Grass carp intensive high-yield character genome selective breeding method

Abstract

The invention belongs to the technical field of fish genetic breeding in aquaculture industry, and discloses a grass carp intensive high-yield character genome selective breeding method, which comprises the following steps: establishing a grass carp intensive high-yield character prediction model; the establishment of the grass carp intensive high-yield character prediction model comprises the following steps: (1) obtaining a grass carp candidate parent fish single nucleotide polymorphism site; (2) constructing an analysis model for predicting breeding values of candidate parents; and (3) constructing a molecular genetic relationship matrix by using the obtained single nucleotide polymorphic sites, and predicting a genome estimation breeding value of the candidate group in combination with the analysis model in the step (2). The improved variety breeding method based on grass carp genome information can be used for screening excellent parents suitable for intensive breeding, the screened parents are directly used for breeding offspring seeds, and the high-density breeding tolerance of grass carp can be directly improved; and a technical support is provided for improving cultivation of new grass carp varieties suitable for intensive and facility cultivation.

Description

Technical Field

[0001] The invention belongs to the technical field of fish genetic breeding in aquaculture, and particularly relates to a method for intensive high-yield trait genome selection breeding of grass carp. Background Art

[0002] Intensive aquaculture is not only an important way to further increase aquaculture production but also a key direction for the future development of the aquaculture industry, crucial for ensuring a stable production and supply of aquatic products. However, research on intensive aquaculture has been slow to gain traction, resulting in a lack of both suitable fish species and efficient breeding technologies for cultivating these superior species. This has severely hampered the development of efficient fish farming models, such as factory-based and facility-based fish farming. Therefore, there is an urgent need to develop efficient breeding technologies for fish suitable for intensive aquaculture.

[0003] grass carp( Ctenopharyngodon idella Grass carp, also known as Anhui fish, is the most productive aquaculture fish and an ideal candidate for intensive aquaculture. However, due to its long sexual maturation cycle, the selection and breeding of high-quality grass carp is a recognized scientific challenge. Currently, only one new grass carp variety, "Husu No. 1," has been developed using family selection combined with molecular marker-assisted selection. A dedicated new grass carp variety suitable for intensive aquaculture has yet to be developed. Therefore, establishing efficient grass carp breeding technologies and cultivating new grass carp varieties suitable for intensive aquaculture are urgent needs for the development of intensive and facility-based aquaculture. Summary of the Invention

[0004] In view of the shortcomings of the prior art, the object of the present invention is to provide a method for intensive high-yield trait genomic selection breeding of grass carp.

[0005] The genomic selection breeding method for intensive high-yield traits of grass carp described in the present invention comprises establishing grass carp families, measuring traits, constructing reference populations, establishing a prediction model for intensive high-yield traits of grass carp, screening and cultivating superior families, and measuring traits. The establishment of the prediction model for intensive high-yield traits of grass carp comprises the following steps: (1) Obtaining single nucleotide polymorphism sites of candidate grass carp broodstock; (2) Construct an analytical model to predict the breeding values ​​of candidate parents; (3) Using the obtained single nucleotide polymorphism sites to construct a molecular kinship matrix, combined with the analysis model in step (2) to predict the genomic estimated breeding value of the candidate population; The present invention has established an efficient grass carp breeding technology model of "precise phenotypic measurement - utilization of genetic variation - screening of excellent parents". By utilizing the genetic variation information within the entire genome of grass carp and combining it with the accurately measured intensive and high-yield phenotype of grass carp, while retaining the genetic diversity of the population, high-yield grass carp that is suitable for intensive breeding is cultivated, providing technical support for the transformation, upgrading and high-quality development of the grass carp farming industry.

[0006] The method for constructing the kinship matrix is ​​as follows: S1. Construct and load the recoded single nucleotide polymorphism genotype matrix M and individual identifier iid to generate the pedigree-based identity matrix A; S2. Calculate the minor allele frequency pi of each SNP site, and construct the minor allele matrix P and the normalization matrix Z=MP in turn; S3. Parallel calculation of the product ZZt of the row vector of the normalization matrix Z and the transposed matrix, and calculation of the normalization factor sum2pq = 2 × (sum(pi × (1-pi))), and the initial genome kinship matrix G = ZZt / sum2pq is obtained by the normalization factor sum2pq; S4. Matrix G is fused with matrix A using weight w, as shown in the formula G = wG + (1-w) A, to obtain a weighted genomic kinship matrix with weight w being 0.95. S5. Calculate the inverse matrix of G and format and store it for use in predicting breeding values ​​for intensive high-yield traits of grass carp.

[0007] The above steps can generate an accurate and reliable kinship matrix that contains both genomic and pedigree information, providing key input for the subsequent prediction of genomic estimated breeding values ​​(GEBVs) for intensive and high-yield traits of grass carp, and ultimately improving the accuracy of screening for excellent families.

