Genome selective breeding method for cashmere goat with cashmere thickness character

Abstract

The invention belongs to the technical field of genetic breeding, and particularly relates to a genome selective breeding method for cashmere thickness traits of cashmere goats, which is carried out by combining an SNP (Single Nucleotide Polymorphism) site data set with the density of 10K and a Bayesian A genome selection model, and comprises the following steps: carrying out genetic typing by adopting the SNP site data set with the density of 10K to construct a reference group; and estimating a genome estimation breeding value of the candidate individual by using a Bayesian A genome selection model so as to realize genome selection breeding of the cashmere thickness character of the cashmere goat. According to the method disclosed by the invention, the influence of a data set based on randomly screened SNP loci with different densities and a genome selection model on the prediction accuracy of the cashmere thickness character genome of the Inner Mongolia cashmere goat is compared, and the heritability, the prediction accuracy and unbiasedness are systematically evaluated; a low SNP site data set with the density of 10K and a Bayesian A genome selection model are found for the first time, and high accuracy of genome selection breeding of the cashmere thickness character of the Inner Mongolia cashmere goat can be ensured while the cost is saved.

Description

Technical Field

[0001] The invention belongs to the technical field of genetic breeding, and in particular relates to a method for genome selection breeding of cashmere goats with a cashmere thickness trait. Background Art

[0002] As the main production area of ​​cashmere goats in my country, the Inner Mongolia Autonomous Region's cashmere goat industry occupies an important position in the domestic and international markets. Cashmere thickness is one of the core traits that affect cashmere quality and yield. Traditional breeding relies on phenotypic and pedigree selection, which has drawbacks such as long cycles and high costs. With the rapid development of genomic technology, whole-genome selection has become an important means to improve the efficiency of cashmere goat breeding. In the existing technology, low-density SNP density chips and SNP density chips below 50K have insufficient coverage for genomic selection breeding, making it difficult to fully capture QTL loci and unable to make accurate predictions. Therefore, genomic selection breeding for cashmere thickness traits currently mostly uses high-density SNP density chips, SNP density chips of 60K and above, but high-density SNP density chips are too expensive.

[0003] Therefore, there is an urgent need to develop a genomic selection breeding method that balances accuracy and cost. Summary of the Invention

[0004] In order to solve the above problems, the present invention provides a method for genomic selection breeding of cashmere goats for the cashmere thickness trait.

[0005] The present invention is achieved through the following technical solutions:

[0006] A method for genomic selection breeding of cashmere goats for the thick cashmere trait is conducted using a 10K SNP locus dataset combined with a Bayesian A genomic selection model, including:

[0007] Construction of reference population: The genotype-cattle thickness trait association database was established using the genomic information of adult Inner Mongolia cashmere goats. Individuals with complete cashmere thickness trait data were selected, and a SNP locus dataset with a density of 10K was used for genotyping to construct the reference population.

[0008] Genomic prediction and selection: Based on the reference population, young Inner Mongolia cashmere goats are used as candidate individuals. The Bayesian A genomic selection model is used to estimate the genomic estimated breeding values ​​of the candidate individuals, and the candidate individuals are sorted from high to low according to the genomic estimated breeding values. According to the actual situation, candidate individuals with excellent cashmere thickness trait performance are selected, thereby realizing genomic selection breeding of cashmere goats for cashmere thickness trait.

[0009] Preferably, the adult Inner Mongolia cashmere goats refer to Inner Mongolia cashmere goats that are over 12 months old.

[0010] Preferably, the young Inner Mongolia cashmere goats refer to Inner Mongolia cashmere goats under 12 months old.

[0011] Preferably, the reference population is dynamically updated: individuals with newly obtained velvet thickness trait data are included in the reference population, and the Bayesian A-genomic selection model is retrained to improve prediction accuracy.

[0012] Preferably, the updating frequency of the dynamically updated reference population is once a year.

