variety plants with flower shape and color characteristics from flower shape and color types that are determined to have stable and consistent genetic backgrounds; the representative flower shape and color characteristics include: the flower shape is double or single, and the plant is in normal growth status and free from pests and diseases; the number of target variety plants selected for each flower shape and color type is not less than 20, and the genetic fingerprinting index number of each plant is recorded; S202, evaluate the yield of a single plant of the target variety to obtain the yield evaluation value of a single plant of the target variety; When evaluating the yield of a single plant of the target variety, the product evaluation includes above-ground trait indicators and underground yield indicators. By combining the above-ground trait indicators and underground yield indicators, the weight of each indicator is determined using the entropy weight method to obtain the yield evaluation value of a single plant of the target variety. The above-ground trait indicators include stem height (cm, measured from the ground to the top of the inflorescence using a tape measure), number of leaves (leaf / plant, counting all fully unfolded leaves), number of flowers (flower / plant, counting all open flowers), and number of fruits of single-petaled flower type. The underground yield indicators include root fresh weight (g / plant, weighed using an electronic balance) and dry weight (g / plant, dried at 60℃ to constant weight), number of taproots (roots with a diameter ≥0.5cm / plant), root width (cm, maximum root spread), number of buds (buds / plant, number of robust buds), length of the longest taproot (cm), and diameter of the thickest taproot (cm).

[0054] In this embodiment of the invention, when the entropy weight method is used to determine the weight of a single indicator, the weights of the single indicators are as follows: stem height weight 0.08, number of leaves weight 0.06, number of flowers weight 0.05, number of fruits weight 0.03, fresh weight of roots weight 0.18, dry weight weight 0.25, number of taproots weight 0.15, root spread weight 0.10, length of the longest taproot weight 0.05, and diameter of the thickest taproot weight 0.05.

[0055] In this embodiment of the invention, the weight of each indicator is determined based on the entropy weight method. After the range normalization of each indicator of each target plant, the weighted sum is obtained to obtain the yield evaluation value of each plant. Table 3 shows the mean yield evaluation results of target plants of different flower types and colors in this embodiment of the invention.

[0056] Table 3 As shown in Table 4, the yield evaluation index of individual plants with different flower shapes and colors is continuously distributed. There are individuals with higher and lower yield evaluation values ​​in both single-petal and double-petal flower types, indicating that the yield advantage is not determined by the single flower shape and color type. Therefore, fine screening at the individual level is required.

[0057] S203. The quality of the target variety is evaluated on individual plants to obtain a comprehensive quality evaluation value. When evaluating the quality of the target variety, the following methods are used: hyperspectral analysis, UPLC-MS chemical composition identification, DESI-MSI mass spectrometry imaging analysis, UPLC multi-component content determination, and qNMR total glycoside content determination. After normalizing the measurement results, the grey relational analysis method is used to calculate the correlation degree between the target variety and the ideal high-quality medicinal material reference sequence, and the correlation degree is used as the comprehensive quality evaluation value of the medicinal material. In this embodiment of the invention, during hyperspectral analysis, peony root powder was spread evenly on a petri dish, and hyperspectral image data in the 400-1000nm band was collected. After standard normal variable correction, principal component analysis and partial least squares method were used to extract characteristic bands, and the differences in overall chemical information of peony roots of different flower types and colors were compared.

[0058] For chemical component identification by UPLC-MS, peony root powder was extracted with 50% ethanol by ultrasonic extraction, filtered through a 0.22 μm microporous membrane, and detected by tandem Q-TOF mass spectrometry using a Waters ACQUITY UPLC system. The chromatographic column was ACQUITY BEHC18 (2.1 mm × 100 mm, 1.7 μm), the mobile phase was an aqueous solution containing 0.1% formic acid-acetonitrile system, gradient elution was used, the flow rate was 0.3 mL / min, the column temperature was 30 °C, and the electrospray ionization source was used in negative ion mode for scanning. The chemical components corresponding to the chromatographic peaks were identified by comparison with the database using the UNIFI scientific information system.