[0008] The screening of the single nucleotide polymorphism sites in step (1) meets the following conditions: the minimum allele frequency ≥ 0.05, the genotype missing rate ≤ 5%, the individual genotype missing rate ≤ 10% and conformity to the Hardy-Weinberg equilibrium ( p =0.0001).

[0009] This parameter combination significantly improves the reliability and data quality of SNP markers by strictly filtering low-quality variants, providing a key guarantee for the accuracy of grass carp genome selection breeding, and can realize the "precise phenotypic measurement-genetic variation utilization-excellent parent screening" technical model.

[0010] Preferably, the analysis model in step (2) satisfies the following conditions: phenotype = fixed effect + random effect + residual.

[0011] Preferably, the fixed effect is the group mean, or a combination of the group mean and age or body weight at marking.

[0012] Preferably, the individual random effect is selected from either the mother or the individual.

[0013] The mixed linear model constructed in the present invention can accurately separate various effects affecting the intensive and high-yield traits of grass carp by clarifying the phenotypic composition and optimizing fixed effects (mean, age, weight at marking, etc.) and individual random effects (mother, individual), thereby improving the accuracy of genomic estimated breeding value (GEBV) prediction and providing reliable model support for the screening of excellent families.

[0014] Preferably, the screening and cultivation of superior families and the determination of traits include the following steps: A1. Select individuals with high genomic estimated breeding values ​​as the core population; A2. Conduct high-density stocking stress trials to evaluate the effectiveness of selection by comparing the growth performance of grass carp offspring not selected using genomic estimated breeding values ​​or bred from parents with low genomic estimated breeding values ​​with that of offspring from parents with high genomic estimated breeding values.

[0015] Preferably, the grass carp family establishment comprises the following steps: (B1) Collect grass carp parents from different geographical locations and construct full-sib families through one-to-one artificial insemination and hatching in buckets; (B2) Family seedlings are grown in cages in stages until they are 11-12 days old and vaccinated. When they reach a full length of 11-13 cm, they are electronically marked and their initial growth data is measured. Optional lengths include 11 cm, 11.5 cm, 12 cm, 12.5 cm, and 13 cm. Specific values ​​within these ranges can be selected.

[0016] The method of the present invention obtains a bred grass carp variety.

[0017] Compared with the prior art, the present invention has the following beneficial effects: (1) The present invention uses broodstock from different geographical locations to construct more than 120 full-sib families. The family seedlings are cultured in cages to 12 cm and injected with electronic markers. Then, more than 30 individuals are randomly sampled and measured from each family as the initial growth data before the grass carp family experiment. After mixed breeding, they are used for high-density breeding experiments. Finally, the growth data obtained from each family under high-density breeding environment are used as phenotypes. According to the individual weight, 1251 samples are randomly selected for resequencing using the frequency accumulation method to construct a reference population. The high-quality SNP sites screened out are used to establish the GEBV of grass carp's high-density breeding resistance trait using the genomic kinship matrix + mixed linear model method. High-GEBV parents are used to breed seedlings to obtain intensive and high-yield grass carp.

[0018] (2) This invention uses grass carp as the research object, conducts high-density farming, obtains growth traits, and uses genomic information for breeding. It establishes an efficient breeding technology based on grass carp genomic information for breeding grass carp suitable for intensive farming. This technology addresses the shortcomings of existing technologies such as low accuracy and low efficiency, solves the problem of the lack of high-density breeding methods for grass carp intensive farming, and provides technical means to meet the urgent need for new specialized grass carp varieties in my country's facility farming. It has important significance and application value for the breeding of high-yield and high-quality fish.

[0019] (3) The method of the present invention can produce a grass carp variety with an intensive and high-yield phenotype. This variety is bred by accurately screening parents with high genomic estimated breeding values ​​(GEBV). Under high-density farming conditions, its growth traits (such as body weight, total weight gain, daily weight gain, etc.) and survival rate are excellent. For example, the average daily weight gain of some key families is much higher than the overall average, and the survival rate is also significantly higher than the group average. At the same time, this variety can be directly applied to grass carp seed breeding, effectively improving the ability of grass carp to tolerate high-density farming, providing a special high-quality variety for intensive and facility-based farming models, helping the transformation, upgrading and high-quality development of the grass carp farming industry, and meeting the urgent needs of the aquaculture industry for high-yield and high-efficiency farming varieties. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a graph of the average daily weight gain of grass carp families in a high-density culture environment; Figure 2 This is a graph of the survival rate of grass carp families in a high-density culture environment; Figure 3 This is the average daily weight gain graph of grass carp family under normal density breeding environment; Figure 4 It is a growth comparison chart of offspring of broodstock with high and low breeding values ​​for intensive high-yield traits, where Group 1 is the offspring individuals of broodstock with high breeding values, and Group 2 is the offspring individuals of broodstock with low breeding values. (a) is the growth comparison chart of offspring after the first measurement, (b) is the growth comparison chart of offspring after the second measurement, (c) is the growth comparison chart of offspring after the third measurement, (d) is the growth comparison chart of offspring after the fourth measurement, and (e) is the growth comparison chart of offspring after the fifth measurement. DETAILED DESCRIPTION

[0021] The present invention will be further described below in conjunction with specific embodiments.

