Method for improving livestock and poultry character uniformity
By using the residual selection method of the linear mixed model of animal breeding in livestock and poultry breeding, single-trait and double-trait models were constructed, and livestock and poultry with large breeding values were eliminated. This solved the problem of improving the uniformity of broiler weight, and achieved the improvement of the consistency of group traits and the enhancement of economic benefits.
Patent Information
- Application Number
- CN202511659528.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-13
- Publication Date
- 2026-01-09
AI Technical Summary
Existing technologies are not very effective in improving the uniformity of livestock and poultry populations, especially in broiler breeding, where it is difficult to effectively improve weight uniformity through genetic parameters, thus limiting economic benefits.
Genetic evaluation was performed using residuals from a linear mixed model of animal breeding. By constructing single-trait and dual-trait animal models, selection was conducted using residual variance to eliminate livestock and poultry with higher breeding values, thereby improving the uniformity of livestock and poultry traits.
It significantly improved the uniformity of traits in livestock and poultry groups, enhanced the uniformity of broiler weight, increased the number of eggs laid and the fertilization rate of hens, strengthened individual stress resistance, optimized the efficiency of automated segmentation and packaging in the processing, and improved the economic benefits of breeding enterprises.
Smart Images
Figure SMS_13 
Figure SMS_14 
Figure SMS_15
Abstract
Description
Technical Field
[0001] This invention relates to the field of livestock and poultry breeding technology, specifically a method for improving the uniformity of traits in livestock and poultry. Background Technology
[0002] Uniformity, also known as consistency or uniformity, is a concept describing the degree of consistency in characteristics, attributes, or distribution. Uniformity has different interpretations in different fields and with different indicators. In livestock and poultry, uniformity refers to the uniformity of growth and development within a group, mainly categorized as weight uniformity, body shape uniformity, and sexual maturity uniformity. For broilers, uniformity primarily refers to the range of weight variation within the group, typically defined as the percentage of individuals within ±10% of the group's average weight. For laying hens, uniformity refers not only to weight but also to egg quality; improving egg weight uniformity means reducing differences in egg weight, quantified by the standard deviation of egg weight. For piglets, piglet uniformity refers to the degree of weight dispersion among piglets within the same litter. In recent years, with the rapid development of intensive farming, the demand for production performance and product consistency has become increasingly strong. This is mainly because significant differences in body shape and weight within livestock and poultry groups can affect growth rate, survival rate, feed utilization efficiency, and ultimately lead to economic losses.
[0003] Studies have shown that improving the uniformity of livestock and poultry herds also leads to an increase in overall herd weight. For poultry farmers, increased herd weight uniformity improves the number of qualified hatching eggs laid by hens, fertilization rates, and hatchability, and also enhances individual hens' stress resistance, thus improving the profitability of farming enterprises. For processing, higher weight uniformity facilitates automated processing and packaging of carcasses on production lines, increasing the yield of high-quality processed products.
[0004] Currently in China, the main methods for improving the uniformity of livestock and poultry populations are through enhanced feeding management and optimized feed formulation. During the rearing process, since male and female chickens have different nutritional needs and growth rates, separating them reduces competition for feed, thereby improving production performance and uniformity. Adding glucosamine to sow diets can increase litter size and birth weight. Controlling the temperature, humidity, and ventilation in livestock sheds to create a suitable environment also improves population uniformity. In selective breeding, the primary goal is to improve uniformity by culling individuals with excessively high or low extreme values in certain traits.
[0005] Compared to China, international research has made significant progress in using genetic parameters for selection to improve uniformity. However, current methods primarily focus on reducing parameters such as standard deviation and coefficient of variation, with less than ideal results. Therefore, there is an urgent need for new methods to select for uniform broiler weight, thereby accelerating the genetic advancement of broiler uniformity and improving economic efficiency. Based on this, this invention utilizes residuals from breeding models to perform genetic assessments of uniformity, thereby improving the uniformity of livestock and poultry traits and ultimately enhancing the industry's economic benefits. Summary of the Invention
[0006] The purpose of this invention is to provide a method for improving the uniformity of livestock and poultry traits, thereby providing a reference for improving size consistency and increasing weight within livestock and poultry populations. To achieve the above objective, this invention is the first in China to use residuals from a linear mixture model of animal breeding to select for uniformity. In the animal breeding model, after excluding all effects affecting phenotypic values, the residual variance can be defined as the environmental variance; that is, the environmental variance in the model is the residual variance.
