Method for constructing genome selective breeding technology system of villus sheep

Abstract

The invention provides a method for constructing a genome selective breeding technical system of downy sheep, and relates to the technical field of genome selective breeding of downy sheep, the method comprises the following steps: constructing a reference group, selecting samples from downy sheep with multiple regions and multiple genetic backgrounds, collecting phenotypic data, and carrying out genetic typing; then establishing a genome selection model, processing genotype and phenotype data, and training by using GBLUP and Bayesian models; then, candidate groups are screened, a genome estimation breeding value is calculated, and individuals ranked in the first 20%-30% are selected; a breeding scheme is formulated, breeding is carried out without too large inbreeding coefficient, and progenies are determined and analyzed; and finally, collecting new data every year, and retraining the model every 2-3 years. Regional ecological differences of a reference group, a model algorithm, a convergence threshold value and candidate group phenotype evaluation index weights are limited, and the breeding effect of the villus sheep can be optimized.

Description

Technical Field

[0001] This invention relates to the field of genomic selection breeding technology, and more specifically, to a method for constructing a genomic selection breeding technology system for cashmere sheep. Background Technology

[0002] In the cashmere sheep farming industry, developing breeds with high wool and cashmere yields, superior cashmere quality, and good growth performance has always been the core objective of breeding work. Traditional cashmere sheep breeding methods mainly rely on pedigree information and phenotypic selection. These methods have played an important role in past breeding practices, but with the development of the industry and the increasing demands for breeding efficiency, they have gradually revealed many limitations.

[0003] Pedigree-based breeding involves analyzing the family lineage of individual cashmere sheep to understand the genetic characteristics of their ancestors and relatives, thereby inferring the individual's genetic potential. This method is based on the laws of gene transmission, assuming that individuals with superior ancestors and relatives are more likely to inherit superior genes. Phenotypic selection selects breeding individuals based on the cashmere sheep's physical characteristics and production performance phenotypic data. Common phenotypic indicators include wool length, fineness, cashmere yield, and weight. This method is intuitive and relatively simple to operate, but it also has significant drawbacks. Phenotype is the result of the interaction between genes and environment, with environmental factors having a greater impact on phenotype. For example, under different feeding conditions and climates, the wool yield and quality of cashmere sheep can vary greatly. Therefore, selecting solely based on phenotype may include individuals with good performance due to environmental factors but low genetic potential in the breeding population, while missing individuals with superior genes whose phenotypes are not fully expressed due to environmental suppression. Moreover, phenotypic selection can only be accurately evaluated after individuals have reached a certain stage of growth, making the breeding cycle long and hindering rapid genetic progress.

[0004] Because current technologies cannot accurately distinguish the influence of genes and environment on phenotypes, there are significant errors in the assessment of an individual's genetic value. In pedigree-based breeding, the complex changes in genes during transmission cannot be considered; in phenotypic selection breeding, environmental factors interfere with the phenotype, preventing it from truly reflecting an individual's genetic potential. Current technologies rely primarily on phenotypic data for selection, requiring individuals to reach a certain growth stage before evaluation, resulting in long generation intervals. Moreover, due to the low accuracy of assessments, multiple generations of selection are often needed to improve the economic traits of the population, slowing the breeding process. Current technologies have limited understanding and utilization of genes, failing to accurately identify gene loci related to wool and cashmere yield, quality, and economic traits. Therefore, it is difficult to conduct targeted gene selection and combination during breeding, hindering the rapid and effective improvement of cashmere sheep's economic traits. Furthermore, current breeding methods often lack a systematic consideration of genetic diversity when constructing breeding populations, potentially leading to a narrow gene pool and increased risk of inbreeding. Furthermore, due to the inability to accurately understand the genetic makeup of individuals, it is difficult to selectively introduce genes with different environmental adaptability, resulting in poor adaptability of the population to environmental changes. Therefore, a method for constructing a genomic selection breeding technology system for cashmere sheep is proposed. Summary of the Invention