[0059] For DESI-MSI mass spectrometry imaging analysis, peony roots were transversely sectioned into frozen sections with a thickness of 10-20 μm, attached to conductive glass slides, and mass spectrometry imaging was performed using the DESI-MSI system in positive ion mode. The spray solvent was methanol-water (1:1, v / v), the spray voltage was 4.5 kV, the sampling rate was 100 μm / s, and the spatial resolution was 50-100 μm. MassLynx software was used for data acquisition, and HDI software was used for base peak intensity normalization and ion image reconstruction to obtain the spatial distribution heatmap of paeoniflorin, paeoniflorin lactone, oxidized paeoniflorin, and benzoyl paeoniflorin on the root transverse section.

[0060] For UPLC multi-component content determination, peony root powder was extracted with 50% ethanol using ultrasound. The same UPLC chromatographic conditions as used for UPLC-MS chemical component identification were applied, and the contents of pentagalloyl glucose, paeoniflorin, and hydroxypaeoniflorin were simultaneously determined at a detection wavelength of 230 nm. For qNMR total glycoside content determination, 20 mg of crude peony total glycosides was accurately weighed, dissolved in 50 μL of 10 mg / mL 1,4-dinitrobenzene internal standard solution and 550 μL of deuterated methanol. ¹H-NMR quantitative analysis was performed using a Bruker 700 MHz nuclear magnetic resonance spectrometer with a relaxation time ≥10 s and ≥64 scans. The total peony glycoside content was calculated using the absolute quantitative method of the Chinese Pharmacopoeia.

[0061] S204. The expression levels of key genes in individual plants of the target variety were detected, and the contents of pengaloyl glucose, paeoniflorin, and hydroxypaeoniflorin were detected by UPLC-MS chemical composition identification method to obtain the total amount of medicinal components. The expression levels of key genes and the total amount of medicinal components were positively normalized, and the entropy weight method was used to determine the weights of the expression levels of key genes and the total amount of medicinal components to obtain the comprehensive value of medicinal components. In this process, the expression levels of key expression genes and the total amount of medicinal components are positively normalized. When determining the weights of key expression gene expression levels and the total amount of medicinal components using the entropy weight method, the expression levels of key expression genes (absolute value after Z-score standardization) and the total amount of medicinal components are positively normalized to the [0,1] interval, and the weights of the two are determined using the entropy weight method (key expression gene weight 0.45, total amount of medicinal components weight 0.55). The weighted sum is then used to obtain the comprehensive value of medicinal components for each individual plant. Table 4 shows the test results of the comprehensive value of medicinal components for different flower types and colors in the embodiments of this invention.

[0062] Table 4 As shown in Table 4, the expression level of the AG gene was significantly downregulated in the double-flowered type (Z-score was negative), while the CHS gene was highly expressed in all types. The total amount of medicinal components was significantly positively correlated with the expression level of CHS, indicating that the expression level of genes related to flower color formation can predict the accumulation capacity of medicinal components in a single plant to a certain extent.

[0063] S205 integrates the yield evaluation value, comprehensive quality evaluation value, and comprehensive value of medicinal components of a single plant of the target variety into a multi-dimensional quality evaluation result, and stores it in association based on the genetic fingerprint index number of a single plant of the target variety.

[0064] In this embodiment of the invention, the obtained yield evaluation value, comprehensive medicinal material quality evaluation value, and comprehensive medicinal component value are linked one-to-one according to the genetic fingerprint index number of each individual plant to construct a multi-dimensional quality evaluation database. The multi-dimensional quality evaluation database contains 240 records (8 types × 30 plants), and each record includes: index number, flower type and color code, yield evaluation value, comprehensive medicinal material quality evaluation value, comprehensive medicinal component value, and corresponding molecular identifier code, providing a data foundation for the subsequent construction of regression screening models.

[0065] This invention uses genetic fingerprinting as a screening prerequisite constraint to simultaneously conduct three-dimensional evaluations of target varieties with verified genetic backgrounds, including yield evaluation, medicinal quality evaluation, and efficacy evaluation. Yield evaluation covers 10 indicators of aboveground agronomic traits and underground root yield, objectively weighted using entropy weighting. Medicinal quality evaluation integrates hyperspectral analysis, UPLC-MS chemical composition identification, DESI-MSI mass spectrometry imaging, UPLC multi-component content determination, and qNMR total glycoside content determination, quantifying the correlation using grey relational analysis. Efficacy evaluation integrates key gene expression levels and core efficacy component content, comprehensively using entropy weighting. This multi-dimensional evaluation system overcomes the limitations of existing single-indicator evaluations, achieving full-chain quality control from genetic stability determination, phenotypic yield measurement, qualitative and quantitative chemical composition analysis, and efficacy material basis analysis, ensuring that selected individual plants possess comprehensive traits of high yield, high quality, and high efficacy.