[0007] The technical solution adopted in this invention is as follows: One method to improve the evenness of livestock and poultry traits involves using pedigree information and quality-controlled phenotypic information as data. First, a mixed linear model is used to estimate the genetic variance of the livestock and poultry traits. Then, the residual effect of individual i is calculated. e i The result of squaring and logarithmic transformation is ln( e i 2 Using the single-trait animal model as the response variable, a two-trait animal model is constructed based on the single-trait model. The three models are then used to conduct genetic evaluation of the offspring population to obtain the final breeding value, and culling is carried out based on the final breeding value.
[0008] Furthermore, the method includes the following specific steps: S1. Data Preparation and Quality Control Pedigree information and quality-controlled phenotypic information were used as data; S2, Analysis of Variance Components in the Breeding Model The genetic variance of livestock and poultry traits was estimated using the following mixed linear model; Calculate the residual effect of individual i using a mixed linear model e i Additive genetic effects A i Maternal genetic effects M g Maternal environmental effects M e The genetic variance and heritability; S3, Model Conversion The residual effect of individual i ei After squaring and logarithmic transformation, we get the result ln( e i 2 ); ln( e i 2 Using ) as the response variable, construct a single-trait animal model; Using a single-trait animal model, calculate ln(i) for individual i. e i 2 Additive genetic effects A res and residual effect e res ; The phenotypic value of trait y of individual i and ln( e i 2 Each of the two groups of response variables corresponds to an individual ID and forms two sets of response variables. Based on the relevant parameters in the expressions of the mixed linear model and the single-trait animal model, a two-trait animal model is constructed, and the genetic evaluation of the offspring population is carried out using the two-trait animal model. Calculating breeding value ln( using a bitrait animal model) e 2 A) res The genetic variance and heritability; S4. Solving for each genetic parameter in the breeding model The genetic parameters of the three models—the mixed linear model, the single-trait animal model, and the two-trait animal model—are calculated and solved to obtain the final breeding value ln(…). e 2 Thus, elimination is carried out based on the final breeding value.
[0009] Furthermore, the mixed linear model is as follows: y ige =μ+batch +w i +A i +M g +M e +e i Model I; Among them, y ige Let μ represent the phenotypic value of a certain trait of individual i, μ represent the mean of a certain trait in the paternal lineage of individual i, batch represent the generation batch corresponding to individual i, and w represent the phenotypic value of a certain trait of individual i. i A represents the age at which individual i is born from the mother. i M represents the additive inheritance effect of a certain trait in individual i. g M represents random maternal genetic effects. e Represents random parental environmental effects. ei The residual effect representing individual i; Single-trait animal models are: ln(e i 2 ) = μ + batch + w i +A res +e res Model II; Among them, A res ln(representing individual i) e i 2 The additive genetic effect of ), μ represents the paternal lineage mean of individual i, batch represents the generation batch corresponding to individual i, w i e represents the age at which individual i is born from the mother. res The residual effect for individual i in Model II; The bisexual animal model is: Model III; Where y represents y ige vector, ln( e 2 ) represents ln( e i 2 The vector of ) X y The phenotypic value of trait y represents individual i. ln(representing trait y of individual i) e i 2 The correlation matrix of b and b) It is the solution vector for fixed effects. Z y and These are additive effects A mi A resi The correlation matrix, and They are A i A res The solution vector, W g and d g They are M g The correlation matrix, solution vector, W e and d e They are M e The correlation matrix and solution vector, e y and e res They are e i e res The vector.
[0010] Furthermore, based on the culling rate set according to the breeding objectives, livestock and poultry with higher breeding values are culled.
[0011] Furthermore, the specific process for culling livestock and poultry with higher breeding values is to eliminate the corresponding individuals ranked higher according to the culling rate, based on the breeding value from highest to lowest.
[0012] Furthermore, pedigree information includes the IDs and birth dates of livestock and poultry individuals across generations, as well as their parents' IDs.
[0013] Furthermore, the phenotypic information includes the individual ID, batch, gender, and corresponding parent ID of each generation with phenotypic data.
[0014] Furthermore, quality control involves controlling the phenotypic information by removing phenotypic information that is more than three standard deviations above or below the mean of the population trait, thus obtaining the quality-controlled phenotypic information.
[0015] Furthermore, the method uses genetic evaluation software such as DMU, ASReml, or BLUPF90 to solve for various genetic parameters in the three models: mixed linear model, single-trait animal model, and bitrait animal model.