[0005] The purpose of this invention is to address the problems raised in the existing background technology. To achieve the above-mentioned objective, this invention provides the following technical solution: a method for constructing a genomic selection breeding technology system for cashmere sheep, comprising the following steps: Step 1, reference population construction: Selecting no fewer than 1000 individuals as reference population samples from cashmere sheep populations with at least 5 different breeding areas and more than 3 different genetic backgrounds; collecting phenotypic data of the reference population samples for 3-5 consecutive years, covering wool length, fineness, cashmere yield, weight, body height, and body length indicators, wherein the wool fineness is measured to an accuracy of 0.1 micrometers; collecting blood or tissue samples to extract genomic DNA; and performing genotyping using a chip with a density of no less than 500,000 SNPs. Step 2, Genomic selection model establishment: Perform quality control on genotype data to remove SNP markers with a missing rate greater than 10% and an allele frequency less than 0.05; standardize phenotypic data to make the mean 0 and the standard deviation 1; select GBLUP and Bayesian models, and train them using genotype and phenotypic data of the reference population, with no less than 100 iterations to estimate the effect value of SNP markers. Step 3, Candidate Population Screening and Evaluation: Based on pedigree information and preliminary phenotypic evaluation, select individuals from the candidate population that account for no less than 30% of the total number of candidates as candidate individuals; perform genotyping on the candidate population to obtain genotype data, input the trained model to calculate the genome estimated breeding value (GEBV), rank the individuals according to GEBV, and select the top 20% to 30% of individuals as breeding candidates. Step 4: Breeding program development and implementation: Based on the GEBV, pedigree information and phenotypic characteristics of the breeding candidates, artificial insemination or natural mating is carried out to produce offspring using a mating scheme that avoids inbreeding coefficients greater than 0.0625; phenotypic determination and genotypic analysis are conducted on the offspring for 2-3 years. Step 5: Optimization and updating of the technical system: Collect new phenotypic and genotypic data every year, retrain the model and adjust parameters every 2-3 years; pay attention to new technologies in the industry, and evaluate whether to introduce new analytical methods or technologies every 1-2 years.

[0006] As a preferred technical solution of the present invention, in the construction of the reference population, the differences in the ecological environment of different breeding areas include an altitude difference of not less than 500 meters, an average annual temperature difference of not less than 5°C, and an annual precipitation difference of not less than 200 millimeters.

[0007] As a preferred embodiment of the present invention, in establishing the genomic selection model, a hybrid linear model algorithm is used when training the GBLUP model, and the model convergence threshold is set to 10. -6 .

[0008] As a preferred technical solution of the present invention, in the candidate population screening and evaluation, the weight allocation of the indicators for preliminary phenotypic evaluation is as follows: wool yield accounts for 30%, cashmere fineness accounts for 30%, growth rate and weight gain account for 20%, and other indicators account for 20%.

[0009] As a preferred technical solution of the present invention, in the formulation and implementation of the breeding program, during artificial insemination, the amount of insemination for each ewe is 0.2-0.3 ml, and the sperm motility is not less than 0.6.

[0010] As a preferred technical solution of the present invention, in the construction of the reference population, when collecting phenotypic data, the wool length is measured to the millimeter, the weight is measured to the 0.1 kg, and the body height and body length are measured to the centimeter.

[0011] As a preferred technical solution of the present invention, in the establishment of the genome selection model, a Bayesian model is selected, and one of BayesA, BayesB or BayesCπ is used; when using the particle swarm optimization algorithm (PSO) to optimize and adjust the model hyperparameters, the particle swarm size is set to 50, the maximum number of iterations is 200, and the hyperparameters are set according to the prior distribution of the reference population data for the above optimization and adjustment.

[0012] As a preferred technical solution of the present invention, in the candidate population screening and evaluation, the chip density used for genotyping is not less than 80% of the chip density used for the reference population; if the chip density used for the reference population is 60K SNP, then the chip density used for genotyping of the candidate population is not less than 48K SNP.

[0013] As a preferred technical solution of the present invention, in the formulation and implementation of the breeding program, when performing phenotypic determination on offspring, the measurement frequency of the measurement indicators is as follows: body weight is measured once a month, and wool and cashmere related indicators are measured once a year; when measuring body weight, the measurement accuracy is accurate to 0.1 kg; the wool length is measured accurately to 0.1 cm, and the cashmere fineness is measured accurately to 0.1 micrometers.

[0014] As a preferred technical solution of the present invention, when introducing new analytical methods or technologies in the optimization and updating of the technical system, a small-scale verification experiment of at least one year is required; the number of wool sheep in the small-scale verification experiment is set at 50.

[0015] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention uses high-density SNP chips to genotype cashmere sheep, obtaining a large amount of genomic information, and combines it with years of accurately measured phenotypic data, using advanced genomic selection models for analysis. These models can more accurately estimate the effect value of each SNP marker on important economic traits, thereby achieving precise assessment of the genetic value of individual cashmere sheep. This invention, based on genomic information for breeding, enables a more comprehensive and in-depth understanding of individual genetic characteristics, greatly reducing selection errors. By screening individuals with truly excellent gene combinations as breeding candidates, the accuracy and reliability of breeding are improved, making the offspring more likely to inherit excellent economic traits, such as high wool yield, good cashmere quality, and fast growth rate.

