A fine wool sheep genomic selection method

Abstract

The application provides a fine wool sheep genome selection method, relates to the field of animal genetic breeding, and comprises the following steps: assembling a sheep genome reference population, determining main phenotypic traits of the sheep genome reference population; designing a liquid phase chip based on resequencing data of a plurality of sheep breeds and performing fine wool sheep genome selection to obtain GWAS prior information; estimating genetic variance and error variance of different traits based on a RRBLUP model to obtain statistical results of trait heritability; performing genome estimated breeding value prediction analysis based on the statistical results of trait heritability; constructing breeding target traits according to breeding targets, constructing a genome comprehensive breeding value estimation model based on the breeding target traits; calculating comprehensive breeding values of sheep individuals based on the genome comprehensive breeding value estimation model, performing comprehensive breeding value ranking, and selecting and mating sheep according to the comprehensive breeding values; and performing important economic trait genome breeding value prediction analysis based on the GWAS prior information.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of animal genetic breeding technology, and in particular to a fine wool sheep genomic selection method. BACKGROUND

[0002] Gao Shan Merino sheep is a new breed developed by using modern breeding techniques for 20 years, taking Australian Merino sheep as the father and Gansu Gao Shan fine wool sheep as the mother (Yue Yaojing et al., 2014). It has gone through three main breeding stages of crossbreeding improvement, horizontal cross fixation and breeding improvement. It lives in the alpine grassland of Qilian Mountain with an altitude of 2400-4070m all year round, and is the only fine wool sheep breed in the world that can adapt to high mountains, cold, drought and harsh ecological environment (Yang Bohui, 2017). Gao Shan Merino sheep has large body size, rich wool quantity, soft and delicate wool texture, good elasticity and warmth, and is widely used in the production of wool products such as wool sweaters and blankets. In terms of meat quality, Gao Shan Merino sheep has tender and delicious meat with excellent quality and multi-functionality, which is often loved by people. Gao Shan Merino sheep has an important position in animal husbandry, not only brings economic benefits to local herdsmen, but also provides high-quality raw materials for wool and meat product industries, and becomes a kind of animal husbandry breed that is loved and valued by people, which has been widely concerned and protected.

[0003] Since Gaomeinu sheep was approved as a new breed in 2015, many researchers have conducted research on the whole genome association analysis and genomic breeding value prediction of important traits such as wool quality, reproductive performance, and meat of Gaomeinu sheep. In terms of wool quality, Li et al. found that the genetic correlation between wool bundle strength and yearling weight, wool fiber diameter, coefficient of variation of wool fiber diameter, and clean wool weight was significant, indicating that wool bundle strength should be one of the breeding goals (Li et al., 2023). Zhu et al. conducted whole genome association analysis on wool quality traits of Gaomeinu sheep combined with high-density (Single nucleotide polymorphism, SNP) SNP microarray typing information, and identified four genes PBX1, TRPC3, SLITRK5, and PVRL1 as candidate genes related to wool quality traits (Zhu et al., 2022). Qiao et al. statistically recorded the phenotypic data of wool traits of Gaomeinu sheep from 2003 to 2018, and estimated the genetic correlation of estimated wool production, clean wool rate, clean wool weight, wool fiber diameter, coefficient of variation of wool fiber diameter, and wool length as 0.3253, 0.4394, 0.2893, 0.4222, 0.3174, and 0.3669, respectively (Qiao et al., 2021). Lei et al. applied selection signal analysis method to analyze the selection signal of 21 fine wool sheep breeds including Gaomeinu sheep, and found that genes FGF1, SOX18, ALX4, and LRP4 were closely related to the development cycle of hair follicles and hair formation (Lei et al., 2022). Zhao et al. conducted whole genome association analysis on the wool quality of one-year-old Gaomeinu sheep based on whole genome resequencing, and detected 16 significant SNP sites, which were annotated to 14 candidate genes, providing theoretical support for marker-assisted selection in Gaomeinu sheep wool breeding (Zhao et al., 2021b). In terms of body weight, Li et al. applied 50k chip density combined with whole genome association analysis results to construct G matrix for genomic selection of body weight traits of Gaomeinu sheep, and found that the genomic prediction accuracy of body weight traits was improved (Li et al., 2023). Xiao et al. screened 20 SNPs related to important traits such as birth weight, bundle strength, clean rate, and fiber diameter of Gaomeinu sheep based on genomic resequencing data, of which 8 were monomorphic sites and 12 were polymorphic sites (Xiao et al., 2023). Lu et al. conducted whole genome association analysis on birth weight, weaning weight, adult weight, and one-year weight of Gaomeinu sheep based on resequencing data, and found that 113 single nucleotide polymorphisms (SNPs) reached the whole genome significance level of 4 body weight traits, and 30 genes were effectively annotated, including AADACL3, VGF, NPC1, and SERPINA12.It can be seen that although many researchers have carried out extensive research on the molecular genetic basis of important economic traits such as body weight and wool of the Merino sheep, they have not yet mined genes with breeding value and established a comprehensive breeding index based on genomic breeding value, and a set of genomic selection techniques for Merino sheep has not yet been established, which limits the application of molecular breeding techniques in improving the breeding of Merino sheep.