[0016] The beneficial effects of the method for improving the uniformity of livestock and poultry traits according to the present invention are as follows: To improve the uniformity of livestock and poultry traits, this invention is the first in China to use residuals in a linear animal breeding model for selection of uniformity. By conducting in-depth research on the variance components and genetic parameters in the animal breeding model, the breeding method is simplified and the selection accuracy is improved. Compared with using standard deviation and coefficient of variation to improve uniformity, using residual selection significantly improves the uniformity of livestock and poultry traits. Detailed Implementation
[0017] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Many specific details are set forth in the following description to provide a thorough understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the spirit of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.
[0018] Example 1: A method for improving the uniformity of livestock and poultry traits This embodiment is a method for improving the uniformity of traits in livestock and poultry, specifically including the following steps: S1. Data Preparation and Quality Control The data is mainly divided into pedigree information and phenotypic information.
[0019] The pedigree information includes the ID and date of birth of each generation of livestock and poultry individuals, as well as their parents' IDs.
[0020] Phenotypic information includes the individual ID, batch, gender, and corresponding parent ID of individuals with phenotypic data in each generation.
[0021] To minimize the impact of extreme and outlier values on subsequent data analysis, phenotypic information needs to be quality controlled, and phenotypic information that is more than three times the standard deviation above or below the mean of the population trait should be removed.
[0022] S2, Analysis of Variance Components in the Breeding Model The following mixed linear model was used to estimate the genetic variance of livestock and poultry traits. The model was used to analyze both male and female offspring populations. The specific mixed linear model is as follows: y ige =μ+batch +w i +A i +M g +M e + e i Model I; Among them, y ige Let μ represent the phenotypic value of a certain trait of individual i, μ represent the mean of a certain trait in the paternal lineage of individual i, batch represent the generation batch corresponding to individual i, and w represent the phenotypic value of a certain trait of individual i. i The ages of the mother at birth of individual i, batch size, and w are represented by the number of weeks of birth. i Both belong to fixed effects, A i M represents the additive inheritance effect of a certain trait in individual i. g M represents random maternal genetic effects. e Represents random parental environmental effects. e i This represents the residual effect of individual i.
[0023] Calculate the residual effect of individual i using model I e i Additive genetic effects A i Maternal effect M g M e Genetic variance and heritability.
[0024] S3, Model Conversion The purpose of this invention is to utilize residual variance to select for the uniformity of livestock and poultry, specifically as follows: The residual effect of individual i e i After squaring and logarithmic transformation, we get the result ln( e i 2 ); ln( e i2 Using trait as the response variable, we construct a single-trait animal model, with the aim of building a two-trait animal model based on the single-trait model.
[0025] In mixed linear models, the expected value of the residuals is 0. Directly using the residuals as a trait for model transformation fails to reflect their magnitude. Squaring the residuals overcomes this deficiency, better reflecting the degree of variation in the residuals. Furthermore, the expected value of the squared residuals is the variance, thus the squared residuals are directly related to the variance. Logarithmic transformation of the squared residuals makes the residual variance closer to a normal distribution and reduces the impact of extreme values on the model, thereby increasing the accuracy of the residual fit.
[0026] The specific steps for constructing a single-trait animal model are as follows: ln( e i 2 ) = μ + batch + w i +A res +e res Model II; Among them, A res ln(representing individual i) e i 2 The additive genetic effect of ), μ represents the paternal lineage mean of individual i, batch represents the generation batch corresponding to individual i, w i The ages of the mother at birth of individual i, batch size, and w are represented by the number of weeks of birth. i Both are fixed effects, e res For the residual effect of individual i in Model II, that is, ln( e i 2 () is considered as a trait, and obtained by linear fitting, e res It's about ln( e i 2 The residuals in the mixed linear model are similar to those in Model I, e res It is also calculated based on paternal pedigree, and its value has the same meaning as in Model I, because ln( e i 2 Unlike traits such as body weight and egg weight, which can be directly measured, A is calculated through breeding models. Therefore, in the construction of Model II, the A included in Model I is not considered. i M g M e effect.
[0027] Using Model II, calculate ln( ) for individual i. ei 2 Additive genetic effects A res and residual effect e res .