[0016] This invention enables accurate assessment of the genetic potential of cashmere sheep at an early stage by analyzing their genomic information, allowing for the early selection of individuals with high genetic value for breeding. This skips unnecessary generations of selection, shortening the generation interval from the traditional 4-5 years to about 3 years, greatly accelerating the breeding process and significantly reducing the time required to develop superior breeds.

[0017] This invention, through precise genomic selection, enables the rapid transmission and accumulation of superior genes within a population. In practical applications, after three years of selective breeding, wool sheep populations using this invention have shown significant improvements in important economic traits such as wool length, cashmere yield, and wool fineness. This invention can accurately identify gene loci associated with wool and cashmere yield and improve the yield potential of offspring by selecting individuals with these superior genes for reproduction. In experiments, after three years of selective breeding, the average wool yield of wool sheep populations using this invention increased by 10% and the average cashmere yield increased by 15% compared to traditional breeding groups. This not only increases the economic income of farmers but also improves the overall economic benefits of the wool sheep farming industry.

[0018] In addition to increasing yield, this invention also focuses on improving the quality of wool and cashmere. By selecting genes related to wool fineness, strength, and elongation, the quality of the bred offspring has been significantly improved. The reference population includes cashmere sheep from different regions and genetic backgrounds, allowing this invention to fully utilize the genetic diversity within the population. During the breeding process, a reasonable mating scheme combines superior genes from different sources, enriching the population's gene pool. This not only helps avoid the harmful homozygosity problems caused by inbreeding and ensures the genetic health of offspring, but also provides more genetic resources to cope with future environmental changes and market demands.

[0019] This invention introduces individuals with different genetic backgrounds and environmental adaptability, and through genomic selection and breeding, produces a cashmere sheep population with stronger environmental adaptability. For example, some selected cashmere sheep are better able to adapt to cold, arid, and harsh environmental conditions, reducing production losses caused by environmental changes and improving the stability and sustainability of sheep farming. Attached Figure Description

[0020] Figure 1 A flowchart illustrating the steps involved in constructing the cashmere sheep genome selection breeding technology system provided by this invention; Figure 2 A data diagram is constructed for the reference group provided by this invention; Figure 3 A data flowchart is established for the genomic selection model provided by this invention; Figure 4 This is a data flowchart for candidate population screening and evaluation provided by the present invention; Figure 5 A data flowchart for implementing the breeding program provided by this invention is provided. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention.

[0022] Therefore, the following detailed description of the embodiments of the present invention is not intended to limit the scope of the claimed invention, but merely illustrates some embodiments of the invention. 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. It should be noted that, in the absence of conflict, the embodiments and features and technical solutions in the embodiments of the present invention can be combined with each other. It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0023] Example 1: A method for constructing a genomic selection breeding technology system for cashmere sheep, comprising the following steps: Step 1, reference population construction: Select no less than 1,000 individuals from cashmere sheep populations with at least 5 different breeding regions and more than 3 different genetic backgrounds as reference population samples; collect phenotypic data of the reference population samples for 3 to 5 consecutive years, covering wool length, fineness, cashmere yield, weight, body height, and body length indicators, wherein the wool fineness is measured to an accuracy of 0.1 micrometers; collect blood or tissue samples to extract genomic DNA; and perform genotyping using a chip with a density of no less than 500,000 SNPs. Step 2, Genomic selection model establishment: Perform quality control on genotype data to remove SNP markers with a missing rate greater than 10% and an allele frequency less than 0.05; standardize phenotypic data to make the mean 0 and the standard deviation 1; select GBLUP and Bayesian models, and train them using genotype and phenotypic data of the reference population, with no less than 100 iterations to estimate the effect value of SNP markers. Step 3, Candidate Population Screening and Evaluation: Based on pedigree information and preliminary phenotypic evaluation, select individuals from the candidate population that account for no less than 30% of the total number of candidates as candidate individuals; perform genotyping on the candidate population to obtain genotype data, input the trained model to calculate the genome estimated breeding value (GEBV), rank the individuals according to GEBV, and select the top 20% to 30% of individuals as breeding candidates. Step 4: Breeding program development and implementation: Based on the GEBV, pedigree information and phenotypic characteristics of the breeding candidates, artificial insemination or natural mating is carried out to produce offspring using a mating scheme that avoids inbreeding coefficients greater than 0.0625; phenotypic determination and genotypic analysis are conducted on the offspring for 2-3 years. Step 5: Optimization and updating of the technical system: Collect new phenotypic and genotypic data every year, retrain the model and adjust parameters every 2-3 years; pay attention to new technologies in the industry, and evaluate whether to introduce new analytical methods or technologies every 1-2 years.

