Whole genome selection method for genetic improvement of multiple characters

By establishing a statistical genetic model under multitrait conditions and calculating the genetic variance and covariance matrix, the problem of difficulty in distinguishing between main effect genes and micro-effect genes in the existing technology is solved, the prediction accuracy is improved, and the needs of multitrait breeding are adapted.

CN119068994AInactive Publication Date: 2024-12-03SICHUAN AGRI UNIV
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
CN202411108550.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-13
Publication Date
2024-12-03
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing prediction models are difficult to distinguish between main effect genes and micro effect genes under multitrait conditions, resulting in a decrease in prediction accuracy.

Method used

Establish a statistical genetic model under multi-trait conditions, and locate the main effect gene loci and construct a mixed model equation group. The genetic variance and covariance matrix are calculated by using the minimum norm quadratic unbiased estimation method, and combined with the best linear unbiased prediction calculation Marking effect values ​​and fixed effect values ​​are then sorted and selected for individual genome breeding values.

Benefits of technology

The accuracy of the distinction between main effect genes and micro-effect genes under multiple traits has been improved, the accuracy of prediction has been improved, and it has been adapted to actual breeding conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure BDA0004991514930000021
    Figure BDA0004991514930000021
  • Figure BDA0004991514930000031
    Figure BDA0004991514930000031
  • Figure BDA0004991514930000034
    Figure BDA0004991514930000034
Patent Text Reader

Abstract

The invention discloses a whole genome selection method for genetic improvement of multiple characters, and belongs to the technical field of biological breeding. The whole genome selection method comprises the following steps: (1) establishing a statistical genetic model; (2) positioning a major gene; (3) constructing a hybrid model equation set to estimate an individual whole genome breeding value; and (4) selection based on an individual whole genome breeding value. By adopting the whole genome selection method disclosed by the invention, a major gene segment for controlling a plurality of characters can be obtained, actual breeding conditions are further met, and when the characters are controlled by major genes and minor genes, compared with the prior art, the whole genome selection method disclosed by the invention has higher prediction accuracy.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of biological breeding, and in particular to a whole genome selection method for genetic improvement of multiple traits. Background Art

[0002] Most of the important traits of plants and animals are quantitative traits controlled by multiple genes with minor effects. Traditional marker-assisted breeding (MAS) and gene editing (CRISPR) technologies are only effective for genes with large effects, but rarely for genes with minor effects. In 2001, Meuwissen et al. pioneered the theory of genome selection (GS), which brought revolutionary progress to the genetic improvement of quantitative traits of plants and animals. This technology is currently the only molecular breeding technology that can genetically improve quantitative traits. It is characterized by short cycle, high efficiency, and wide adaptability. It has become the main breeding method for many large animal and plant breeding companies.

[0003] Different from the traditional MAS technology that uses a few major effect genes for individual breeding, the whole genome selection technology uses molecular markers covering the whole genome to estimate the breeding value of individuals, which can maximize the use of genetic effects and thus has a high prediction accuracy. In order to effectively use the whole genome selection technology, some statistical models such as genomic best linear unbiased prediction (GBLUP), BayesA, random forest (RF) and so on have been proposed. At present, these statistical models have been successfully applied to the breeding of corn, wheat and dairy cows.

[0004] With global climate change and the improvement of human demand, the coordinated improvement of multiple traits in breeding high-yield, high-quality and multi-resistant animal and plant varieties has become the main direction of current breeding. However, the existing prediction models lack the ability to distinguish and estimate major effect genes from minor effect genes under multi-trait conditions, which will inevitably lead to a decrease in prediction accuracy. Summary of the invention

[0005] In order to solve the above-mentioned deficiencies in the prior art, the purpose of the present invention is to provide a whole genome selection method for genetic improvement of multiple traits, so as to solve the problem that the existing prediction model lacks the ability to distinguish and estimate major effect genes from minor effect genes under multiple trait conditions and the prediction accuracy is reduced.