[0013] Preferably, the combination of the SNP site dataset with a density of 10K and the Bayesian A genome selection model is obtained by screening in the following manner:

[0014] (1) Acquire phenotypic data of cashmere goats, wherein the phenotypic data include a measured value of cashmere thickness trait.

[0015] (2) Extracting genomic DNA from cashmere goats and performing genotyping using a high-density SNP chip to obtain initial genotype data; the high-density SNP chip is a SNP chip of 60K or more.

[0016] (3) Performing quality control on the initial genotype data, eliminating markers with a missing rate greater than 10%, individuals with missing markers greater than 10%, and markers with an allele frequency less than 5%.

[0017] (4) Randomly screen out SNP site data sets with different densities from the quality-controlled genotype data, including 60K, 50K, 40K, 30K, 20K, 10K, and 5K.

[0018] (5) Based on the screened SNP sites with different densities, genomic selection models were constructed respectively, including GBLUP, ssGBLUP, Bayesian A and Bayesian B models.

[0019] (6) The cross-validation method was used to screen out a SNP site dataset with a density of 10K and combine it with the Bayesian A genome selection model.

[0020] Preferably, the cashmere goats are Inner Mongolia cashmere goats over 12 months old.

[0021] Preferably, in (2), the high-density SNP chip is a 70K chip, and the initial genotype data includes at least 47,790 SNP sites.

[0022] Compared with the prior art, the present invention has the following beneficial effects:

[0023] The present invention provides a method for genomic selection breeding of cashmere goats for cashmere thickness trait. The method is carried out by combining a SNP site data set with a density of 10K with a Bayesian A genome selection model, and comprises the following steps: constructing a reference population: using genome information of adult Inner Mongolia cashmere goats to establish a genotype-cattle thickness trait association database, selecting individuals with complete cashmere thickness trait data, and using the SNP site data set with a density of 10K to perform genotyping to construct the reference population; and genomic prediction and selection: based on the reference population, using young Inner Mongolia cashmere goats as candidate individuals, estimating genomic estimated breeding values ​​of the candidate individuals by using the Bayesian A genome selection model, sorting the candidate individuals from high to low according to the genomic estimated breeding values, and selecting candidate individuals with excellent cashmere thickness trait performance according to actual conditions, thereby realizing genomic selection breeding of cashmere goats for cashmere thickness trait. This study compared the effects of randomly selected SNP locus datasets of varying densities and genomic selection models on the accuracy of genomic predictions for the cashmere thickness trait in Inner Mongolian cashmere goats. By systematically evaluating heritability, prediction accuracy, and unbiasedness, the study found for the first time that a low-density SNP locus dataset of 10,000 combined with a Bayesian A genomic selection model can achieve high accuracy in genomic selection breeding for the cashmere thickness trait in Inner Mongolian cashmere goats while saving costs. This approach provides a solution for low-cost, high-precision breeding. This approach will bring greater economic benefits to the cashmere goat industry and promote its sustainable development.

[0024] In addition, the existing SNP density chip is a solid-phase chip and cannot detect large-scale variations such as copy number variation or insertion and deletion. These variations may significantly affect the cashmere thickness trait, resulting in reduced accuracy. The present invention uses a SNP site data set, which can dynamically update the reference population, incorporate individuals with newly obtained cashmere thickness trait data into the reference population, and retrain the Bayesian A genome selection model to improve prediction accuracy, providing an effective tool for improving cashmere production and quality and accelerating its genetic improvement process, promoting the high-quality development of China's cashmere industry, and bringing higher economic benefits to the cashmere goat industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0026] Figure 1The heritability of the cashmere thickness trait of Inner Mongolian cashmere using the various SNP locus datasets and statistical methods of the present invention is shown. The X-axis represents the different methods used for each SNP locus dataset, and the Y-axis represents the heritability value. With decreasing SNP density, heritability decreases somewhat, with the exception of the 50K case where the Bayesian A genomic selection model was used, where the heritability peaked at 0.4207, but the downward trend was modest. Overall, the Bayesian method outperformed the GBLUP and ssGBLUP methods.