[0024] In constructing the reference population, the ecological environment differences between different aquaculture areas include an altitude difference of no less than 500 meters, an average annual temperature difference of no less than 5℃, and an annual precipitation difference of no less than 200 millimeters.

[0025] In establishing the genomic selection model, a mixed linear model algorithm was used when training the GBLUP model, and the model convergence threshold was set to 10. -6 .

[0026] In the screening and evaluation of candidate populations, the weight allocation of the indicators for preliminary phenotypic evaluation is as follows: wool yield accounts for 30%, cashmere fineness accounts for 30%, growth rate and weight gain account for 20%, and other indicators account for 20%.

[0027] During the development and implementation of breeding programs, when artificial insemination is performed, the amount of insemination for each ewe is 0.2-0.3 ml, and the sperm motility is not less than 0.6.

[0028] In the construction of the reference population, during the collection of phenotypic data, wool length was measured to the millimeter, weight to the 0.1 kg, and body height and length to the centimeter.

[0029] In establishing the genomic selection model, the Bayesian model was selected, using one of BayesA, BayesB, or BayesCπ. When using the particle swarm optimization algorithm (PSO) to optimize and adjust the model hyperparameters, the particle swarm size was set to 50, the maximum number of iterations was 200, and the hyperparameter settings were optimized and adjusted based on the prior distribution of the reference population data.

[0030] In the screening and evaluation of candidate populations, the chip density used for genotyping shall not be less than 80% of the chip density used for the reference population; if the chip density used for the reference population is 60K SNPs, then the chip density used for genotyping of the candidate population shall not be less than 48K SNPs.

[0031] In the formulation and implementation of breeding programs, when conducting phenotypic testing on offspring, the measurement frequency of the indicators is as follows: body weight is measured once a month, and wool and cashmere related indicators are measured once a year; when measuring body weight, the measurement accuracy is accurate to 0.1 kg; the wool length is measured accurately to 0.1 cm, and the cashmere fineness is measured accurately to 0.1 micrometers.

[0032] When introducing new analytical methods or technologies during the optimization and updating of the technical system, a small-scale verification test of at least one year is required; the number of wool sheep in the small-scale verification test is set at 50.

[0033] The working principle of the cashmere sheep genome selection breeding technology system: The core principle of this cashmere sheep genome selection breeding technology system is to combine the genomic information of cashmere sheep with the phenotypic information of important economic traits using modern genomics technology and statistical methods. Through a series of steps, including constructing a reference population, establishing a genome selection model, screening candidate populations, formulating breeding programs, and continuously optimizing the technology system, the system can accurately assess the genetic value of cashmere sheep and achieve efficient breeding, thereby accelerating the process of cultivating superior breeds.

[0034] The reference population forms the foundation of the entire genomic selection breeding technology system. Selecting representative individuals from wool sheep populations of different regions and genetic backgrounds ensures a broader range of genetic variations. Differences in ecological environments, such as altitude, temperature, and precipitation, lead to the development of different genetic traits in wool sheep during long-term adaptation. Selecting at least 1000 individuals from at least five different breeding regions with more than three different genetic backgrounds guarantees a rich genetic diversity in the reference population, thus more accurately reflecting the genetic structure of the entire wool sheep population.

[0035] Phenotypic data collection from individuals in the reference population for 3–5 years is necessary because some economic traits of cashmere sheep, such as wool and cashmere yield and quality, and growth performance, are affected by growth stage and environmental factors. Continuous measurement over many years allows for a more comprehensive and accurate understanding of each individual's true phenotype, reducing the interference of environmental factors on phenotypic data. Precise measurement of various phenotypic indicators, such as wool fineness to 0.1 micrometers, cashmere yield to grams, wool length to millimeters, and weight to 0.1 kilograms, improves the accuracy of phenotypic data and provides high-quality input information for subsequent genomic selection models.

[0036] Genotyping of individuals in the reference population is performed using microarrays with a density of at least 500,000 SNPs because single nucleotide polymorphisms (SNPs) are the most common form of genetic variation in the genome. High-density SNP microarrays can detect a large number of SNP markers, which are widely distributed throughout the genome and can cover most of the genetic information. By analyzing these SNP markers, we can understand the genomic characteristics of individuals and provide a genotypic data foundation for the subsequent establishment of genomic selection models.