[0028] Compared to single-trait animal models, multi-trait animal models can fully utilize the genetic and environmental correlations between traits to conduct genetic assessments of individuals. Therefore, the phenotypic value of trait y for individual i is ln( e i 2 Each of the variables corresponds sequentially to an individual ID, forming two sets of response variables. Based on the relevant parameters in the expressions of Model I and Model II, a bitrait animal model is constructed. This model is then used to conduct genetic evaluation of the offspring population. The specific bitrait animal model is as follows: Model III; Where y represents y ige vector, ln( e 2 ) represents ln( e i 2 The vector of ) X y The phenotypic value of trait y represents individual i. ln(representing trait y of individual i) e i 2 The correlation matrix of b and b) It is the solution vector for fixed effects. Z y and These are additive effects A i A res The correlation matrix, and They are A i A res The solution vector, W g and d g They are M g The correlation matrix, solution vector, W e and d e They are M e The correlation matrix and solution vector, e y and e res They are e i e res The vector.
[0029] The association matrix is constructed based on the structure of Model II, which determines the parameters and uses the association matrix to describe the relationship between the model and the individual. Then, the corresponding association matrix is constructed based on the pedigree information and model settings.
[0030] The solution vector is obtained by solving Model II using genetic evaluation software (such as DMU, ASReml, and BLUPF90). These software programs can estimate the values of various effects based on the model specifications and data. During the model solving process, the software estimates the values of fixed effects, additive genetic effects, random maternal effects, etc. These estimates constitute the solution vector.
[0031] Calculate the breeding value ln( using Model III) e 2 A) res Genetic variance and heritability.
[0032] S4. Solving for each genetic parameter in the breeding model After constructing the animal breeding models, genetic evaluation software is needed to calculate the various genetic parameters in the three models, including the additive genetic effect A. i A res Genetic variance and heritability, maternal effect M g M e Genetic variance and heritability, breeding value ln( e 2 ), residual e i e res Common genetic evaluation software such as DMU, ASReml, and BLUPF90 are all suitable for solving the various genetic parameters in this embodiment. In actual operation, simply follow the instructions of each software to gradually build the above model to solve for the various genetic parameters, thereby obtaining the final breeding value ln( e 2 ).
[0033] S5. Under the elimination rate set by the breeding target, the breeding value ln( ) obtained from relevant genetic evaluation software. e 2 ) to select and breed livestock and poultry populations. Typically, ln( e 2 The trait y is negatively inherited, therefore, when selecting and breeding, it should be based on ln( e 2 From largest to smallest, eliminate the individuals ranked higher according to the elimination rate. For example, in a population of 1000 individuals, if 30% needs to be eliminated, then ln( e 2 The top 30% of individuals, sorted from largest to smallest, are eliminated.
[0034] Example 2: Application of methods to improve the uniformity of livestock and poultry traits This embodiment illustrates a specific application of a method for improving the uniformity of traits in livestock and poultry. It employs the method described in Embodiment 1, which is applicable to breeds such as broilers, pigs, cattle, and sheep. This embodiment uses broiler breeding as an example for explanation, and the specific steps are as follows: S1. Using pedigree information of 142,920 broilers from the 7th to 14th generations and phenotypic information of 34,670 broiler individuals from the 11th to 14th generations of a purebred broiler population and a full-sib population, an analysis of the uniformity selection method was conducted.
[0035] Phenotypic data were quality controlled by removing phenotypic data that were more than three standard deviations above or below the population trait mean. Maternal parents with at least five offspring were selected from the population, leaving 28,797 broiler chickens.
[0036] The population was divided into female and male offspring groups. In each group, paternal lines with at least 30 offspring were selected, resulting in a new female offspring dataset with 262 paternal lines and 12,086 individuals.
[0037] S2. Using DMU genetic assessment software, the variance components were estimated according to Model I in Example 1. The results are shown in Table 1.
[0038] Table 1. Estimated genetic variance of broiler weight using the residual variance model.
[0039] Note: σ A 2 It is the mean A m Estimation of genetic variance for additive genetic effects; σ Dg 2 It is the estimated variance of maternal genetic effects, σ De 2 It is the estimated variance of the parental environmental effect; σ e 2 It is the average estimated residual variance for each paternity; σ p 2 It is to estimate the phenotypic variance (σ) p 2 =σ A 2 +σ Dg 2 +σ De 2 +σ e 2 h A 2 The mean A m Estimation of heritability of additive effects, h A 2 =σ A 2 / σp 2 h D 2 To estimate the heritability of the maternal effect, h D 2 =(σ Dg 2 +σ De 2 ) / σ p 2 h A 2 and h D 2 Both represent heritability.
[0040] Due to gender differences in actual processing, roosters generally weigh more than hens. Therefore, during processing, offspring are calculated separately according to gender.