[0027] Figure 2 The accuracy of the cashmere thickness trait of Inner Mongolia cashmere under each SNP locus data set and statistical method of the present invention is shown in FIG. The X-axis represents the different methods under each SNP locus data set, and the Y-axis represents the accuracy value.

[0028] Figure 3 The unbiasedness of the cashmere thickness trait of Inner Mongolia cashmere under the various SNP locus data sets and statistical methods of the present invention. The X-axis represents the different SNP locus data sets, and the Y-axis represents the unbiased value. DETAILED DESCRIPTION

[0029] To facilitate understanding of the present invention, the present invention will be described more fully below, along with preferred embodiments of the present invention. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and comprehensive understanding of the disclosure of the present invention.

[0030] Unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0031] The beneficial effects of the present invention are described below through specific examples.

[0032] Example 1

[0033] 1. Experimental Animals and Phenotype Sources:

[0034] The sheep used in this study were all from the Erlangshan Ranch of Inner Mongolia Beiping Textile Co., Ltd., a national-level conservation farm for Inner Mongolia Cashmere Goats. All of the goats were of the Erlangshan type. They were aged 12 months and older. The wool phenotypic data for 1,518 individuals from 2020 to 2024 were measured using the methods described in Table 1. These individuals included 224 rams and 1,294 ewes. Pedigrees were traceable for two generations. For the phenotypic analysis, 1,518 records were obtained for wool thickness.

[0035] Table 1 Determination method of thickness properties of Inner Mongolia cashmere cashmere

[0036]

[0037] 2. Genomic DNA Extraction

[0038] Ear samples were collected from 1518 Inner Mongolia cashmere goats. DNA was extracted from ear tissue samples using the phenol-chloroform method. DNA concentration, the absorption wavelength ratio of the highest absorption peaks of nucleic acids, proteins, and phenols (260 nm / 280 nm), and the absorption wavelength ratio of the highest absorption peak of carbohydrates (260 nm / 230 nm) were measured using a NanoDrop 2000 spectrophotometer.

[0039] 3. Sources of Genotype Data

[0040] A total of 1,115 samples were genotyped using the 70K chip and merged with the 404 resequencing samples from the research team based on the shared SNP sites. SNP sites containing multiple alleles and duplicate sample individuals were eliminated, ultimately retaining 47,790 sites and 1,518 samples. The 70K chip was used to fill in 63,906 sites for subsequent analysis.

[0041] 4. Randomly screening SNP site datasets with different densities

[0042] Using the dplyr package of R language, we randomly screened SNP locus data sets with different densities of 60K, 50K, 40K, 30K, 20K, 10K, and 5K, and then performed data quality control on each of them. The quality control standards were as follows:

[0043] (1) Delete markers whose deletion rate in individuals is greater than 10%.

[0044] (2) Delete individuals with missing markers greater than 10%.

[0045] (3) Delete markers with allele frequencies less than 5%.

[0046] 5. Building a genomic selection model

[0047] Four genomic selection models, namely GBLUP, ssGBLUP, Bayesian A and Bayesian B, were used to construct genomic prediction models for the velvet thickness trait based on SNP locus datasets of seven densities.

[0048] GBLUP and ssGBLUP are statistical models based on the BLUP method:

[0049] y=Xb+Za+e.

[0050] Where y is the phenotypic value vector of the trait; b is the fixed effect vector, including field, measurement year, the interaction effect of measurement year and field, and age; a is the additive genetic effect vector, which follows a normal distribution: a~N is the additive genetic variance of the trait, A is the additive genetic correlation matrix between individuals; X and Z are the correlation matrices of fixed effects and additive genetic effects, respectively; e is the random residual effect vector, which obeys the normal distribution: e~N I is the identity matrix, is the residual variance. A is the kinship matrix constructed based on the pedigree. The GBLUP model replaces the A matrix with the G matrix, a matrix constructed based on whole-genome markers. In the ssGBLUP model, the H matrix, constructed using the A and G matrices, replaces the A matrix.