[0037] Before establishing a genomic selection model, quality control of genotype data is essential. Removing SNP markers with a deletion rate greater than 10% and an allele frequency less than 0.05 can eliminate markers that may have measurement errors or have extremely low frequencies in the population and contribute little to genetic assessment, thus improving the quality and reliability of genotype data. Standardizing phenotypic data to have a mean of 0 and a standard deviation of 1 eliminates dimensional differences between different traits, making the data for different traits comparable and facilitating model training and analysis.

[0038] The selection of a suitable genomic selection model, such as the GBLUP genome-optimal linear unbiased prediction model or the Bayesian model, is based on the characteristics of these models and the genetic characteristics of wool sheep. The GBLUP model assumes that all SNP markers have a small effect on the trait and estimates the breeding value of individuals by constructing a genomic relation matrix, offering advantages such as high computational efficiency and good stability. The Bayesian model, on the other hand, can estimate the effect of SNP markers more flexibly based on different prior assumptions, making it suitable for the genetic assessment of some complex traits. The selected model was trained using genotype and phenotypic data from a reference population, and the effect value of each SNP marker was estimated through at least 100 iterations using an iterative algorithm. When training the GBLUP model, a mixed linear model algorithm was used, combining genomic and phenotypic information, and the model was converged through continuous iteration, with a convergence threshold set at 10. -6 This is to ensure that the model can accurately estimate the effect of SNP markers, thereby enabling accurate prediction of individual genetic value.

[0039] Candidate populations are selected from the prospective population based on pedigree information and preliminary phenotypic assessment. Pedigree information provides information on an individual's kinship, helping to understand their genetic background. Preliminary phenotypic assessment allows for initial screening of individuals based on important economic traits such as wool yield, cashmere fineness, and growth rate. By setting weights for different indicators, such as 30% for wool yield, 30% for cashmere fineness, 20% for growth rate and weight gain, and 20% for other indicators, multiple traits can be comprehensively considered to evaluate individuals and identify those with potential breeding value.

[0040] Genotyping was performed on the candidate population to obtain genotype data, which was then input into a trained genomic selection model to calculate the Genomic Estimated Breeding Value (GEBV). GEBV is calculated based on individual genomic information and model-estimated SNP marker effect values, reflecting an individual's genetic potential. Individuals in the candidate population were ranked according to GEBV, and the top 20%–30% were selected as breeding candidates. These individuals possess high genetic value and are considered to have the greatest potential to produce superior offspring.

[0041] Developing mating programs based on the GEBV, pedigree information, and phenotypic characteristics of breeding candidates aims to achieve the combination and transmission of superior genes. Avoiding mating programs with a close-kin coefficient greater than 0.0625 can reduce the risk of harmful gene homozygosity from inbreeding, ensuring the genetic health of offspring. During artificial insemination, the insemination volume per ewe is 0.2-0.3 ml, with sperm motility not lower than 0.6; this insemination volume and sperm motility ensure a high conception rate.

[0042] Phenotypic and genotypic analysis of the offspring for 2-3 years is conducted to evaluate the effectiveness of the breeding program. By comparing the differences in phenotype and genotype between the offspring and the parents, we can understand the transmission and expression of genes during the breeding process and determine whether the selection program is effective. Simultaneously, continuous monitoring can also promptly identify superior individuals among the offspring, providing new material for further breeding.

[0043] New phenotypic and genotypic data are collected annually to continuously expand the reference population and update data. Over time and with ongoing breeding efforts, new data reflects the genetic changes and environmental adaptations of the wool sheep population. The model is retrained and parameters adjusted every 2-3 years because existing models may become inapplicable as data accumulates and the population's genetic structure changes. Retraining the model allows it to better adapt to new data and population characteristics, improving its accuracy and predictive power.

[0044] Pay close attention to new technologies in the industry and assess every 1-2 years whether to introduce new analytical methods or technologies. As genomics and breeding technologies continue to evolve, new analytical methods and technologies may provide more accurate and efficient breeding methods. When introducing new methods or technologies, conduct small-scale validation trials for at least one year to verify their significant improvement in breeding outcomes. This can avoid the risks of blindly introducing new technologies and ensure that they truly improve the breeding efficiency and quality of cashmere sheep.

[0045] In summary, this method for constructing a genomic selection breeding technology system for cashmere sheep, through the coordinated work of each step and the combination of genomic and phenotypic information, has achieved accurate assessment of the genetic value of cashmere sheep and efficient breeding. Furthermore, through continuous optimization and updates, the effectiveness and adaptability of the technology system have been ensured.