[0006] The technical solution of the present invention to solve the above technical problem is as follows: a whole genome selection method for genetic improvement of multiple traits is provided, comprising the following steps:

[0007] (1) Establishment of statistical genetic model:

[0008] A statistical genetic model based on major effect genes and minor effect genes under multi-trait conditions was established for phenotypic prediction. Assuming that the number of traits is n, the statistical genetic model is as follows:

[0009]

[0010] Among them, y1 and y n are the phenotypic vectors of the first and nth traits, respectively; μ1 and μ n are the means of the first and nth traits respectively; f1 and f n are the marker matrices of the first and nth trait main effect genes, respectively; w1 and w n are the marker effect vectors of the first and nth major effect genes of the traits respectively; Z1 and Z n are the marker matrices of the first and nth trait minor effect genes, respectively; u1 and u n are the marker effect vectors of the first and nth trait minor effect genes, respectively; e1 and e n are the residual vectors of the 1st and nth traits respectively; I is the identity matrix;

[0011] (2) Based on the statistical genetic model of step (1), the main effect gene is first located. For the i-th trait, the calculation formula of its trait phenotype vector is as follows:

[0012] y i =μ i +f i w i +Z i u i +e i

[0013] Among them, y i is the phenotype vector of the i-th trait; μ i is the phenotypic mean of the i-th trait; f i is the main effect gene marker matrix of the i-th trait; w i is the marker effect vector of the main effect gene of the i-th trait, which is a fixed effect; Z i is the marker matrix of the minor effect gene of the ith trait; u i is the marker effect vector of the minor effect gene of the i-th trait; e i is the residual vector of the ith trait;

[0014] Then the major effect gene loci were located using GCTA software;

[0015] (3) Substituting the major effect gene loci obtained in step (2) into the statistical genetic model of step (1), and then constructing a mixed model equation group for estimating the individual whole genome breeding value; wherein the mixed model equation group is as follows:

[0016]

[0017] in, is the marker effect value, is the fixed effect value; X is the fixed effect matrix; Z is the marker matrix of the minor effect gene; T is the matrix transpose; G is the genetic variance and covariance matrix; D is the residual variance matrix; y is the trait phenotype vector;

[0018]

[0019] Among them, y1 and y n are the phenotypic vectors of the first and nth traits, respectively; μ1 and μ n are the means of the first and nth traits respectively; f1 and f n are the marker matrices of the first and nth trait main effect genes, respectively; w1 and w n are the marker effect vectors of the first and nth major effect genes of the traits respectively; Z1 and Z n are the marker matrices of the first and nth trait minor effect genes, respectively; u1 and u n are the marker effect vectors of the first and nth trait minor effect genes, respectively; and are the residual variances of the 1st and nth traits, respectively;

[0020] (4) Selection based on individual whole genome breeding values: After obtaining the individual whole genome breeding values ​​described in step (3), genome selection is performed based on the size of the individual whole genome breeding values.

[0021] Furthermore, the marker effect vectors of the minor effect genes in step (1) and step (2) are all subject to The normal distribution of i is the genetic relationship matrix of the ith trait; is the genetic variance of the ith trait.

[0022] Further, G. i The calculation formula is as follows:

[0023]

[0024] Where m is the number of markers; Z i is the marker matrix of the minor effect gene of the i-th trait; T is the matrix transpose.

[0025] Furthermore, the residual vector in step (1) obeys Normal distribution of; where I is the unit matrix; is the residual variance of the ith trait.

[0026] Furthermore, in step (3), the genetic variance and covariance matrix G and the residual variance matrix D are calculated using the minimum norm quadratic unbiased estimation method, and the calculation formula is as follows:

[0027]

[0028] Q=V -1 -V -1 X(X T V -1 X) + X T V -1

[0029] V=G+D

[0030] Among them, tr is the trace of the matrix; U m To solve the variable matrix; U n is the background variable matrix; y is the phenotype vector; G represents the genetic variance and covariance matrix; D represents the residual matrix; X is the fixed effect matrix; T represents the matrix transpose; represents genetic variance; Q is the intermediate parameter; V represents the total variance matrix.

[0031] Furthermore, the marker effect value in step (3) is calculated using the best linear unbiased prediction and fixed effect values The calculation formula is as follows:

[0032]

[0033] Q=V -1 -V -1 X(X T V -1 X) + X T V -1

[0034] Among them, y is the phenotype vector; X is the fixed effect matrix; V is the total variance matrix; Z is the marker matrix; T is the matrix transpose; is the estimated genetic variance; Q is the intermediate parameter.