[0004] Therefore, it is necessary to design a fine wool sheep genomic selection method to mine genes with breeding value and establish a comprehensive breeding index based on genomic breeding value, so as to fill the gap in the prior art that there is no systematic genomic selection technique for Merino sheep. SUMMARY

[0005] In view of this, the present application provides a fine wool sheep genomic selection method to fill the gap in the prior art that there is no systematic genomic selection technique for Merino sheep.

[0006] To achieve the above-mentioned purpose, the present application provides a fine wool sheep genomic selection method, comprising:

[0007] Assembling a sheep genomic reference population, and determining the main phenotypic traits of the sheep genomic reference population;

[0008] Designing a liquid phase chip based on resequencing data of several sheep breeds;

[0009] Performing fine wool sheep genomic selection based on the liquid phase chip to obtain GWAS prior information;

[0010] Estimating the heritability of genetic variance and error variance for different traits based on the RRBLUP model to obtain trait heritability statistical results;

[0011] Based on the trait heritability statistical results, using GBLUP, BayesA, BayesB, BayesC and BayesLasso model for genomic estimated breeding value prediction analysis;

[0012] Constructing breeding target traits according to breeding goals, and constructing a genomic comprehensive breeding value estimation model based on the breeding target traits;

[0013] Calculating the comprehensive breeding value of the sheep individual based on the genomic comprehensive breeding value estimation model, and ranking the comprehensive breeding value, and selecting and mating the sheep according to the comprehensive breeding value;

[0014] Performing important economic trait genomic breeding value prediction analysis based on the GWAS prior information.

[0015] Further, the main phenotypic trait determination includes mating record, lamb birth identification, weaning identification, growing identification, adult identification;

[0016] The phenotype categories identified in the lamb birth include birth weight, birth type, gestation time, gender, birth rank, birth date, and lamb group;

[0017] The phenotype categories identified in the weaning include weaning weight, weaning wool length, horn type, wool fineness, daily weight gain before weaning, weaning date, and weaning group;

[0018] The phenotype categories identified in the growing include body weight, daily weight gain, wool length, wool fiber diameter, coefficient of variation of wool fiber diameter, clean wool rate, wool clip, clean wool weight, and growing group;

[0019] The phenotype categories identified in the adult include body weight, wool length, wool fiber diameter, coefficient of variation of wool fiber diameter, clean wool rate, wool clip, clean wool weight, and group.

[0020] Further, the process of wool genome selection based on the liquid chip includes:

[0021] Based on the liquid chip, the sample is sequenced, the sample alignment rate is counted, and the genomic selection individuals obtained after detection are filtered for completeness, heterozygosity, minimum allele frequency, and multiple allelic variant sites;

[0022] PCA analysis is performed on the genomic selection individuals to determine whether there is a grouping phenomenon, and whether to perform population structure correction according to the influence of population structure on genomic prediction.

[0023] Further, the process of constructing a genomic comprehensive breeding value estimation model based on the breeding target traits includes:

[0024] The weighting coefficients of each trait are determined according to the marginal benefit and importance of the breeding target traits, and a genomic comprehensive breeding value estimation model is constructed based on the weighting coefficients of each trait.

[0025] Further, in the process of genomic estimated breeding value prediction analysis, the group and age are taken as fixed factors.

[0026] Further, the genomic comprehensive breeding value estimation model is as follows:

[0027]

[0028] wherein, GEBV i is the genomic estimated breeding value of a certain individual of a certain trait of the Merino sheep; is the mean of the genomic estimated breeding value of the trait; σ GEBV is the standard deviation of the genomic estimated breeding value; GPI is the standardized result of the trait estimated breeding value; GPI is the comprehensive breeding selection index model of high mountain merino sheep, 、 and respectively represent the trait comprehensive selection index, the comprehensive selection index mean and the comprehensive selection index standard deviation.