[0066] Example 4: Construction of Regression Screening Model and Screening of Superior Varieties In the selection of superior varieties, multi-dimensional evaluation results of individual plants of the target variety are obtained. Combined with the modal correlation matrix of flower shape and color, multi-dimensional evaluation results and genetic fingerprint, a regression screening model is constructed based on the modal correlation matrix to calculate the comprehensive closeness score of the target variety and select superior varieties.

[0067] The calculation of the comprehensive similarity score of the target variety includes: S301, using the flower shape and color of a single plant of the target variety as the row dimension, the multi-dimensional evaluation results of a single plant of the target variety as the column dimension, and combining the molecular identifier combination of the genetic fingerprint as the depth dimension, constructs a third-order data tensor to characterize the excellence of a single plant of the target variety. In this embodiment of the invention, the third-order data tensor is represented as: , These are the evaluation indicators and molecular identifiers for column-dimensional and multi-dimensional evaluation results, respectively.

[0068] S302, using high-order singular value decomposition (HOSVD) to reduce the dimensionality of the third-order data tensor, obtains the core tensor and three sets of modal correlation matrices: flower type / color-evaluation index matrix, flower type / color-genetic information matrix, and evaluation index-genetic information matrix. Factor matrices corresponding to the modal correlation matrices are extracted, and the factor matrices are solved using the HOSVD algorithm. The top r principal components with a cumulative variance contribution rate ≥ 85% are retained. Based on the element values ​​of the core tensor, the collaborative correlation patterns among the three modalities of flower type / color, evaluation index, and genetic information are identified. When the element value is greater than the preset significance threshold (0.2), it is determined that there is a significant synergistic association between flower type, evaluation index, and genetic information, and the significant association triplet is recorded. In this embodiment, r=5 (flower type and color modality), r=2 (evaluation index modality), and r=3 (genetic information modality) have cumulative variance contribution rates of 88.7%, 90.2%, and 86.5%, respectively. In this embodiment, a total of 27 significant association triplets were identified, covering all 8 flower type and color types, indicating that each type has molecular marker combinations that are synergistically expressed with yield, quality, and efficacy indicators. Table 5 shows some examples of significant association triplets in this embodiment.

[0069] Table 5 Table 5 shows that significantly associated triplets are evenly distributed across all eight flower type and color types, exhibiting both positive associations (high expression of molecular markers / specific SNPs synergistically with high yield, high quality, and high efficiency) and negative associations (low AGL9 expression synergistically with high efficiency). This further confirms that the genetic basis of desirable traits is distributed across different flower type and color types, and there is no situation where one type is completely dominant or completely inferior.

[0070] S303, a reference sequence of ideal high-quality varieties is pre-constructed. The entropy weight-TOPSIS method is used to calculate the comprehensive closeness score between the target variety and the ideal reference sequence. The comprehensive closeness score between the target variety and the ideal reference sequence is used as the dependent variable, and the principal component scores of flower type and color type, the principal component scores of evaluation indicators extracted from the factor matrix, and the quantitative coding values ​​of molecular identifier dimension are used as independent variables to construct a multivariate nonlinear regression screening model. in, , representing the comprehensive similarity score and the basic predicted value of the comprehensive similarity between the i-th target variety single plant and the ideal reference sequence, respectively. This represents the regression coefficient, the m-th original independent variable (the principal component score of flower pattern and color type, the principal component score of the evaluation index extracted from the factor matrix, and the quantitative encoding value of the molecular identifier dimension). Represents the random error term. Represents a nonlinear transformation function; In the process of pre-constructing the reference sequence for ideal high-quality varieties, for each flower type and color type, significantly associated individuals in the significantly associated triplet within that flower type and color type are selected whose yield evaluation value, comprehensive medicinal material quality evaluation value, and comprehensive medicinal component value are all in the top 10 percentile. The mean of the multi-dimensional evaluation results of the significantly associated individuals is used as the ideal reference value for the flower type and color type. At the same time, the common molecular marker features of the flower type and color type are extracted as ideal genetic markers. The ideal high-quality variety reference sequence is constructed by combining the ideal reference value and the ideal genetic marker.

