A method for screening quantitative classification traits of crabapple germplasm

Abstract

This invention designs a method for screening traits in the quantitative classification of ornamental plant germplasm, mainly consisting of the following steps: trait selection and coding, intra-germplasm consistency test, inter-germplasm discrimination analysis, principal component analysis, and Pearson correlation analysis. The scientific validity and effectiveness of the trait screening technology system proposed in this invention are verified by including species in the population sample and evaluating the kinship clustering probability distribution between species, species and varieties, and varieties, respectively, thereby indirectly reflecting the effectiveness of the trait screening technology system proposed in this invention.

Description

Technical Field

[0001] This invention relates to the field of plant breeding, and particularly to a method for screening quantitative classification traits of crabapple germplasm. Background Technology

[0002] Genetic diversity assessment and specific germplasm mining are fundamental requirements for the innovation of ornamental plant resources. These can be achieved through evaluation of morphological, phenological, and agronomic traits, as well as analysis using biochemical and molecular markers. Among these, morphological parameters are one of the most important factors determining the taxonomic and agronomic value of plants. Since information obtained from morphological characteristics comes from large datasets composed of qualitative and quantitative features, multivariate analysis is particularly favored. Quantitative taxonomy, as a type of multivariate analysis, uses quantitative methods to evaluate the morphological similarity between taxa, accelerating the application of germplasm resources in breeding. However, the objectivity of its classification results is largely affected by the selection of morphological traits. Currently, in quantitative taxonomy research on ornamental plant germplasm both domestically and internationally, trait selection either artificially defines the number of selected traits or often employs principal component analysis in multivariate statistical analysis for dimensionality reduction, supplemented by one-way ANOVA and correlation analysis, but the selection results are not ideal.

[0003] In summary, existing methods for screening quantitative classification traits of ornamental plant germplasm have the following drawbacks: 1. The existing screening principles are purely based on statistical methods and fail to consider the requirements of taxonomic theory for classification traits; 2. They cannot accurately reveal the importance of the investigated traits, thus hindering scientific trait screening; 3. In data preprocessing, the coding of the investigated traits does not fully consider the evolutionary order between different levels of each trait, and does not strictly follow the order from primitive to evolution; 4. To date, a systematic and scientific theory and technology system for screening classification traits has not been established, and screening is only based on some statistical methods. Summary of the Invention

[0004] The purpose of this invention is to provide a scientific and systematic technology system for screening quantitative classification traits of ornamental plant germplasm.

[0005] The technical problem solved by this invention is achieved by the following technical solution.

[0006] This invention proposes a method for screening quantitative classification traits of crabapple germplasm, which includes:

[0007] The first step is the selection and coding of traits;

[0008] The second step is to test the consistency of germplasm.

[0009] The third step is germplasm discrimination analysis;

[0010] The fourth step is principal component analysis and Pearson correlation analysis.

[0011] The beneficial effects of this invention are as follows: The ornamental plant germplasm quantitative classification trait screening technology system provided by this invention, through encoding the measured traits and performing intra-germplasm consistency tests, inter-germplasm discrimination analysis, principal component analysis, and correlation analysis, successfully screens out important traits that can be used for the quantitative classification of ornamental plant germplasm. This achieves a high degree of dimensionality reduction of classification traits and further enables the scientific classification of all germplasm. The operation of this system meets the requirements of taxonomy for classification traits and has strong scientific validity.

[0012] The scientific validity and effectiveness of the trait screening technology system proposed in this invention are verified by including species in the population sample and evaluating the kinship aggregation distribution probability between species, between species and varieties, and between varieties, respectively. This process indirectly reflects the effectiveness of the trait screening technology system proposed in this invention. Attached Figure Description

[0013] Figure 1 This invention provides a germplasm consistency test and inter-germplasm discrimination analysis for the organ phenotypic traits of crabapple flowers in this embodiment. Figure 1 -a represents the germplasm consistency test for qualitative traits of crabapple flower organs (with... (for testing standards); Figure 1 -b represents the germplasm consistency test for quantitative traits of crabapple flower organs (with... (for testing standards); Figure 1 -c represents the germplasm discrimination analysis of qualitative traits of crabapple flower organs (with MF≤k1 / f as the test standard); Figure 1 -d represents the inter-germination discrimination analysis of quantitative traits of crabapple flower organs (with Cv≥15% as the test standard); * indicates that the differences of each quantitative trait among different crabapple germplasms reached a significant level (P<0.05).

[0014] Figure 2 This is a Pearson correlation analysis of the phenotypic traits of crabapple flower organs in an embodiment of the present invention. In the figure, '×' indicates a correlation coefficient r > 0.80 between phenotypic traits of flower organs.