[0029] Further, in the process of important economic trait genome breeding value prediction analysis based on the GWAS prior information, the number of molecular genetic markers set as prior information based on the genetic parameter estimation and prediction accuracy analysis result of WT traits.

[0030] Compared with the prior art, the beneficial effects of the present application are that:

[0031] 1. The present application performs whole genome association analysis on important economic traits of high mountain merino sheep, such as body weight traits and wool traits, i.e. birth weight, weaning weight, one-year weight, net wool weight, net wool rate, wool fiber diameter, wool fiber diameter coefficient of variation, and wool length, screens out significant SNP sites related to target traits, realizes fine positioning of important trait associated genes, and improves the accuracy of genome selection prediction by combining GWAS prior marker information for genome selection analysis, and optimizes the high mountain merino sheep genetic selection breeding model.

[0032] 2. The present application applies the genome selection breeding technology based on the GBLUP theory and the Bayes theory to estimate the genome estimated breeding value of important economic traits of high mountain merino sheep, obtains individual comprehensive selection index by weighting different genome optimal models of each trait, and selects and retains individuals according to the constructed comprehensive breeding index, so that the genetic value and economic value of individuals can be comprehensively evaluated in early selection, and scientific and reasonable breeding targets can be constructed according to the actual production needs of individuals, so as to meet the demand of the market for different trait uses, and improve the breeding efficiency and economic benefits of breeders. BRIEF DESCRIPTION OF DRAWINGS

[0033] Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are for purposes of illustration only and are not intended to limit the present application thereto. Moreover, the use of the same reference numerals in different figures indicates same or similar components. In the drawings:

[0034] Figure 1 It is a fine wool sheep genome selection method flowchart in the embodiments of the present application;

[0035] Figure 2Figure 1 is a Manhattan plot of the wool fiber diameter trait of the high mountain merino sheep in the liquid chip GWAS research results of the embodiments of the present application, and Figure 2 is a Q-Q plot of the wool fiber diameter trait of the high mountain merino sheep.

[0036] Figure 3 Figure 3 is a PCA analysis result of the genomic selection individual in the embodiments of the present application.

[0037] Figure 4 Figure 4 is a comprehensive breeding value ranking result in the embodiments of the present application.

[0038] Figure 5 Figure 5 is an estimation and prediction accuracy analysis result of the WT trait genetic parameters in the embodiments of the present application, wherein Figure 5(a) is a Manhattan plot of the one-year-old weight trait of the high mountain merino sheep, Figure 5(b) is a Q-Q plot of the one-year-old weight trait of the high mountain merino sheep, and Figure 5(c) is a population genetic parameter statistical graph of the one-year-old weight trait. DETAILED DESCRIPTION

[0039] Exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the drawings, it is understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be accurately conveyed to those skilled in the art. It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict. The present application will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

[0040] The present embodiment proposes a fine wool sheep genomic selection method, aiming to measure the phenotypic values of the high mountain merino's wool production, growth, reproduction and other traits, and establish a genomic selection reference population; develop a genomic breeding chip, and establish a high mountain merino genomic genetic evaluation technology system, such as Figure 1 The steps include:

[0041] Assembling a sheep genomic reference population, and measuring the main phenotypic traits of the sheep genomic reference population;

[0042] Designing a liquid chip based on the resequencing data of several sheep breeds;

[0043] Performing fine wool sheep genomic selection based on the liquid chip to obtain GWAS prior information;

[0044] Estimating the heritability of different traits based on the RRBLUP model to estimate the genetic variance and error variance, and obtaining the trait heritability statistical result;

[0045] Based on the statistical results of the genetic correlation of the traits, the GBLUP, BayesA, BayesB, BayesC and BayesLasso models are used for the prediction analysis of the genomic estimated breeding value;

[0046] According to the breeding target traits, the genomic comprehensive breeding value estimation model is constructed based on the breeding target traits;

[0047] Based on the genomic comprehensive breeding value estimation model, the comprehensive breeding value of the sheep individual is calculated, and the comprehensive breeding value ranking is performed, and the sheep is selected and matched according to the comprehensive breeding value;

[0048] Based on the GWAS prior information, the important economic trait genomic breeding value prediction analysis is performed

[0049] As a preferred embodiment, a genomic reference population of alpine merino sheep is constructed in the embodiment, the population scale is 2663, including 397 super fine type, fine type alpine merino sheep mating rams and 2266 core group ewes. The reference population mainly carries out 4 stages and 33 main phenotypic trait measurement records:

[0050] a. Mating record (pedigree).