[0051] The Bayesian model is based on the Bayesian statistical framework in computer science:

[0052]

[0053] Where y is the phenotypic value vector of the trait; b is the fixed effect vector; g i is the effect value of the i-th site; X is the correlation matrix of fixed effects; Z i is the genotype at the i-th locus, 0 / 1 / 2; e is the random residual effect vector. The Bayesian A-genomic selection model assumes that every SNP has an effect and that the effect variance follows the same prior distribution. The Bayesian B-genomic selection model assumes that only a small number of SNPs have an effect, with a proportion of 1-π. The parameter π represents the prior probability that the SNP has no effect, meaning that 1-π represents the proportion of SNPs with an effect. In the BGLR package, the default value of π for the Bayesian B-genomic selection model is typically 0.95, assuming that 5% of SNPs have an effect.

[0054] The variance components and heritability of each trait were estimated using ASReml software. The heritability results are as follows: Figure 1 As shown in the figure, the X-axis represents the different methods for each SNP locus dataset, and the Y-axis represents the heritability value. As the SNP density decreases, the heritability decreases. Except for the 50K case, the heritability reaches a maximum of 0.4207 using the Bayesian A genomic selection model, but the downward trend is not significant. The Bayesian method is generally superior to the GBLUP and ssGBLUP methods.

[0055] 6. Model Evaluation

[0056] The five-fold cross-validation method was used to evaluate the accuracy and unbiasedness of the prediction model under different densities of SNP site datasets and genomic selection models. The results are as follows Figure 1 、 Figure 2 and as shown in Table 2.

[0057] Table 2 Heritability, accuracy and unbiasedness results of different statistical methods at different densities

[0058]

[0059] 7. Solution Screening

[0060] The prediction accuracy of different schemes was compared, and combinations of SNP site datasets with different densities and genomic selection models were screened. The results show that under SNP site datasets with different densities, the accuracy of the Bayesian method is higher than that of the BLUP method, and the difference between the Bayesian A genomic selection model and the Bayesian B genomic selection model is relatively small. As the SNP density decreases, the accuracy increases, and the accuracy begins to decrease at 5K. The accuracy and unbiasedness of the Bayesian A genomic selection model for the velvet thickness trait are the best at 10K. Compared with the accuracy of 0.3728 obtained by the GBLUP model at 60K density, the accuracy of the Bayesian A genomic selection model at 10K SNP marker density is 0.4661, and the prediction accuracy is improved by 25.03%.

[0061] Example 2: A method for genomic selection breeding of cashmere goats for the thick cashmere trait

[0062] At a 10K SNP marker density, the Bayesian A genomic selection model has high accuracy and unbiased prediction of the genomic thickness trait of Inner Mongolia cashmere goats. Based on this conclusion, the following genomic selection breeding methods can be adopted:

[0063] 1. Construction of a reference population: The genotype-cattle thickness trait association database was established using the genomic information of adult Inner Mongolia cashmere goats. Individuals with complete cashmere thickness trait data were selected, and a SNP locus dataset with a density of 10K was used for genotyping to construct a reference population. The adult Inner Mongolia cashmere goats referred to were Inner Mongolia cashmere goats over 12 months old.

[0064] 2. Genomic Prediction and Selection: Based on the reference population, young Inner Mongolian cashmere goats are selected as candidate individuals. The Bayesian A genomic selection model is used to estimate the genomic breeding values ​​of the candidate individuals. The candidate individuals are ranked from high to low according to the genomic breeding value, and the candidate individuals with excellent cashmere thickness trait are selected according to the breeding goal, thereby achieving genomic selection breeding for the cashmere thickness trait in cashmere goats. The young Inner Mongolian cashmere goats described herein refer to Inner Mongolian cashmere goats under 12 months of age.