[0015] Figure 3 This is a dendrogram of crabapple germplasm clustering analysis based on important phenotypic traits of floral organs in an embodiment of the present invention. Crabapple germplasm of the same subclass is labeled with the same color, and the dynamic flower images of typical crabapple germplasm of each subclass are presented in the corresponding colored rectangular boxes.

[0016] Figure 4This invention presents an analysis of the affinity aggregation distribution characteristics of Malus spectrograms based on a clustering dendrogram. The groups (lines) in the figure are labeled as follows: Ⅰ is the Malus spectabilis group, Ⅱ is the Malus 'Green Apple' group, Ⅲ is the Malus spectabilis lineage, Ⅳ is the Malus 'Longdong' lineage, Ⅴ is the Malus trifoliata lineage, Ⅵ is the Malus spectabilis lineage, and Ⅶ is the Malus spectabilis lineage. Ⅰ→Ⅶ indicates the evolutionary order of the groups and lines, with Ⅰ representing the primitive group and Ⅶ representing the evolved group. Malus species within the same group or lineage are labeled with the same color. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below. Where specific conditions are not specified in the embodiments, conventional conditions or conditions recommended by the manufacturer shall apply. Reagents or instruments whose manufacturers are not specified are all conventional products that can be purchased commercially.

[0018] The following is a detailed description of a method for screening quantitative classification traits of crabapple germplasm according to an embodiment of the present invention.

[0019] Embodiments of the present invention provide a method for screening quantitative classification traits of crabapple germplasm, comprising:

[0020] The first step is the selection and coding of traits;

[0021] The second step is to test the consistency of germplasm.

[0022] The third step is germplasm discrimination analysis;

[0023] The fourth step is principal component analysis and Pearson correlation analysis.

[0024] Furthermore, in a preferred embodiment of the present invention, in the first step of trait selection and coding, the selection of traits refers to the DUS testing guidelines for new plant varieties of each genera (such as Malus) established by UPOV (International Union for the Protection of New Varieties of Plants), and additional traits are selected based on their ornamental and identification value. Regarding trait coding, qualitative traits are coded using a hierarchical numerical coding method, expanding by consecutive non-negative integers 0, 1, 2, 3, ... in the order from primitive to evolutionary. Dimorphic traits whose evolutionary relationship is difficult to determine are generally coded as 1, otherwise 0. Quantitative traits do not require coding and are directly entered into the next step of calculation in the form of the mean.

[0025] Furthermore, in a preferred embodiment of the present invention, in the second step of germplasm uniformity testing, the average mode frequency is used for the qualitative trait uniformity test. It means (Formula 1), if This trait then meets the consistency requirements. The mean coefficient of variation is used to test the germplasm consistency of quantitative traits. (Formula 2) indicates that if Then the trait meets the consistency requirement.

[0026]

[0027] Where n is the number of germplasm resources, and m is the number of replicates. M0 is the rank of the trait that appears most frequently in each germplasm resource across m replicates, and S... i Let m be the standard deviation of the observations for each trait across m replicates for each quality resource. It represents the average of the observations for each trait across m replicates for each quality resource.

[0028] Furthermore, in a preferred embodiment of the present invention, the third step, germplasm discrimination analysis, uses the degree of discrimination of qualitative traits as a simple measure of the closeness between the mode frequency (MF) of the rank exhibited by each trait in all germplasm resources and the theoretical frequency (1 / f) (Formula 3). If MF ≤ k1 / f, then the qualitative trait has a high degree of discrimination among all germplasm resources. The degree of discrimination of quantitative traits is expressed by the coefficient of variation (Cv) of the mean of each trait in all germplasm resources (Formula 4). If Cv ≥ 15%, then the trait is considered to have a high degree of discrimination among all germplasm resources.

[0029] Furthermore, for quantitative traits, analysis of variance (Tukey's method) is required to determine whether the differences in the quantitative trait among all germplasms reach a significant level.

[0030]

[0031] Where n is the number of germplasm resources, M0' is the rank with the highest frequency of each trait in all germplasm resources, k is a coefficient whose magnitude depends on the number of ranks (f) of each trait in all germplasm resources, and S' is the standard deviation of the observed mean of each trait in all germplasm resources. This is the average of the observed mean values ​​for each trait across all germplasm resources.

[0032] Furthermore, in a preferred embodiment of the present invention, the fourth step involves principal component analysis and Pearson correlation analysis: Under the premise of satisfying high consistency and strong discriminative power, principal component analysis and correlation analysis are used to further reduce the dimensionality of the selected traits, thereby further reducing the number of traits with redundant information. To eliminate the influence of different dimensions on data analysis, the original numerical matrix is ​​first standardized using standard deviation (STD), i.e., normalized.