[0071] S304. The least squares method is used to estimate the regression coefficients and a significance test is performed. When the adjusted coefficient of determination of the model is ≥0.80 and the significance test is <0.01, the regression screening model is considered to be effective. S305: Substitute the multi-dimensional evaluation results of the target varieties into the fitted and effective regression screening model, calculate the predicted comprehensive similarity score, sort the comprehensive similarity scores from high to low, and select the top 3 flower type and color type as candidates for superior varieties; for all target variety individual plants within the candidate types of superior varieties, sort them according to the comprehensive similarity score, and select the top 30% of individuals as core breeding materials to complete the screening of superior varieties.

[0072] Table 6 shows the distribution of excellent single plant candidates in different flower shapes and colors.

[0073] Table 6 Table 6 shows that candidate superior individual plants are distributed across all flower types and colors, but the distribution ratios differ. Double-petaled white (PP-1), single-petaled deep purple-pink (PP-4), and single-petaled bright purple-red (PP-5) types had the highest selection ratios (40.0%), while single-petaled white (PP-8) had a relatively lower selection ratio. This result indicates that superior individual plants are not concentrated in a specific flower type or color, but are dispersed across multiple types, confirming the necessity and effectiveness of cross-type screening. Furthermore, from the perspective of dominance dimensions, the main dominance dimensions of core breeding materials for single-petaled flower types are yield and efficacy, while the main dominance dimension of core breeding materials for double-petaled flower types is medicinal quality. This result provides a basis for subsequent targeted breeding for different application needs. In addition, the above candidate category ranking results only represent the specific test results of this sample of 240 plants and should not be considered as an absolute judgment of the superiority or inferiority of a particular flower type or color. In practical applications, as the sample size increases and more generations of data accumulate, the composition of candidate categories may change, and the individual-level screening conclusions may also be adjusted. The value of this method lies in providing a quantifiable, reproducible, and iterative screening framework, rather than providing a one-time ranking of types.

[0074] This invention constructs a third-order data tensor from flower shape and color phenotypes, multi-dimensional evaluation results, and molecular identifiers from genetic fingerprinting. It employs High-Order Singular Value Decomposition (HOSVD) to extract modal association matrices and identify significant co-associated triples. Then, it combines the entropy-weighted TOPSIS method to calculate a comprehensive proximity score, constructing a multivariate nonlinear regression screening model with principal component scores of flower shape and color, principal component scores of evaluation indicators, and quantitative coding values ​​of molecular identifiers as independent variables. This regression screening model transforms traditional experience-based selection into quantitative screening based on multi-source data fusion. It can overcome the limitations of single flower shape and color types in all target varieties, enabling precise cross-type individual-level screening. The selected core breeding materials cover multiple flower shape and color types, each with its own differentiated advantages, shortening the breeding cycle and providing diversified high-quality germplasm resources for subsequent targeted breeding and hybridization for different application needs.

[0075] Example 5 This invention provides a method for screening superior peony varieties based on genetic information of flower shape and color. Figure 4 This diagram illustrates the implementation flow of a method for screening superior peony varieties based on genetic information of flower shape and color. The method specifically includes: S10. During the peak blooming period of peonies, peonies were classified into two types according to flower shape: double-petaled and single-petaled, and the flower shape was marked. The flower color was also classified according to petal color and marked. Leaf and petal tissues were collected from the marked peony plants of different flower shapes and colors as test samples. The flower colors are divided into: double-petaled white, double-petaled deep purple-pink, double-petaled light purple-pink, single-petaled deep purple-pink, single-petaled bright purple-red, single-petaled purple-pink, single-petaled light purple-pink, and single-petaled white. S20. Complete leaf and petal tissues of peonies with different flower shapes and colors were obtained, and genomic DNA and total RNA of peonies were extracted. Based on the extracted genomic DNA and total RNA, multidimensional genetic information analysis of peonies was performed to establish a genetic fingerprint map based on flower shape and color phenotype. The genetic fingerprint map is an independent variety / strain with a stable genetic background, and the genetic fingerprint map of hybrid individuals is excluded. S30, using genetic fingerprinting as a priori, conduct multi-dimensional quality evaluation on at least one group of target variety individual plants, wherein the multi-dimensional quality evaluation includes efficacy evaluation, yield evaluation, and medicinal material quality evaluation, to obtain multi-dimensional evaluation results of target variety individual plants; S40: Obtain multi-dimensional evaluation results of individual plants of the target variety. Combine flower shape and color, multi-dimensional evaluation results, and modal correlation matrix of genetic fingerprint. Construct a regression screening model based on the modal correlation matrix, calculate the comprehensive closeness score of the target variety, and screen out superior varieties.