[0022] It should be understood that the examples are only used to illustrate the present invention and are not used to limit the scope of the present invention.

[0023] In addition, after reading the contents of the present invention, those skilled in the art may make various changes or modifications to the present invention, and these equivalent forms also fall within the scope defined by the claims attached to this application.

[0024] Unless otherwise specified, all raw materials used in the examples are commercially available.

[0025] Example 1 This example provides a process for establishing a grass carp family.

[0026] Grass carp breeding and farming currently have relatively mature, economical, and family-independent technical processes. It's difficult to find a farm that can meet all the requirements for large-scale grass carp family establishment. Therefore, this application separates grass carp family establishment and seedling cultivation into three phases: family establishment, seedling hatching, and water fry rearing, to be completed in stages. After the grass carp family seedlings reach a length of 12.5±2.5 cm, intensive high-yield experiments will be conducted.

[0027] (1) Establishment of grass carp family Mature grass carp individuals or semen were collected from farmed populations in different regions. Individuals with normal body color, smooth, healthy skin, and plump abdomens nearing maturity or mature were selected from farmed populations in Enping and Heshan, Jiangmen (April and May 2024), Yangchun, Yangjiang (April and May 2024), and Sanshui, Foshan (April and May 2024). These individuals were used for artificial insemination and indoor hatching. Semen from grass carp was also collected from Chongqing (May 2024) for artificial insemination.

[0028] Using the "two-injection method," male and female fish are injected with induced fertilization, followed by one-on-one insemination. The sperm is squeezed out first, followed by the eggs. After mixing, the eggs are rinsed with distilled water for activation and then placed into corresponding packaging bags. At least 2000ml of well water is added based on the number of fertilized eggs. After oxygenation and packaging, a parental pairing is complete. For successfully paired families, fin ray samples are collected from both parents and preserved in 100% alcohol.

[0029] The coding rule for grass carp family is F24&pairing number. The pairing number part is coded incrementally starting from 001. If the pairing is unsuccessful, only the pairing number is recorded without family coding.

[0030] (2) Hatching of grass carp fry Fertilized grass carp eggs are oxygenated and packaged for transport in batches. Water is pre-filled in the hatching bucket. Upon arrival, the eggs are allowed to acclimate to the water temperature for 10 minutes before the bags are opened and the eggs are added. Pure oxygen is used to control the appearance of slight bubbles on the water surface. Recirculating water is used during incubation to gently flush the eggs. When the water temperature is between 22-25°C, the eggs hatch 21-30 hours after fertilization. Once the eggs have broken through the membrane and grown a waist point, they are packaged and shipped within a day. Each bag represents a family, which is then raised in cages. A total of 129 families have been successfully cultivated in the ponds. The detailed family construction is shown in Table 1.

[0031] Table 1

[0032] (3) Grass carp family seedling cultivation A 2m x 3m x 1.5m cage culture system was constructed. Cages were replaced every 30 days according to the specifications of grass carp fry and feed. The cages had 100-mesh, 7-mesh, and 9-mesh mesh sizes, respectively. When the primary cages were replaced by the secondary cages, the fry were screened through a 7-mesh sieve, retaining the fry of the family with a body length of 4±1cm. When the secondary cages were replaced by the tertiary cages, the fry were screened through an 11-mesh sieve, retaining the fry of the family with a body length of 8±1cm. After the fry were raised to 10±1cm, the experimental fish were vaccinated with a live viral hemorrhagic disease vaccine and a bacterial vaccine.