[0035] The present invention has the following beneficial effects: the whole genome selection method of the present invention is based on constructing a statistical genetic model under multiple traits to distinguish and estimate major effect genes and minor effect genes. The whole genome selection method of the present invention can not only obtain major effect gene segments that control multiple traits, but also is more in line with actual breeding conditions. When the traits are controlled by major genes and minor effect genes, the whole genome selection method of the present invention has higher prediction accuracy than the prior art. DETAILED DESCRIPTION

[0036] The principles and features of the present invention are described below in conjunction with the examples, which are only used to explain the present invention and are not intended to limit the scope of the present invention. If no specific conditions are specified in the examples, the conditions are carried out according to conventional conditions or the conditions recommended by the manufacturer. If the manufacturer of the reagents or instruments is not specified, they are all conventional products that can be purchased commercially.

[0037] Example 1

[0038] A genome-wide selection method for genetic improvement of multiple traits, comprising the following steps:

[0039] (1) Establishment of statistical genetic model:

[0040] A statistical genetic model based on major effect genes and minor effect genes under multi-trait conditions (Multi-traitmajor-minor Genomic Best Linear Unbiased Prediction, referred to as MTMGBLUP) was established for phenotypic prediction. Here, only the additive effect is concerned. Assuming that the number of traits is n, the statistical genetic model is as follows:

[0041]

[0042] Among them, y1 and y n are the phenotypic vectors of the first and nth traits, respectively; μ1 and μ n are the means of the first and nth traits respectively; f1 and f n are the marker matrices of the first and nth trait main effect genes, respectively; w1 and w n are the marker effect vectors of the first and nth major effect genes of the traits respectively; Z1 and Z n are the marker matrices of the first and nth trait minor effect genes, respectively; u1 and u n are the marker effect vectors of the first and nth trait minor effect genes, respectively, and obey Normal distribution of is the genetic relationship matrix of the ith trait; is the genetic variance of the ith trait; e1 and e n are the residual vectors of the 1st and nth traits, respectively, and obey The normal distribution of is the residual variance of the ith trait; I is the unit matrix;

[0043] (2) Based on the statistical genetic model of step (1), the main effect gene is first located. For the i-th trait, the calculation formula of its trait phenotype vector is as follows:

[0044] y i =μ i +f iw i +Z i u i +e i

[0045] Among them, y i is the phenotype vector of the i-th trait; μ i is the phenotypic mean of the i-th trait; f i is the main effect gene marker matrix of the i-th trait; w i is the marker effect vector of the main effect gene of the i-th trait, which is a fixed effect; Z i is the marker matrix of the minor effect gene of the ith trait; u i is the marker effect vector of the minor effect gene of the ith trait, obeying Normal distribution of is the genetic relationship matrix of the ith trait; is the genetic variance of the ith trait; e i is the residual vector of the ith trait;

[0046] Then, the major gene loci were located using GCTA software (Yang J, Lee SH, Goddard ME, Visscher PM. (2011) GCTA: a tool for genome-wide complex trait analysis. The American Journal of Human Genetics, 88(1): 76–82.).

[0047] (3) Substituting the major effect gene loci obtained in step (2) into the statistical genetic model of step (1), and then constructing a mixed model equation group (MME) for estimating the individual genome-wide breeding value; wherein the mixed model equation group is as follows:

[0048]

[0049] in, is the marker effect value, is the fixed effect value; X is the fixed effect matrix; Z is the marker matrix of the minor effect gene; T is the matrix transpose; G is the genetic variance and covariance matrix; D is the residual variance matrix; y is the trait phenotype vector;

[0050]

[0051] Among them, y1 and y n are the phenotypic vectors of the first and nth traits, respectively; μ1 and μ n are the means of the first and nth traits respectively; f1 and f nare the marker matrices of the first and nth trait main effect genes, respectively; w1 and w n are the marker effect vectors of the first and nth major effect genes of the traits respectively; Z1 and Z n are the marker matrices of the first and nth trait minor effect genes, respectively; u1 and u n are the marker effect vectors of the first and nth trait minor effect genes, respectively; and are the residual variances of the 1st and nth traits, respectively;