[0076] In summary, this invention provides a method for screening superior peony varieties based on genetic information of flower shape and color. This invention systematically analyzes the genetic background of double / single petals and different flower colors using multidimensional molecular markers such as ITS, SSR, SNP, chloroplast whole genome, and transcriptome, effectively eliminating mixed individuals. Furthermore, using genetic fingerprinting as a screening prerequisite, it conducts multidimensional and three-dimensional evaluations of yield, medicinal quality, and efficacy on selected individual plants. By utilizing higher-order tensor decomposition and entropy weight-TOPSIS methods, it deeply integrates flower color phenotype, molecular data, and quality indicators to quantitatively assess the comprehensive similarity of each individual plant, thereby accurately identifying high-yielding, high-quality, and highly effective superior varieties. This transforms traditional experience-based selection into quantitative screening based on multi-source data fusion, shortening the breeding cycle and achieving precise identification at the individual level, providing reliable technical support for the breeding of superior medicinal peony varieties.

[0077] It should be noted that, for the sake of simplicity, the foregoing embodiments are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to the present invention. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0078] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit the scope of protection of the invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on these embodiments, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art can still combine, add, delete, or otherwise adjust the features of the various embodiments of the present invention according to the circumstances without conflict or creative effort, thereby obtaining different technical solutions that do not fundamentally depart from the concept of the present invention. These technical solutions are also within the scope of protection of the present invention.

Claims

1. A method for screening superior peony varieties based on genetic information of flower shape and color, characterized in that, The method includes: S10. During the peak blooming period of peonies, peonies were classified into two types according to flower shape: double-petaled and single-petaled, and the flower shape was marked. The flower color was also classified according to petal color and marked. Leaf and petal tissues were collected from the marked peony plants of different flower shapes and colors as test samples. S20. Complete leaf and petal tissues of peonies with different flower shapes and colors were obtained, and genomic DNA and total RNA of peonies were extracted respectively. Based on the extracted genomic DNA and total RNA, multidimensional genetic information analysis of peonies was carried out to establish a genetic fingerprint map based on flower shape and color phenotype. S30, using genetic fingerprinting as a priori, conduct multi-dimensional quality evaluation on at least one group of target variety individual plants, wherein the multi-dimensional quality evaluation includes efficacy evaluation, yield evaluation, and medicinal material quality evaluation, to obtain multi-dimensional evaluation results of target variety individual plants; S40: Obtain multi-dimensional evaluation results of individual plants of the target variety. Combine flower shape and color, multi-dimensional evaluation results, and modal correlation matrix of genetic fingerprint. Construct a regression screening model based on the modal correlation matrix, calculate the comprehensive closeness score of the target variety, and screen out superior varieties.

2. The method for screening superior peony varieties based on flower shape and color genetic information as described in claim 1, characterized in that: The multidimensional genetic information analysis of peony included: Based on peony genomic DNA, nuclear genome polymorphism analysis was performed on peonies with different flower types and colors to obtain genetic distances, SNP fingerprints, and chloroplast whole genome sequences of peonies with different flower types and colors, and to generate nuclear genome polymorphism analysis results. Based on total RNA from peonies, differential transcriptome expression analysis was performed on peonies with different flower shapes and colors to obtain genetic information differences. Based on the genetic information differences, differentially expressed genes related to flower shape development and structural genes and transcription factors related to flower color formation were screened. Genetic fingerprints based on flower pattern and color phenotypes were constructed by combining nuclear genome polymorphism analysis results and key expressed genes.