[0046] The working process of constructing a genomic selection breeding technology system for cashmere sheep: Reference population construction stage: First, based on the main distribution of cashmere sheep farming areas, at least five farming regions with significant ecological differences are selected, such as high-altitude cold regions, low-altitude warm regions, arid regions, and humid regions. These regions have significant differences in ecological environment, such as an altitude difference of no less than 500 meters, an average annual temperature difference of no less than 5℃, and an annual precipitation difference of no less than 200 mm.

[0047] Individual selection: Within the selected region, carefully select no fewer than 1,000 individuals from a wool sheep population with at least three different genetic backgrounds. These individuals should cover different ages, sexes, and growth stages to ensure that the reference population can fully represent the genetic diversity of the entire wool sheep population.

[0048] Phenotypic Data Collection: A 3-5 year phenotypic data collection plan was developed for each cashmere sheep in the reference population. Measured parameters included wool length, fineness, cashmere yield, body weight, height, and body length. Professional measuring tools and standard methods were used for data collection. For example, a fineness meter was used to accurately measure wool fineness to 0.1 micrometers; a scale was used to measure body weight to 0.1 kilograms; a measuring tape was used to measure wool length to millimeters; specific measuring instruments were used to measure height and length to centimeters; and cashmere yield was measured by weighing to gram. Measurements were conducted at prescribed intervals: body weight was measured monthly, and wool and cashmere-related parameters were measured annually.

[0049] Samples are collected from blood or tissue samples of individuals within the reference population. For blood samples, professional blood collection equipment is used, and strict aseptic techniques are followed to ensure sample quality.

[0050] Genomic DNA is extracted from collected samples using the phenol-chloroform method or other efficient DNA extraction methods. During the extraction process, experimental conditions must be strictly controlled to ensure the purity and integrity of the DNA.

[0051] Genotyping of extracted DNA samples was performed using a chip with a density of at least 500,000 SNPs. The genotyping process was conducted strictly according to the operating procedures provided by the chip manufacturer to ensure the accuracy and reliability of the genotyping results.

[0052] Genomic selection model establishment phase: Genotype data quality control: Quality control is performed on the genotype data obtained from genotyping. Using specialized bioinformatics software, SNP markers with a deletion rate greater than 10% and an allele frequency less than 0.05 are removed. This step eliminates markers that may have measurement errors or have extremely low frequencies in the population, contributing little to genetic assessment, thus improving the quality of the genotype data.

[0053] The collected phenotypic data were standardized. By calculating the mean and standard deviation of each trait, the phenotypic data were converted into standardized data with a mean of 0 and a standard deviation of 1, eliminating dimensional differences between different traits and facilitating subsequent model analysis.

[0054] Based on the genetic characteristics and breeding goals of cashmere sheep, a suitable model was selected from the GBLUP optimal linear unbiased prediction model and various genomic selection models such as the Bayesian model. If the genetic structure of the trait is relatively simple and computational efficiency and stability are important, the GBLUP model can be selected; if the trait is more complex and a more flexible estimation of SNP marker effects is required, the Bayesian model should be selected.

[0055] Taking the GBLUP model as an example, a mixed linear model algorithm is used for training. Preprocessed genotype and phenotypic data are input into the model, and the effect value of each SNP marker is estimated through an iterative algorithm. The number of iterations is no less than 100, and the convergence threshold is set to 10. -6 This ensures that the model can converge accurately and obtain reliable estimates of SNP labeling effects.

[0056] Candidate Population Screening and Evaluation Phase: From the prospective cashmere sheep population, candidate populations are screened based on pedigree information and preliminary phenotypic evaluation. Pedigree information can be obtained by reviewing breeding records and genealogies to understand the kinship and genetic background of individuals. Preliminary phenotypic evaluation is conducted based on important economic traits such as wool yield, cashmere fineness, growth rate, and weight gain. The weighting of each indicator is as follows: wool yield 30%, cashmere fineness 30%, growth rate and weight gain 20%, and other indicators 20%.

[0057] Based on the preliminary assessment results, individuals representing no less than 30% of the total number of candidates were selected as candidates.

[0058] Genotyping of individuals in the candidate population was performed using a chip density no less than 80% of that used in the reference population to ensure that sufficient genomic information could be obtained.

[0059] Genotype data of the candidate population are input into a trained genomic selection model to calculate the genomic estimated breeding value (GEBV) for each individual.

[0060] Breeding program development and implementation phase: A comprehensive analysis is conducted based on the GEBV, pedigree information, and phenotypic characteristics of breeding candidates. The kinship between individuals is considered to avoid mating combinations with a closeness coefficient greater than 0.0625, thereby reducing the risk of harmful gene homozygosity from inbreeding. Detailed mating plans are developed, clearly defining the combination method for each parent pair to ensure the effective combination and transmission of superior genes.