[0052] (4) In order to solve the above MME equation, it is necessary to use the minimum normquadratic unbiased estimator (MINQUE) method to calculate the genetic variance and covariance matrix G and the residual variance matrix D. The calculation formula of the minimum normquadratic unbiased estimator method is as follows:

[0053]

[0054] Q=V -1 -V -1 X(X T V -1 X) + X T V -1

[0055] V=G+D

[0056] Among them, tr is the trace of the matrix; U m To solve the variable matrix; U n is the background variable matrix; y is the phenotype vector; G represents the genetic variance and covariance matrix; D represents the residual matrix; X is the fixed effect matrix; T represents the matrix transpose; represents genetic variance; Q is the intermediate parameter; V represents the total variance matrix;

[0057] (5) Calculate the marker effect value using the best linear unbiased prediction (BLUP) and fixed effect values The formulas are as follows:

[0058]

[0059] Q=V -1 -V -1 X(X T V -1 X) + X T V -1

[0060] Among them, y is the phenotype vector; X is the fixed effect matrix; V is the total variance matrix; Z is the marker matrix; T is the matrix transpose; is the estimated genetic variance; Q is the intermediate parameter;

[0061] (6) Selection based on individual whole genome breeding values: After obtaining the individual whole genome breeding values ​​described in step (3), genome selection is performed based on the size of the individual whole genome breeding values.

[0062] The whole genome selection method of the present invention and the prior art GBLUP algorithm were used to calculate the simulated population. The simulated population parameters were as follows: the natural population size was 8700, the number of molecular markers was 13501, the number of major effect genes was 10, the true heritability of trait 1 was 0.13, the true heritability of trait 2 was 0.72, the genetic correlation coefficient of the two traits was 0.68, the test population size of the two traits was 100, and the number of simulations was 10.

[0063] The results are as follows: the prediction accuracy of the whole genome selection method of the present invention is 0.12 for trait 1 and 0.71 for trait 2; the prediction accuracy of the GBLUP algorithm is 0.09 for trait 1 and 0.67 for trait 2. It can be seen that the prediction accuracy of the whole genome selection method of the present invention is better than that of the GBLUP method in the presence of major effect genes.

[0064] Example 2

[0065] A genome-wide selection method for genetic improvement of multiple traits, comprising the following steps:

[0066] (1) Establishment of statistical genetic model:

[0067] A statistical genetic model based on major effect genes and minor effect genes under multi-trait conditions (Multi-traitmajor-minor Genomic Best Linear Unbiased Prediction, referred to as MTMGBLUP) was established for phenotypic prediction. Here, only the additive effect is concerned. Assuming that the number of traits is n, the statistical genetic model is as follows:

[0068]

[0069] Among them, y1 and y n are the phenotypic vectors of the first and nth traits, respectively; μ1 and μ n are the means of the first and nth traits respectively; f1 and f n are the marker matrices of the first and nth trait main effect genes, respectively; w1 and w nare the marker effect vectors of the first and nth major effect genes of the traits respectively; Z1 and Z n are the marker matrices of the first and nth trait minor effect genes, respectively; u1 and u n are the marker effect vectors of the first and nth trait minor effect genes, respectively, and obey Normal distribution of is the genetic relationship matrix of the ith trait; is the genetic variance of the ith trait; e1 and e n are the residual vectors of the 1st and nth traits, respectively, and obey The normal distribution of is the residual variance of the ith trait; I is the unit matrix;

[0070] (2) Based on the statistical genetic model of step (1), the main effect gene is first located. For the i-th trait, the calculation formula of its trait phenotype vector is as follows:

[0071] y i =μ i +f i w i +Z i u i +e i

[0072] Among them, y i is the phenotype vector of the i-th trait; μ i is the phenotypic mean of the i-th trait; f i is the main effect gene marker matrix of the i-th trait; w i is the marker effect vector of the main effect gene of the i-th trait, which is a fixed effect; Z i is the marker matrix of the minor effect gene of the ith trait; u i is the marker effect vector of the minor effect gene of the ith trait, obeying Normal distribution of is the genetic relationship matrix of the ith trait; is the genetic variance of the ith trait; e i is the residual vector of the ith trait;

[0073] Then, the major gene loci were located using GCTA software (Yang J, Lee SH, Goddard ME, Visscher PM. (2011) GCTA: a tool for genome-wide complex trait analysis. The American Journal of Human Genetics, 88(1): 76–82.).