[0041] S3. The residuals in Model I... e i After squaring and logarithmic transformation, we get ln( e i 2 ), and used it as a response variable to construct Model II.
[0042] Model II was solved using genetic evaluation software (such as DMU) to estimate the additive genetic effect A. res residual effect e res The values of parameters such as...
[0043] The phenotypic value of trait y of individual i and ln( e i 2 These variables correspond sequentially to individual IDs, forming two sets of response variables, thus constructing Model III; Model III was used to estimate the weight and variance of the weight residuals of broilers. The specific results are shown in Tables 2 and 3.
[0044] Table 2. Genetic variance of broiler body weight residuals estimated using the bitrait model.
[0045] Note: σ 2 Ares It is ln( e 2 The estimated additive genetic variance; σ 2 eres It is ln( e 2 The estimated residual variance; h 2 res It is the heritability of the squared residual after logarithmic transformation, h 2res =σ 2 Ares / (σ 2 Ares +σ 2 eres );σ 2 av It is the estimated genetic variance of the residuals in a heterogeneous residual model, σ. 2 av =2(σ e 2 ) 2 h 2 res h 2 v It refers to the heritability of residuals in the residual model, h. 2 v =σ 2 av / (2σ p 4 +3σ 2 av ).
[0046] Table 3. Estimated genetic variance of broiler weight using the residual variance model.
[0047] Note: The meanings of the parameters in Table 3 are the same as those in Tables 1 and 2.
[0048] Higher heritability indicates a greater influence of genetic factors on the trait, and a higher likelihood of selective breeding. The remaining parameters represent the variance of each effect; the larger the variance, the easier it is for the trait to be improved through selection. As can be seen from Tables 1-3, σ... e 2 The maxima indicates that for ln( e 2 It is feasible to carry out selective breeding.
[0049] S4. Comparison of the selection effects of different selection methods on the uniformity of rooster weight. Different selection methods were used to simulate selection using historical real data, and the selection effect was evaluated by the weight and evenness of the offspring population. Five methods were compared and named Method 1 to Method 5.
[0050] Method 1 served as the control group, without selecting roosters from the 13th generation, and directly calculated the mean weight, standard deviation, coefficient of variation, and uniformity of the male and female populations from the 14th generation.
[0051] Method 2 selects based on weight phenotypic values, using 13th generation breeding roosters as the candidate group, and culling breeding roosters that are too large or too small based on the culling rate.
[0052] Method 3 selects based on the standard deviations of the 13th generation breeding roosters. Using the 13th generation breeding roosters as the candidate group, breeding roosters with excessively large or small standard deviations are eliminated according to the elimination rate.
[0053] Method 4 selects roosters based on the coefficients of variation of each 13th generation rooster. Using 13th generation roosters as the candidate group, roosters with excessively large or small coefficients of variation are eliminated based on the elimination rate.
[0054] Method 5: Based on the bisexual animal model (i.e., Model III), the ln( e 2 The breeding value is used to select roosters, and the culling rate is used to eliminate the roosters with higher breeding values.
[0055] With a culling rate of 5-60%, the effects of different methods are relatively consistent. Taking a culling rate of 25% as an example, the effects of weight uniformity selection on roosters and hens are shown in Tables 4 and 5, respectively.
[0056] Table 4 Performance levels of selected male progeny at a 25% elimination rate
[0057] Table 5 Performance levels of selected female offspring at a 25% elimination rate
[0058] The comparison results of the above five selection methods show that, compared with the other selection methods, the method of the present invention (method five) uses the logarithm of the residual square to select for body weight uniformity. The offspring male and female chickens have the largest body weight uniformity and body weight, and the smallest standard deviation and coefficient of variation. This indicates that using residuals to select for livestock and poultry traits can not only improve the uniformity of traits, but also improve the phenotypic values of some traits in the optimized population.
[0059] This invention utilizes residuals to select for uniformity of livestock and poultry traits, which can significantly improve the consistency of traits in livestock and poultry groups, and is of great significance for improving the economic benefits of the industry.
[0060] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.
Claims
1. A method for improving the uniformity of traits in livestock and poultry, characterized in that, The method uses pedigree information and quality-controlled phenotypic information as data. First, a mixed linear model is used to estimate the genetic variance of livestock traits. Then, the residual effect of individual i is calculated. e i The result of squaring and logarithmic transformation is ln( e i 2 Using the single-trait animal model as the response variable, a two-trait animal model is constructed based on the single-trait model. The three models are then used to conduct genetic evaluation of the offspring population to obtain the final breeding value, and culling is carried out based on the final breeding value.