[0061] If artificial insemination is used for breeding, the amount of insemination per ewe should be 0.2-0.3 ml, and the sperm motility should be no less than 0.6. During artificial insemination, strict adherence to operating procedures is essential to ensure a high success rate.

[0062] Phenotypic and genotypic analyses are conducted on the offspring for 2–3 years. Offspring weight, wool and cashmere-related indicators are measured regularly, and genotyping is performed to understand gene transmission and expression, and to evaluate the effectiveness of the breeding program.

[0063] Technical system optimization and updating phase: New phenotypic and genotypic data are collected annually and added to the reference population, continuously expanding the size of the reference population and updating the data. This helps reflect the genetic changes and environmental adaptation of the cashmere sheep population, providing richer data support for model optimization.

[0064] The genomic selection model is retrained every 2–3 years, and its parameters are adjusted based on new data and research findings. For example, as data accumulates and population genetic structure changes, the effect estimates of SNP markers are updated to better adapt the model to new data and population characteristics, thereby improving the model's accuracy and predictive ability.

[0065] Monitor the latest developments in genomics and breeding technologies, and assess every 1-2 years whether to introduce new analytical methods or technologies. Before introducing a new method or technology, conduct at least one year of small-scale validation trials to compare the breeding effects before and after using the new technology and verify its significant improvement in breeding results. Only after successful validation will the new technology be fully applied to the wool sheep genomic selection breeding technology system.

[0066] Test case Experimental Objective To verify the effectiveness and superiority of the proposed method for constructing a genomic selection breeding technology system for cashmere sheep, and to evaluate the practical effects of this technology system in improving the accuracy of cashmere sheep breeding, accelerating the breeding process, and improving important economic traits.

[0067] Experimental Materials and Methods Test materials 1200 wool sheep from five different breeding regions with more than three different genetic backgrounds were randomly selected as experimental subjects. They were divided into two groups: one group (800 sheep) used the genomic selection breeding technology system of this invention for breeding experiments, and the other group (400 sheep) used the traditional breeding method as a control group.

[0068] Test methods Genomic selection breeding technology system experimental group: constructing a reference population according to the method of this invention, performing genotyping, data processing and correction, breeding value prediction, candidate population genetic evaluation and selection operations.

[0069] Traditional breeding method control group: Selection and mating are carried out based on traditional pedigree information and production performance test results.

[0070] Measurement indicators The growth rate, weight, wool production, annual shearing volume, and wool fineness (in micrometers) of two groups of cashmere sheep were recorded.

[0071] Experimental Results and Analysis growth rate The growth rate data shows that the experimental group had significantly higher body weights than the control group at all growth stages. This indicates that the genomic selection breeding technology system of this invention can more effectively screen individuals with superior growth genes, accelerate the growth and development of cashmere sheep, shorten the time to reach market weight, and improve breeding efficiency.

[0072] Wool production In terms of wool production, the experimental group consistently produced significantly more wool than the control group each year, showing a trend of year-on-year growth. This indicates that the technology system helps improve the wool production performance of cashmere sheep and increase the economic income of farmers.

[0073] Wool fineness Wool fineness is an important indicator of wool quality; the lower the value, the finer the wool and the higher its quality. The wool fineness of the experimental group was lower than that of the control group in all years, and over time, the wool fineness of the experimental group became more stable and tended to be better. This indicates that the breeding technology system of this invention can effectively improve wool quality while increasing wool yield.

[0074] Experimental conclusions Analysis of experimental data shows that the cashmere sheep genomic selection breeding technology system proposed in this invention has significant advantages in improving the growth rate, wool yield, and wool quality of cashmere sheep, which are important economic traits. Compared with traditional breeding methods, this technology system can more accurately screen individuals with superior genes, accelerate the breeding process, and improve breeding efficiency and accuracy. Therefore, this technology system has broad application prospects and promotional value in the field of cashmere sheep breeding.

[0075] The above embodiments are only used to illustrate the present invention and are not intended to limit the technical solutions described herein. Although the present invention has been described in detail with reference to the above embodiments, the present invention is not limited to the specific embodiments described above. Therefore, any modifications or substitutions to the present invention, as well as all technical solutions and improvements that do not depart from the spirit and scope of the invention, are covered within the scope of the claims of the present invention.