3. The method for screening superior peony varieties based on flower shape and color genetic information as described in claim 2, characterized in that: The nuclear genome polymorphism analysis of peonies with different flower types and colors includes: The internal transcribed spacer region sequence of ribosomal DNA was amplified by PCR to obtain the ITS amplified sequence. The ITS amplified sequence was then subjected to Sanger sequencing. The sequencing results were compared with the GenBank database to calculate the genetic distance of peonies with different flower shapes and colors. Based on peony genomic DNA, peony-specific SSR primer pairs were designed and synthesized. PCR amplification of peony genomic DNA of different flower types and colors was performed. Microsatellite fingerprints of each flower type and color type were constructed using allele fragment length polymorphism data. The Nei genetic similarity coefficient method was used to calculate the genetic similarity coefficient. Combining the genetic similarity coefficient matrix and clustering results, flower type and color type groups with similar genetic backgrounds were identified, and the genetic boundaries of the flower type and color type groups were determined. Using the genetic boundary of flower shape and color type grouping as a prior constraint, similar groups and distant groups of different flower shape and color types are determined based on the genetic boundary. For similar groups, fixed differential SNPs within the group are screened, and for distant groups, conserved common SNPs within the group are screened. A pre-constructed paired-end sequencing library was used to mine candidate SNP sites based on simplified genome sequencing data, resulting in a set of targeted SNP markers. SNP genotyping was performed on the target SNP marker set, an SNP fingerprint map was constructed based on the SNP genotyping results, and the genetic distance matrix was calculated using the Nei genetic distance method. A chloroplast genome map was constructed based on genomic DNA, and structural variation features of chloroplast genomes of different flower types and colors were extracted by combining the chloroplast genome map.

4. The method for screening superior peony varieties based on flower shape and color genetic information as described in claim 3, characterized in that: The transcriptome differential expression analysis of peonies with different flower types and colors includes: Total RNA was extracted from peony using a plant total RNA extraction kit. The total RNA was transcribed into cDNA. An Illumina standard cDNA library was constructed based on the cDNA. The Illumina standard cDNA library was sequenced to obtain high-quality sequencing reads. High-quality sequencing reads were aligned to the peony reference genome, and transcripts were assembled and their expression levels were quantified to obtain the reference gene expression level. Differential expression analysis was performed based on the reference gene expression level to obtain differential expression values. Differential expression genes were screened using differential expression thresholds, and functional annotation and pathway enrichment analysis were performed on the differentially expressed genes of peony plants. Based on functional annotation and pathway enrichment analysis of differentially expressed genes, differentially expressed genes associated with stamen petalization were screened, and structural genes and transcription factors associated with flower color formation were also screened to obtain the key expressed genes.

5. The method for screening superior peony varieties based on flower shape and color genetic information as described in claim 4, characterized in that: Based on the results of genetic information differences, differentially expressed genes related to flower shape development were identified, including the AGAMOUS gene and the AGL9 gene; structural genes related to flower color formation included the flavonol synthase and chalcone synthase gene CHS; and transcription factors related to flower color formation included the R2R3-MYB transcription factor and the SPL9 transcription factor.

6. The method for screening superior peony varieties based on flower shape and color genetic information as described in claim 4, characterized in that: The construction of a genetic fingerprint based on flower shape and color phenotypes, combining nuclear genome polymorphism analysis results and key expressed genes, includes: The results of nuclear genome polymorphism analysis were obtained, and the genetic distance, SNP fingerprints, and structural variation features of chloroplast genomes of different flower types and colors of peonies were identified. The qRT-PCR expression levels of key expression genes were also obtained, and the results of nuclear genome polymorphism analysis and qRT-PCR expression levels of key expression genes were standardized. A hierarchical weighted integration strategy is adopted to assign weights to the distribution of multi-dimensional genetic data, and then the weighted multi-dimensional genetic data are linearly fused according to the weights to obtain a comprehensive genetic distance matrix. Based on the comprehensive genetic distance matrix, hierarchical clustering was performed using the UPGMA method to construct genetic clustering trees for different flower patterns and colors. Molecular identifier combinations were labeled on the genetic clustering trees, and a genetic fingerprint map based on flower pattern and color phenotypes was output with flower pattern and color type as the index unit.