[0065] The reference population is dynamically updated, the newly obtained individuals with velvet thickness phenotypic data are included in the reference population, and the Bayesian A genome selection model is retrained to improve the prediction accuracy, with an update frequency of once a year.

[0066] Preselection at a young age based on genomic information shortens the generation interval and reduces the feeding and time costs required for traditional phenotyping. The present invention combines a 10K-density SNP locus dataset with a Bayesian A-genomic selection model for genomic selection breeding, which not only improves the genetic progress of the down thickness trait but also optimizes the allocation of breeding resources, achieving efficient and low-cost genomic selection breeding.

[0067] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0068] The above-described embodiments merely represent several implementation methods of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the patent. A person skilled in the art would be able to make various modifications and improvements without departing from the spirit of the present invention, and these modifications and improvements fall within the scope of protection of the present invention. Therefore, the scope of protection of the patent for this invention shall be based on the appended claims.

Claims

1. A method for genomic selection breeding of cashmere goats for the thick cashmere trait, characterized in that: The results were performed using a 10K-density SNP locus dataset combined with a Bayesian A-genomic selection model, including: Construction of a reference population: Genotype-to-cash thickness trait association database was established using the genomic information of adult Inner Mongolia cashmere goats. Individuals with complete cashmere thickness trait data were selected and genotyped using a 10K SNP locus dataset to construct a reference population. Genomic prediction and selection: Based on the reference population, young Inner Mongolia cashmere goats are used as candidate individuals, and the Bayesian A genomic selection model is used to estimate the genomic estimated breeding values ​​of the candidate individuals. The candidate individuals are sorted from high to low according to the genomic estimated breeding values, and the candidate individuals with excellent cashmere thickness trait performance are selected, thereby realizing genomic selection breeding of cashmere goats for cashmere thickness trait.

2. The method according to claim 1, characterized in that The adult Inner Mongolia cashmere goats refer to Inner Mongolia cashmere goats that are over 12 months old.

3. The method according to claim 1, characterized in that The young Inner Mongolia cashmere goats refer to Inner Mongolia cashmere goats under 12 months old.

4. The method according to claim 1, wherein Dynamically update the reference population: individuals with newly obtained velvet thickness trait data are included in the reference population, and the Bayesian A-genomic selection model is retrained to improve prediction accuracy.

5. The method according to claim 4, characterized in that The updating frequency of the dynamically updated reference group is once a year.

6. The method according to claim 1, wherein The combination of the 10K SNP locus dataset and the Bayesian A-genome selection model was obtained by screening in the following way: (1) obtaining phenotypic data of cashmere goats, wherein the phenotypic data includes a measured value of a cashmere thickness trait; (2) extracting genomic DNA from cashmere goats and performing genotyping using a high-density SNP chip to obtain initial genotype data; the high-density SNP chip is a SNP chip with 60K or more SNPs; (3) performing quality control on the initial genotype data, eliminating markers with a missing rate greater than 10%, individuals with missing markers greater than 10%, and markers with an allele frequency less than 5%; (4) Randomly screen out SNP site data sets with different densities from the quality-controlled genotype data, including 60K, 50K, 40K, 30K, 20K, 10K, and 5K; (5) Based on the screened SNP sites with different densities, genomic selection models were constructed respectively, including GBLUP, ssGBLUP, Bayesian A and Bayesian B models; (6) The cross-validation method was used to screen out a SNP site dataset with a density of 10K and combine it with the Bayesian A genome selection model.

7. The method according to claim 6, characterized in that The cashmere goats are Inner Mongolia cashmere goats that are over 12 months old.

8. The method according to claim 6, characterized in that In (2), the high-density SNP chip is a 70K chip, and the initial genotype data includes at least 47,790 SNP sites.

Images