[0074] (3) Substituting the major effect gene loci obtained in step (2) into the statistical genetic model of step (1), and then constructing a mixed model equation group (MME) for estimating the individual genome-wide breeding value; wherein the mixed model equation group is as follows:

[0075]

[0076] in, is the marker effect value, is the fixed effect value; X is the fixed effect matrix; Z is the marker matrix of the minor effect gene; T is the matrix transpose; G is the genetic variance and covariance matrix; D is the residual variance matrix; y is the trait phenotype vector;

[0077]

[0078] Among them, y1 and y n are the phenotypic vectors of the first and nth traits, respectively; μ1 and μ n are the means of the first and nth traits respectively; f1 and f n are the marker matrices of the first and nth trait main effect genes, respectively; w1 and w n are the marker effect vectors of the first and nth major effect genes of the traits respectively; Z1 and Z n are the marker matrices of the first and nth trait minor effect genes, respectively; u1 and u n are the marker effect vectors of the first and nth trait minor effect genes, respectively; and are the residual variances of the 1st and nth traits, respectively;

[0079] (4) In order to solve the above MME equation, it is necessary to use the minimum normquadratic unbiased estimator (MINQUE) method to calculate the genetic variance and covariance matrix G and the residual variance matrix D. The calculation formula of the minimum normquadratic unbiased estimator method is as follows:

[0080]

[0081] Q=V -1 -V -1 X(X T V -1 X) + X T V -1

[0082] V=G+D

[0083] Among them, tr is the trace of the matrix; U mTo solve the variable matrix; U n is the background variable matrix; y is the phenotype vector; G represents the genetic variance and covariance matrix; D represents the residual matrix; X is the fixed effect matrix; T represents the matrix transpose; represents genetic variance; Q is the intermediate parameter; V represents the total variance matrix;

[0084] (5) Calculate the marker effect value using the best linear unbiased prediction (BLUP) and fixed effect values The formulas are as follows:

[0085]

[0086] Q=V -1 -V -1 X(X T V -1 X) + X T V -1

[0087] Among them, y is the phenotype vector; X is the fixed effect matrix; V is the total variance matrix; Z is the marker matrix; T is the matrix transpose; is the estimated genetic variance; Q is the intermediate parameter;

[0088] (6) Selection based on individual whole genome breeding values: After obtaining the individual whole genome breeding values ​​described in step (3), genome selection is performed based on the size of the individual whole genome breeding values.

[0089] The whole genome selection method of the present invention and the prior art GBLUP algorithm were used to calculate the simulated population. The simulated population parameters were as follows: the natural population size was 9500, the number of molecular markers was 14801, the number of major effect genes was 15, the true heritability of trait 1 was 0.11, the true heritability of trait 2 was 0.62, the genetic correlation coefficient of the two traits was 0.35, the test population size of the two traits was 100, and the number of simulations was 10.

[0090] The results are as follows: the prediction accuracy of the whole genome selection method of the present invention is 0.10 for trait 1 and 0.61 for trait 2; the prediction accuracy of the GBLUP algorithm is 0.07 for trait 1 and 0.58 for trait 2. It can be seen that in the case of major effect genes, the prediction accuracy of the whole genome selection method of the present invention is better than that of the GBLUP method.

[0091] The above results show that in the presence of major effect genes, the prediction accuracy of the MTMGBLUP method is higher than that of the GBLUP method.