[0051] b. Lamb birth identification: birth weight, birth type (single / double / multiple), gestation time, gender, birth grade, birth date, group.

[0052] c. Weaning identification: weaning weight, weaning wool length, horn type, wool fineness (count), pre-weaning daily weight gain, weaning date, group.

[0053] d. Growing identification (14 months old): body weight, daily weight gain, wool length, wool fiber diameter, wool fiber diameter coefficient of variation, clean wool rate, shearing weight, net wool weight, group, etc.

[0054] e. Adult identification (≥26 months): body weight, wool length, wool fiber diameter, wool fiber diameter coefficient of variation, clean wool rate, shearing weight, net wool weight, group, etc.

[0055] Among them, the body weight, wool yield, wool length and horn type are subjective identification traits; the clean wool rate and wool fiber diameter are tested by the test center.

[0056] Based on the re-sequencing data of 620 Merino sheep from 12 breeds around the world, a 50k chip containing 45,213 sites (1K functional site chip containing 1,188 functional sites) was designed, the probe design coverage was 87.745%, the total number of probe design was 84,035, the sample site capture efficiency of 45,213 sites was 93.27%-95.35%, and the average detection rate was 94.2%. The GWAS research results based on the 50K SNP liquid chip of fine wool sheep are shown in Figure 2 .

[0057] In this embodiment, 2178 samples were sequenced, and a total of 26,052,146,716 CleanReads were aligned to the reference sequence, with the highest sample alignment rate of 96.9% and the lowest of 87.23%. After detection, a total of 46178 markers were obtained, and after integrity filtering, the number of markers was 44604, after heterozygosity filtering, the number of markers was 43847, after minimum allele frequency filtering, the number of markers was 40549, and after multiple allele variation site filtering, the number of markers was 40539.

[0058] As shown in Figure 3 , PCA analysis of genome selection individuals showed that there was no grouping phenomenon in the fine wool sheep population, and the population structure in this embodiment did not significantly affect the genome prediction, so the population structure was not corrected in the model.

[0059] The genetic variance and error variance of different traits were estimated by using the kin.blup function of R software package (rrBLUP), and the genetic forces of wool fiber diameter, wool fiber diameter variation coefficient, net wool rate, net wool weight, wool length, and birth weight, weaning weight and one-year-old weight were 0.534, 0.193, 0.400, 0.385, 0.293, 0.157, 0.205 and 0.401 respectively. Among them, wool fiber diameter, net wool rate and one-year-old weight are high genetic force, and other traits are medium genetic force. The statistical results are shown in Table 1.

[0060] Table 1

[0061]

[0062] Genome selection prediction accuracy

[0063] Using GBLUP, BayesA, BayesB, BayesC and BayesLasso models, the flock (Flock) and age (DOB) were used as fixed factors for gEBV prediction analysis, and the analysis statistical results are shown in Table 2.

[0064] Table 2

[0065]

[0066] Integrated breeding value estimation

[0067] According to the breeding improvement and strain breeding target of the Gaoshan Merino sheep, the body weight, wool length, net wool weight and wool fiber diameter are determined as the breeding target traits. According to the marginal benefit and importance of the breeding target traits, the weighting coefficients of the respective traits are determined, and the integrated breeding value estimation model of the Gaoshan Merino sheep genome is as follows:

[0068]

[0069] Among them, GEBV i is the genome estimated breeding value of a certain trait of a certain individual of the Gaoshan Merino sheep; is the mean of the genome estimated breeding value of the trait; σ GEBV is the standard deviation of the genome estimated breeding value; is the result of the trait estimated breeding value after standardization; GPI is the integrated breeding selection index model of the Gaoshan Merino sheep, , and respectively represent the trait integrated selection index, the mean of the integrated selection index and the standard deviation of the integrated selection index.

[0070] The integrated breeding value of the individual is calculated according to the integrated breeding value model of the target trait, the integrated breeding value ranking is obtained, and the selection and matching of the sheep individual are carried out according to the integrated breeding value, as shown in Figure 4 .

[0071] Based on the GWAS prior information obtained by the liquid phase chip, the important economic trait genome breeding value prediction analysis is carried out, and the results are shown in Figure 5 . The genetic parameter estimation and prediction accuracy analysis results of the WT trait show that the accuracy of the breeding value of the selected Top 20% snps as prior information is improved by 2.59% to 7.79%.