7. The method for screening superior peony varieties based on flower shape and color genetic information as described in claim 3, characterized in that: The multi-dimensional quality evaluation includes efficacy evaluation, yield evaluation, and medicinal material quality evaluation, resulting in multi-dimensional evaluation results for individual plants of the target variety. Methods for multi-dimensional quality evaluation of at least one group of target variety individual plants, including: Based on the genetic fingerprint profile, target varieties with flower shape and color characteristics were selected from flower shape and color types that were determined to have stable and consistent genetic backgrounds. The yield of a single plant of the target variety is evaluated to obtain the yield evaluation value of a single plant of the target variety. The quality of medicinal materials was evaluated on individual plants of the target variety to obtain a comprehensive quality evaluation value. When evaluating the quality of medicinal materials on individual plants of the target variety, hyperspectral analysis, UPLC-MS chemical composition identification, DESI-MSI mass spectrometry imaging analysis, UPLC multi-component content determination, and qNMR total glycoside content determination were used. After normalizing the measurement results, the correlation degree between the individual plants of the target variety and the ideal high-quality medicinal material reference sequence was calculated using the grey relational analysis method, and the correlation degree was used as the comprehensive quality evaluation value of the medicinal materials. The expression levels of key genes in individual plants of the target variety were detected, and the contents of pengaloyl glucose, paeoniflorin, and hydroxypaeoniflorin were detected by UPLC-MS chemical composition identification method to obtain the total amount of medicinal components. The expression levels of key genes and the total amount of medicinal components were positively normalized, and the entropy weight method was used to determine the weights of the expression levels of key genes and the total amount of medicinal components to obtain the comprehensive value of medicinal components. The yield evaluation value, comprehensive quality evaluation value, and comprehensive value of medicinal components of the target variety are integrated into a multi-dimensional quality evaluation result, and stored in association based on the genetic fingerprint index number of the target variety.

8. The method for screening superior peony varieties based on flower shape and color genetic information as described in claim 7, characterized in that: When evaluating the yield of a single plant of a target variety, the product evaluation includes aboveground trait indicators and underground yield indicators. By combining the aboveground trait indicators and underground yield indicators, the weight of each indicator is determined using the entropy weight method to obtain the yield evaluation value of a single plant of the target variety. The aboveground trait indicators include stem height, number of leaves, number of flowers, and number of fruits of single-petaled flower type. The underground yield indicators include fresh and dry weight of roots, number of taproots, root width, number of buds, length of the longest taproot, and diameter of the thickest taproot.

9. The method for screening superior peony varieties based on flower shape and color genetic information as described in claim 7, characterized in that: The calculation of the comprehensive similarity score of the target variety includes: Using the flower shape and color of individual plants of the target variety as the row dimension, the multi-dimensional evaluation results of individual plants of the target variety as the column dimension, and the molecular identifier combination of the genetic fingerprint as the depth dimension, a third-order data tensor representing the excellence of individual plants of the target variety is constructed. The third-order data tensor was decomposed by high-order singular value decomposition to obtain the core tensor and three sets of modal correlation matrices. The factor matrix was solved by the HOSVD algorithm, and the first r principal components with a cumulative variance contribution rate ≥85% were retained. Based on the element values ​​of the core tensor, the synergistic correlation patterns among the three modalities of flower pattern and color type, evaluation index and genetic information were identified, and significant correlation triplets were recorded. An ideal high-quality variety reference sequence was pre-constructed. The entropy weight-TOPSIS method was used to calculate the comprehensive closeness score between the target variety and the ideal reference sequence. The comprehensive closeness score between the target variety and the ideal reference sequence was used as the dependent variable, and the principal component scores of flower type and color type, the principal component scores of evaluation indicators extracted from the factor matrix, and the quantitative coding values ​​of molecular identifier dimension were used as independent variables to construct a multivariate nonlinear regression screening model. Substitute the multi-dimensional evaluation results of the target varieties into the fitted regression screening model to calculate the predicted comprehensive similarity score. Sort the varieties by comprehensive similarity score from high to low and select the top 3 flower type and color type as candidates for superior varieties. For all target variety individual plants within the candidate types of superior varieties, sort them by comprehensive similarity score and select the top 30% of individuals as core breeding materials to complete the screening of superior varieties.

10. The method for screening superior peony varieties based on flower shape and color genetic information as described in claim 9, characterized in that: When constructing the reference sequence for ideal high-quality varieties, for each flower type and color type, significantly associated individuals in the significantly associated triplet within that flower type and color type are selected whose yield evaluation value, comprehensive medicinal material quality evaluation value, and comprehensive medicinal component value are all in the top 10 percentile. The mean of the multi-dimensional evaluation results of the significantly associated individuals is used as the ideal reference value for the flower type and color type. At the same time, the common molecular marker features of the flower type and color type are extracted as ideal genetic markers. The ideal reference sequence for high-quality varieties is constructed by combining the ideal reference value and the ideal genetic markers.

Images