Claims

1. A method for constructing a genomic selection breeding technology system for cashmere sheep, characterized in that, Includes the following steps: Step 1: Reference population construction: Select no fewer than 1,000 individuals from at least 5 different breeding regions and cashmere sheep populations with more than 3 different genetic backgrounds as reference population samples; collect phenotypic data of the reference population samples for 3 to 5 consecutive years, covering wool length, fineness, cashmere yield, weight, body height, and body length indicators. Wool fineness is measured to an accuracy of 0.1 micrometers. Collect blood or tissue samples to extract genomic DNA and perform genotyping using a chip with a density of no less than 500,000 SNPs; Step 2, Genomic selection model establishment: Perform quality control on genotype data to remove SNP markers with a missing rate greater than 10% and an allele frequency less than 0.05; standardize phenotypic data to make the mean 0 and the standard deviation 1; select GBLUP and Bayesian models, and train them using genotype and phenotypic data of the reference population, with no less than 100 iterations to estimate the effect value of SNP markers. Step 3, Candidate Group Screening and Evaluation: Based on pedigree information and preliminary phenotypic evaluation, select individuals representing no less than 30% of the total number of candidates from the candidate group as candidates; Genotyping is performed on the candidate population to obtain genotype data. The data is then input into the trained model to calculate the genome-estimated breeding value (GEBV). Individuals are ranked according to their GEBV, and the top 20%–30% of individuals are selected as breeding candidates. Step 4: Breeding program development and implementation: Based on the GEBV, pedigree information and phenotypic characteristics of the breeding candidates, artificial insemination or natural mating is carried out to produce offspring using a mating scheme that avoids inbreeding coefficients greater than 0.0625; phenotypic determination and genotypic analysis are conducted on the offspring for 2-3 years. Step 5: Optimize and update the technical system: Collect new phenotypic and genotypic data every year, and retrain the model and adjust the parameters every 2 to 3 years.

2. The method for constructing a genomic selection breeding technology system for cashmere sheep according to claim 1, characterized in that, In the construction of the reference population, the ecological environment differences between different aquaculture areas include an altitude difference of no less than 500 meters, an average annual temperature difference of no less than 5°C, and an annual precipitation difference of no less than 200 millimeters.

3. The method for constructing a genomic selection breeding technology system for cashmere sheep according to claim 1, characterized in that, In establishing the genomic selection model, a mixed linear model algorithm was used when training the GBLUP model, and the model convergence threshold was set to 10. -6 .

4. The method for constructing a genomic selection breeding technology system for cashmere sheep according to claim 1, characterized in that, In the candidate population screening and evaluation, the weight allocation of the preliminary phenotypic evaluation indicators is as follows: wool yield accounts for 30%, cashmere fineness accounts for 30%, growth rate and weight gain account for 20%, and other indicators account for 20%.

5. The method for constructing a genomic selection breeding technology system for cashmere sheep according to claim 1, characterized in that, In the formulation and implementation of the breeding program, during artificial insemination, the amount of insemination for each ewe is 0.2-0.3 ml, and the sperm motility is not less than 0.

6.

6. The method for constructing a genomic selection breeding technology system for cashmere sheep according to claim 1, characterized in that, In the construction of the reference population, during the collection of phenotypic data, wool length was measured to the millimeter, weight to the 0.1 kg, and body height and length to the centimeter.

7. The method for constructing a genomic selection breeding technology system for cashmere sheep according to claim 1, characterized in that, In establishing the genome selection model, the Bayesian model is selected, using one of BayesA, BayesB, or BayesCπ. When using the particle swarm optimization algorithm (PSO) to optimize and adjust the model hyperparameters, the particle swarm size is set to 50, the maximum number of iterations is 200, and the hyperparameter settings are optimized and adjusted based on the prior distribution of the reference population data.

8. The method for constructing a genomic selection breeding technology system for cashmere sheep according to claim 1, characterized in that, In the screening and evaluation of the candidate population, the chip density used for genotyping shall not be less than 80% of the chip density used for the reference population; if the chip density used for the reference population is 60K SNP, then the chip density used for genotyping of the candidate population shall not be less than 48K SNP.

9. The method for constructing a genomic selection breeding technology system for cashmere sheep according to claim 1, characterized in that, In the formulation and implementation of the breeding program, when conducting phenotypic testing on offspring, the measurement frequency of the indicators is as follows: body weight is measured once a month, and wool and cashmere related indicators are measured once a year; when measuring body weight, the measurement accuracy is accurate to 0.1 kg; the wool length is measured accurately to 0.1 cm, and the cashmere fineness is measured accurately to 0.1 micrometers.

10. The method for constructing a genomic selection breeding technology system for cashmere sheep according to claim 1, characterized in that, When new analytical methods or technologies are introduced during the optimization and updating of the technical system, a small-scale verification experiment of at least one year is required; the number of wool sheep in the small-scale verification experiment is set at 50.

Images