[0072] Finally, it should be noted that: the above examples are only used to illustrate the technical solutions of the present application, but not to limit it, although the present application has been described in detail with reference to the above examples, those skilled in the art should understand that the specific embodiments of the present application can be modified or replaced by the same, without departing from the spirit and scope of the present application. Any modification or equivalent replacement, which should be covered within the protection scope of the claims of the present application.

Claims

1. A method for genomic selection in fine-wool sheep, characterized in that, include: A sheep genome reference population was established, and the main phenotypic traits of the sheep genome reference population were determined. A liquid-phase chip was designed based on resequencing data from several sheep breeds; Genome selection of fine-wool sheep was performed based on the liquid phase chip to obtain GWAS prior information; The heritability of genetic variance and error variance for different traits was estimated based on the RRBLUP model, and statistical results of trait heritability were obtained. Based on the statistical results of the heritability of the aforementioned traits, the GBLUP, BayesA, BayesB, BayesC, and BayesLasso species models were used to perform genome estimation breeding value prediction analysis. Based on the breeding objectives, breeding target traits are constructed, and a genome-wide breeding value estimation model is constructed based on the breeding target traits; The comprehensive breeding value of individual sheep is calculated based on the genome comprehensive breeding value estimation model, and the comprehensive breeding value is ranked. The selection and mating of sheep are then carried out according to the comprehensive breeding value. Based on the GWAS prior information, a genomic breeding value prediction analysis of important economic traits was performed. The determination of the main phenotypic traits includes mating records, lamb birth identification, weaning identification, growing-out identification, and adult identification; The phenotypic categories for lamb birth identification include birth weight, birth type, gestation time, sex, birth grade, date of birth, and lamb flock. The phenotypic categories for weaning identification include weaning weight, weaning wool length, horn shape, wool fineness, weight gain the day before weaning, weaning date, and weaning group. The phenotypic categories for breeding identification include body weight, daily weight gain, hair length, wool fiber diameter, wool fiber diameter coefficient of variation, clean hair yield, shearing amount, clean hair amount, and breeding group. The phenotypic categories for adult identification include body weight, hair length, wool fiber diameter, wool fiber diameter coefficient of variation, clean hair yield, shearing amount, clean hair amount, and flock type. The process of wool genome selection based on the liquid phase chip includes: Sample sequencing was performed based on the liquid-phase chip, the sample alignment rate was calculated, and the genomes of selected individuals obtained after detection were filtered for integrity, heterozygosity, minimum allele frequency, and multiple allele variation sites. PCA analysis is performed on the selected individuals to determine whether clustering occurs, and the impact of population structure on genome prediction is used to determine whether population structure correction is necessary. Based on the breeding and improvement goals of Alpine Merino sheep, body weight, wool length, net wool weight, and wool fiber diameter are identified as target breeding traits. The process of constructing a genome-wide breeding value estimation model based on these target traits includes: The weighting coefficients of each trait are determined based on the marginal benefits and importance of the breeding target traits, and a genome-wide breeding value estimation model is constructed based on the weighting coefficients of each trait. In the process of predicting breeding values ​​by estimating genomic values, population type and age are used as fixed factors; The genome-wide breeding value estimation model is shown in the following formula: in, It is the estimated breeding value of a certain trait in an individual of the Alpine Merino sheep; Estimate the mean of the breeding value for this trait based on the genome; Estimating the standard deviation of breeding values ​​for the genome; The results are standardized breeding values ​​for trait estimation. Standardized results for estimating breeding values ​​for the body weight trait of Alpine Merino sheep; Standardized results for estimating breeding values ​​of the clean wool content trait in Alpine Merino sheep; Standardized results for estimating breeding values ​​for the wool length trait of Alpine Merino sheep; Standardized results for estimating breeding values ​​of wool fiber diameter for Alpine Merino sheep; A comprehensive breeding selection index model for alpine Merino sheep. , as well as These represent the overall selection index, the mean of the overall selection index, and the standard deviation of the overall selection index, respectively.

2. The method for selecting fine-wool sheep genomes according to claim 1, characterized in that, In the process of predicting the genomic breeding value of important economic traits based on the GWAS prior information, the number of molecular genetic markers used as prior information is set based on the genetic parameter estimation and prediction accuracy analysis results of the WT trait.

Images