2. The method for improving the uniformity of livestock and poultry traits according to claim 1, characterized in that, The method includes the following specific steps: S1. Data Preparation and Quality Control Pedigree information and quality-controlled phenotypic information were used as data; S2, Analysis of Variance Components in the Breeding Model The genetic variance of livestock and poultry traits was estimated using the following mixed linear model; Calculate the residual effect of individual i using a mixed linear model e i Additive genetic effects A i Maternal genetic effects M g Maternal environmental effects M e The genetic variance and heritability; S3, Model Conversion The residual effect of individual i e i After squaring and logarithmic transformation, we get the result ln( e i 2 ); ln( e i 2 Using ) as the response variable, construct a single-trait animal model; Using a single-trait animal model, calculate ln(i) for individual i. e i 2 Additive genetic effects A res and residual effect e res ; The phenotypic value of trait y of individual i and ln( e i 2 Each of the two groups of response variables corresponds to an individual ID and forms two sets of response variables. Based on the relevant parameters in the expressions of the mixed linear model and the single-trait animal model, a two-trait animal model is constructed, and the genetic evaluation of the offspring population is carried out using the two-trait animal model. Calculating breeding value ln( using a bitrait animal model) e 2 A) res The genetic variance and heritability; S4. Solving for each genetic parameter in the breeding model The genetic parameters of the three models—the mixed linear model, the single-trait animal model, and the two-trait animal model—are calculated and solved to obtain the final breeding value ln(…). e 2 Thus, elimination is carried out based on the final breeding value.
3. The method for improving the uniformity of livestock and poultry traits according to claim 1 or 2, characterized in that, The mixed linear model is: y ige =μ+batch +w i +A i +M g +M e + e i Model I; Among them, y ige Let μ represent the phenotypic value of a certain trait of individual i, μ represent the mean of a certain trait in the paternal lineage of individual i, batch represent the generation batch corresponding to individual i, and w represent the phenotypic value of a certain trait of individual i. i A represents the age at which individual i is born from the mother. i M represents the additive inheritance effect of a certain trait in individual i. g M represents random maternal genetic effects. e Represents random parental environmental effects. e i The residual effect representing individual i; Single-trait animal models are: ln( e i 2 ) = μ + batch + w i + A res + e res Model II; Among them, A res ln(representing individual i) e i 2 The additive genetic effect of ), μ represents the paternal lineage mean of individual i, batch represents the generation batch corresponding to individual i, w i e represents the age at which individual i is born from the mother. res The residual effect for individual i in Model II; The bisexual animal model is: Model III; Where y represents y ige vector, ln( e 2 ) represents ln( e i 2 The vector of ) X y The phenotypic value of trait y represents individual i. ln(representing trait y of individual i) e i 2 The correlation matrix of b and b) It is the solution vector for fixed effects. Z y and These are additive effects A i A res The correlation matrix, and They are A i A res The solution vector, W g and d g They are M g The correlation matrix, solution vector, W e and d e They are M e The correlation matrix and solution vector, e y and e res They are e i e res The vector.
4. The method for improving the uniformity of livestock and poultry traits according to claim 1 or 2, characterized in that, Based on the culling rate set according to the breeding objectives, livestock and poultry with higher breeding values are culled.
5. The method for improving the uniformity of livestock and poultry traits according to claim 3, characterized in that, The specific process for culling livestock and poultry with higher breeding values is to eliminate the individuals ranked higher according to their breeding values from highest to lowest, based on the culling rate.
6. The method for improving the uniformity of livestock and poultry traits according to claim 1, 2, or 5, characterized in that, The pedigree information includes the ID and date of birth of each generation of livestock and poultry individuals, as well as their parents' IDs.
7. The method for improving the uniformity of livestock and poultry traits according to claim 1, 2, or 5, characterized in that, Phenotypic information includes the individual ID, batch, gender, and corresponding parent ID of individuals with phenotypic data in each generation.
8. The method for improving the uniformity of livestock and poultry traits according to claim 1, 2, or 5, characterized in that, Quality control involves removing phenotypic information that is more than three standard deviations above or below the mean of the population trait, thus obtaining the quality-controlled phenotypic information.
9. The method for improving the uniformity of livestock and poultry traits according to claim 1, 2, or 5, characterized in that, The method uses DMU, ASREMIL, or BLUPF90 genetic evaluation software to solve for various genetic parameters in the three models: mixed linear model, single-trait animal model, and bitrait animal model.