[0033] The beneficial effects of this plan are as follows:

[0034] 1. The invention of the classification trait screening principle: strong uniformity within germplasm, high differentiation between germplasm, independence between traits, and consideration of both the ornamental value and recognizability of traits.

[0035] 2. The invention of measurement criteria for each stage of screening of classification traits corresponding to classification principles: mean, mode, and frequency. Mean coefficient of variation The degree of closeness between the mode frequency (MF) and the theoretical frequency (1 / f), and the coefficient of variation (Cv) of the mean.

[0036] 3. Methods for verifying the scientific validity and effectiveness of the trait screening technology system proposed in this invention: Incorporate species into the population sample, and evaluate the scientific validity of the quantitative classification results of ornamental plant germplasm by separately testing the kinship aggregation distribution probability between species, between species and varieties, and between varieties, thereby indirectly reflecting the effectiveness of the trait screening technology system proposed in this invention.

[0037] Example

[0038] This embodiment screened the quantitative classification traits of crabapple germplasm based on the phenotypic diversity of floral organs. Specifically:

[0039] The first step is the selection and coding of traits.

[0040] The selection of phenotypic traits of crabapple flower organs was based on the DUS testing guidelines for new varieties of apple plants developed by UPOV (International Union for the Protection of New Varieties of Plants). Additional traits were selected based on their ornamental and identification value, for a total of 44 traits. The trait descriptions and codes are shown in Table 1.

[0041]

[0042]

[0043] The second step is to test the consistency of germplasm.

[0044] The intragerm consistency test of qualitative traits of crabapple flower organs (see Table 1) was performed using the mean mode frequency. It means (Formula 1), if Then the trait meets the consistency requirement.

[0045] The mean coefficient of variation was used to test the intragerm uniformity of quantitative traits (see Table 1). (Formula 2) indicates that if Then the trait meets the consistency requirement.

[0046]

[0047] Where n is the number of germplasm resources, and m is the number of replicates. M0 is the rank of the trait that appears most frequently in each germplasm resource across m replicates, and S... i Let m be the standard deviation of the observations for each trait across m replicates for each quality resource. It represents the average of the observations for each trait across m replicates for each quality resource.

[0048] Depend on Figure 1 The results of the germplasm consistency test of the qualitative traits of crabapple flower organs show that the germplasm consistency of the qualitative traits of crabapple flower organs meets the requirements.

[0049] Depend on Figure 1 The results of the intragerminal consistency test of quantitative traits of crabapple flower organs showed that, except for the number of pistils, all crabapple flower organs met the test criteria.

[0050] The third step is to analyze the discriminant properties among germplasms.

[0051] The inter-germ discrimination of qualitative traits of crabapple flower organs is simply measured by the degree of closeness between the mode frequency (MF) of the rank of each trait in all germplasm resources and the theoretical frequency (1 / f) (Formula 3). If MF≤k1 / f, then the qualitative trait has a high degree of discrimination among all germplasm resources.

[0052] The distinguishability of quantitative traits in crabapple flower organs is expressed by the coefficient of variation (Cv) of the mean of each trait across all germplasms (Formula 4). If Cv ≥ 15%, the trait is considered to have high distinguishability among all germplasms. Of course, analysis of variance (Tukey's method) is also required for quantitative traits.

[0053]

[0054] Where n is the number of germplasm resources, M0' is the rank with the highest frequency of each trait in all germplasm resources, k is a coefficient whose magnitude depends on the number of ranks (f) of each trait in all germplasm resources, and S' is the standard deviation of the observed mean of each trait in all germplasm resources. This is the average of the observed mean values ​​for each trait across all germplasm resources.

[0055] Depend on Figure 1 The analysis of the interspecific differentiation of qualitative traits of crabapple flower organs revealed that only 15 traits showed high interspecific differentiation (MF≤k1 / f): petal surface wrinkling, calyx recurving, calyx apex shape, crown shape, relative position of petals, calyx color, receptacle pubescence, receptacle color, pedicel pubescence, pedicel color, relationship between stamen and pistil height, style color, petal shape, outer color during full bloom, and large bud color. The remaining 16 traits showed no significant interspecific differences and low trait differentiation, and were therefore not suitable as a basis for crabapple germplasm classification and were deleted.

[0056] Depend on Figure 1The results of the inter-germ discrimination analysis of quantitative traits of crabapple flower organs showed that all 13 quantitative traits had high trait discrimination (Cv≥15%), and the differences of each quantitative trait among different crabapple germplasms reached a significant level (P<0.05).