[0092] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A genome-wide selection method for genetic improvement of multiple traits, It is characterized in that The following steps are involved: (1) Establishment of statistical genetic model: A statistical genetic model based on major effect genes and minor effect genes under multi-trait conditions was established for phenotypic prediction. Assuming that the number of traits is n, the statistical genetic model is as follows: Among them, y 1 and n are the phenotypic vectors of the first and nth traits respectively; μ 1 and μ n are the means of the first and nth traits respectively; f 1 and f n are the marker matrices of the first and nth major effect genes of the traits respectively; w 1 and w n are the marker effect vectors of the first and nth major effect genes of the traits respectively; Z 1 and Z n are the marker matrices of the first and nth trait minor effect genes respectively; u 1 and u n are the marker effect vectors of the first and nth trait minor effect genes, respectively; e 1 and e n are the residual vectors of the 1st and nth traits respectively; I is the identity matrix; (2) Based on the statistical genetic model of step (1), the main effect gene is first located. For the i-th trait, the calculation formula of its trait phenotype vector is as follows: and i =μ i +f i In i +Z i at i +e i Among them, y i is the phenotype vector of the i-th trait; μ i is the phenotypic mean of the i-th trait; f i is the main effect gene marker matrix of the i-th trait; w i is the marker effect vector of the main effect gene of the i-th trait, which is a fixed effect; Z i is the marker matrix of the minor effect gene of the ith trait; u i is the marker effect vector of the minor effect gene of the i-th trait; e i is the residual vector of the ith trait; Then the major effect gene loci were located using GCTA software; (3) Substituting the major effect gene loci obtained in step (2) into the statistical genetic model of step (1), and then constructing a mixed model equation group for estimating the individual whole genome breeding value; wherein the mixed model equation group is as follows: in, is the marker effect value, is the fixed effect value; X is the fixed effect matrix; Z is the marker matrix of the minor effect gene; T is the matrix transpose; G is the genetic variance and covariance matrix; D is the residual variance matrix; y is the trait phenotype vector; Among them, y 1 and n are the phenotypic vectors of the first and nth traits respectively; μ 1 and μ n are the means of the first and nth traits respectively; f 1 and f n are the marker matrices of the first and nth major effect genes of the traits respectively; w 1 and w n are the marker effect vectors of the first and nth major effect genes of the traits respectively; Z 1 and Z n are the marker matrices of the first and nth trait minor effect genes respectively; u 1 and u n are the marker effect vectors of the first and nth trait minor effect genes, respectively; and are the residual variances of the 1st and nth traits, respectively; (4) Selection based on individual whole genome breeding values: After obtaining the individual whole genome breeding values ​​described in step (3), genome selection is performed based on the size of the individual whole genome breeding values.

2. The whole genome selection method for genetic improvement of multiple traits according to claim 1, It is characterized in that The marker effect vectors of the minor effect genes in step (1) and step (2) are all subject to The normal distribution of i is the genetic relationship matrix of the ith trait; is the genetic variance of the ith trait.

3. The whole genome selection method for genetic improvement of multiple traits according to claim 2, It is characterized in that The G i The calculation formula is as follows: Where m is the number of markers; Z i is the marker matrix of the minor effect gene of the i-th trait; T is the matrix transpose.

4. The whole genome selection method for genetic improvement of multiple traits according to claim 1, It is characterized in that The residual vector in step (1) obeys Normal distribution of; where I is the unit matrix; is the residual variance of the ith trait.

5. The whole genome selection method for genetic improvement of multiple traits according to claim 1, It is characterized in that The genetic variance and covariance matrix G and the residual variance matrix D in step (3) are calculated using the minimum norm quadratic unbiased estimation method, and the calculation formula is as follows: Q=V -1 -V -1 X(X T V -1 X) + X T V -1 V=G+D Among them, tr is the trace of the matrix; U m To solve the variable matrix; U n is the background variable matrix; y is the phenotype vector; G represents the genetic variance and covariance matrix; D represents the residual matrix; X is the fixed effect matrix; T represents the matrix transpose; represents genetic variance; Q is the intermediate parameter; V represents the total variance matrix.

6. The whole genome selection method for genetic improvement of multiple traits according to claim 1, It is characterized in that The marker effect value described in step (3) is calculated using the best linear unbiased prediction and fixed effect values The calculation formula is as follows: Q=V -1 -V -1 X(X T V -1 X) + X T V -1 Among them, y is the phenotype vector; X is the fixed effect matrix; V is the total variance matrix; Z is the marker matrix; T is the matrix transpose; is the estimated genetic variance; Q is the intermediate parameter.

Citation Information

Patent Citations

  • Method for performing forest tree multiple-trait pyramiding breeding based on multiple-trait genomic selection

    CN106755441A

  • A method for estimating genomic breeding value integrating dominance effects

    CN109101786A

  • Method for genome selection analysis of livestock and poultry and livestock-and-poultry breeding method

    CN110211635A

  • Genome selection method for complex character genetic improvement

    CN113223606A

  • Whole genome prediction method

    CN117727370A