[0057] The fourth step is principal component analysis and Pearson correlation analysis.

[0058] To achieve high consistency and strong discriminative power, principal component analysis and correlation analysis were employed to further reduce the dimensionality of the selected traits, thereby reducing the number of traits with redundant information. To eliminate the influence of different dimensions on the data analysis, the original numerical matrix was first standardized using standard deviation (STD), i.e., normalized.

[0059] The results of principal component analysis of crabapple flower organs are shown in Table 2. A total of 10 principal components were extracted, with a cumulative contribution rate of 83.692%. The first principal component contributed 19.66%, and the traits with larger absolute values ​​of eigenvectors were bud color, pedicel color, sepal color, outer color during full bloom, receptacle color, and style color, mainly reflecting the color of crabapple flower organs. The second principal component contributed 18.26%, and the traits with larger absolute values ​​of eigenvectors were flower diameter, petal length, and petal width, mainly reflecting the size of crabapple flowers. The third principal component contributed 11.60%, and the traits with larger absolute values ​​of eigenvectors were petal length-to-width ratio, petal shape, and relative position of petals, mainly reflecting the shape and arrangement of crabapple petals. The fourth principal component contributed 8.93%, and the traits with larger absolute values ​​of eigenvectors were petal number and double-petaledness, mainly reflecting the double-petaledness of crabapple flowers. The fifth principal component contributed 6.64%, with traits showing the largest absolute values ​​in their eigenvectors being the pubescence of the receptacle and pedicel, primarily reflecting the pubescence of the receptacle and pedicel of the crabapple. The sixth principal component contributed 4.81%, with traits showing the largest absolute values ​​in their eigenvectors being the length-to-width ratio of the calyx and the shape of the calyx apex, primarily reflecting the shape of the crabapple calyx. The seventh principal component contributed 3.72%, with the largest absolute value in its eigenvector being the length of the pedicel. The eighth principal component contributed 3.53%, with the largest absolute value in its eigenvector being the wrinkling of the petal surface, primarily reflecting the smoothness of the crabapple petal surface. The ninth principal component contributed 3.45%, with the largest absolute value in its eigenvector being the height relationship between the stamen and pistil, primarily reflecting the relative height of the pistil and stamen groups in the crabapple flower. The tenth principal component contributed 3.11%, with the largest absolute value in its eigenvector being the crown shape, primarily reflecting the degree of curling of the outer periphery of the crabapple crown.

[0060] The results of Pearson correlation analysis are as follows: Figure 2In the figure, '×' indicates a correlation coefficient r > 0.80 between floral organ phenotypic traits. It can be seen that most traits are independent of each other, with only a few traits showing complete or close pairwise correlations, such as flower diameter and petal length (r = 0.98), flower diameter and petal width (r = 0.82), petal length and petal width (r = 0.81), number of petals and double-flowered nature (r = 0.87), outer flower color at full bloom and bud color (r = 0.94), pedicel color and sepal color (r = 0.83), and receptacle pubescence and pedicel pubescence (r = 0.83). For these strongly correlated traits, only one of them needs to be considered when classifying the sample size.

[0061] Table 2. Eigenvalues, contribution rates, and cumulative contribution rates of each principal component.

[0062]

[0063]

[0064] Through the above four steps (steps 1 to 4), this scheme selected a total of 15 important phenotypic traits of crabapple flower organs, namely: outer color of full bloom, pedicel color, calyx color, style color, petal shape, relative position of petals, flower shape, pedicel pubescence, sepal tip shape, flower shape, petal surface wrinkling, height relationship between stamen and pistil, flower diameter, pedicel length, and sepal length-to-width ratio. This scheme successfully achieved a high degree of dimensionality reduction of quantitative classification traits of crabapple germplasm.

[0065] Based on the 15 important floral organ traits extracted in this scheme, using the variable class averaging method (a non-traditional class averaging method), the 142 Malus spectabilis germplasm resources can be divided into 2 major categories (A and B) and 5 subcategories (A1, A2, B1, B2, B3). The major categories / subcategories have significantly different floral organ phenotypic characteristics. Figure 3 This indicates that the classification trait screening technology system proposed in this invention can efficiently extract important phenotypic traits of crabapple organs, thereby achieving scientific classification of all germplasm. The operation of this system meets the requirements of taxonomy for classification traits.

[0066] Figure 4 To analyze the kinship-based clustering distribution characteristics of Malus spectrogram germplasm, we found that Malus species within the same group (lineage) are relatively concentrated, with a kinship-based clustering probability of up to 87.10% in the two major clusters (A and B) and up to 61.29% in the five subgroups (A1, A2, B1, B2, and B3). Homologous varieties also exhibited significant kinship-based clustering characteristics in the two major clusters (A and B) and the five subgroups (A1, A2, B1, B2, and B3) (kinship-based clustering probabilities reaching 72.73% and 63.64%, respectively). These favorable kinship-based distribution characteristics indirectly verify the strong scientific validity of the trait screening technology system proposed in this invention.

[0067] In summary, the beneficial effects of the ornamental plant germplasm quantitative classification trait screening technology system of this invention are as follows: This method, by encoding the measured traits and then performing intra-germplasm consistency testing, inter-germplasm discrimination analysis, principal component analysis, and correlation analysis, can effectively screen important traits of ornamental plants, successfully achieving a high degree of dimensionality reduction of their germplasm classification traits, and ultimately achieving a better scientific classification of all germplasm. The operation of this system conforms to the requirements of taxonomy for classification traits and has strong scientific validity.

[0068] The embodiments described above are some, but not all, embodiments of the present invention. The detailed description of the embodiments of the present invention is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

Claims

1. A method for screening of ornamental plant germplasm for quantitative classification traits, characterized by, It includes: First, the selection and coding of traits. The coding of traits includes qualitative trait coding and quantitative trait coding. Qualitative trait coding adopts the rank-quantitative coding method, which expands the trait by taking consecutive non-negative integers 0, 1, 2, 3, ... in the order from primitive to evolutionary. For binary traits whose evolutionary relationship is not easy to determine, it is coded as 1 if yes and 0 if no. Quantitative traits are not coded and directly enter the next step of calculation using the mean. Second, the intra-germplasm consistency test. Third, the inter-germplasm discrimination analysis. Fourth, principal component analysis and Pearson correlation analysis. The principal component analysis and Pearson correlation analysis are used to further reduce the dimensionality of the selected traits under the premise of satisfying high consistency and strong discrimination. At the same time, the original numerical matrix needs to be standardized by standard deviation processing first.

2. The method for screening of quantitative classification traits of ornamental plant germplasm according to claim 1, characterized in that: The selection of traits in the first step refers to the DUS testing guidelines for new plant varieties of each genus established by the International Union for the Protection of New Varieties of Plants (IUPAC), and additional traits are selected based on their ornamental and identification value.

3. The method for screening of quantitative classification traits of ornamental plant germplasm according to claim 1, characterized in that: The germplasm consistency test includes both qualitative and quantitative traits.

4. The method for screening the quantitative classification traits of ornamental plant germplasm according to claim 3, characterized in that: The qualitative trait germplasm consistency test is expressed by average mode frequency, that is The calculation method is shown in formula 1, wherein n is the number of germplasm resources, m is the number of repetitions, M0 is the highest grade of each trait in each germplasm resource m repetitions, and the following formula is shown, 5. The method for screening the quantitative classification traits of ornamental plant germplasm according to claim 3, characterized in that: The quantitative trait germplasm consistency test adopts average variation coefficient, that is The calculation method is shown in formula 2, wherein n is the number of germplasm resources, m is the number of repetitions, S i is the standard deviation of the observed values of each trait in m repetitions of each germplasm resource, is the average value of the observed values of each trait in m repetitions of each germplasm resource, 6. The method for screening quantitative classification traits of ornamental plant germplasm according to claim 1, characterized in that: The germplasm discrimination analysis is divided into qualitative trait discrimination analysis and quantitative trait discrimination analysis.

7. The method for screening the quantitative classification traits of ornamental plant germplasm according to claim 6, characterized in that: The analysis of the qualitative trait between the germplasm degree of distinction is simply measured by the closeness of the mode frequency MF of the grades of each trait in all germplasm resources to the theoretical frequency 1f, see formula 3, wherein n is the number of germplasm resources, M0 ' is the highest frequency of each trait in all germplasm resources, k is a coefficient, and the determination of the size depends on the number of grades of each trait in all germplasm resources is the highest frequency of each trait in all germplasm resources, k is a coefficient, and the determination of the size depends on the number of grades of each trait in all germplasm resources 8. The method for screening the quantitative classification traits of ornamental plant germplasm according to claim 7, characterized in that: The inter-germplasm discrimination analysis of the quantitative traits is expressed by the coefficient of variation (Cv) of the mean of each trait across all germplasms, as shown in Formula 4, where n is the number of germplasm resources, and S... ' The standard deviation of the observed mean for each trait across all germplasm resources. This represents the average of the observed means for each trait across all germplasm resources. Furthermore, the inter-germplasm discrimination analysis of quantitative traits also requires significance testing.

Images