[0033] (4) Grass carp family seedling marking injection and initial measurement When the fry reached 12 cm, families with fewer than 20 fish were eliminated or merged. 310 individuals from each family were electronically tagged intramuscularly. A total of 31,405 grass carp from 118 full-sib families and four mixed families were electronically tagged. Testing was performed after the injection wounds stabilized. A high-throughput growth trait measurement instrument and supporting measurement and analysis system were developed to measure the shape of grass carp. At least 30 individuals from each family were sampled, for a total of 10,532 tagged grass carp. The measurements included 23 traits: body weight, total length, body length, head length, snout length, anal fin length, anal fin base length, pelvic fin length, pelvic-anal fin base distance, pelvic-dorsal fin distance, pectoral fin length, pectoral-pelvic fin distance, pectoral-dorsal fin length, dorsal fin length, dorsal fin base length, dorsal fin posterior distance, dorsal fin anterior distance, eye diameter, head length behind the eye, caudal fin length, caudal peduncle height, caudal peduncle length, and body height. These traits served as initial growth data for the grass carp families. The average grass carp weight was 19.42g. Each grass carp family was divided into three groups: a high-density group, a normal group, and a seed-retention group, which were then mixed-pond reared. The first two groups were tested for intensive high-yield traits, while the last group was shipped to Sanya, Hainan, where they leveraged the high accumulated temperatures there to accelerate the development of candidate parents.

[0034] Example 2 This example provides the relevant process of determining the intensive high-yield traits of grass carp.

[0035] (1) Grass carp high-density breeding experiment A total of 18,200 grass carp from 116 families were selected and divided into two density groups: a normal density of 15,000 fish per mu (approximately 1.5 acres) and a high density of 75,000 fish per mu (approximately 1.5 acres). The rearing areas were set at 0.6 mu (approximately 1.5 acres) and 0.2 mu (approximately 1.5 acres), respectively. Seventy-two fish from each family were randomly selected for the normal density group and placed in the experimental tanks, while 150 fish from each family were randomly selected for the high density group. If the number of fish was insufficient, the density was filled with fish from unmarked families. The normal density group contained 6,000 marked fish, while the high density group contained 12,200 marked fish, supplemented by 2,000 unmarked fish, for a total of 14,200 fish. Each experimental tank was equipped with an aerator to maintain a minimum dissolved oxygen level of 3 mg / L per day. A dissolved oxygen level of 5 mg / L or above promotes faster growth of grass carp.

[0036] (2) Determination of grass carp growth traits After 3 months of breeding in a high-density environment, the stage growth traits and survival rate phenotypes of 116 grass carp families were measured. According to the stage growth data results, the average daily weight gain of 7706 grass carp in the high-density breeding group was 0.51 g / day, and then the daily weight gain of 116 families was calculated separately. Three families, F24008, F24068, and F24057, were screened out. The average daily weight gain of grass carp in the families was 1.26 g / day, 1.16 g / day, and 1.10 g / day, respectively, which were more than 2 times higher than the overall average. The above families can be used as new fast-growing germplasm for grass carp in high-density breeding environments. Figure 1 The average survival rate of the high-density culture group was 61.22%, and then the survival rates of 116 families were calculated. Three families, F24006, F24081, and F24019, were selected. The survival rates of these families were 85.30%, 83.00%, and 82.70%, respectively, which were more than 30% higher than the overall average. They can be used as new germplasms with high survival rates for grass carp in high-density culture environments. Figure 2 shown.

[0037] At the same time, phenotypic data were collected from grass carp families participating in the experiment using a combination of manual and computer-generated measurements. Based on the growth data of 2,796 grass carp collected by equipment in the normal density group, weight, total length, body length, head length, snout length, anal fin length, anal fin base length, pelvic fin length, pelvic-anal fin base distance, pelvic-dorsal fin distance, pectoral fin length, pectoral-pelvic fin distance, pectoral-dorsal fin length, dorsal fin length, dorsal fin base length, dorsal fin posterior distance, dorsal fin anterior distance, eye diameter, head length behind the eye, caudal fin length, caudal peduncle height, caudal peduncle length, and body height were calculated. The results showed that, with the exception of eye diameter, snout length, and head length behind the eye, the correlation coefficients between all other length traits and weight were above 0.85.

[0038] In the normal density group, the average daily weight gain of 3431 grass carp was 1.19 g / day. Then the daily weight gain of 116 families was calculated. The families F24004, F24086, and F24084 were selected. The average daily weight gain of the families was 2.00 g / day, 1.98 g / day, and 1.97 g / day, respectively. They were all more than 60% higher than the overall average and were selected as new fast-growing grass carp germplasms. Figure 3 shown.

[0039] Example 3 Establishment of a reference population for intensive and high-yield traits of grass carp (1) Whole genome resequencing and variation detection of grass carp reference population Individual growth data from grass carp families, obtained by measuring growth traits after three months of rearing in a high-density environment, were sorted in descending order by the primary keyword "weight" and in ascending order by the secondary keyword "family." One hundred families with 12 or more individuals were then selected. Family F24057, although consisting of only eight individuals, was also included as a reference population due to its high-density, rapid growth. Finally, individuals were selected using the cumulative weight frequency method within each family. From each family, 12-15 individuals were randomly selected, for a total of 1,251 fin rays.

[0040] Grass carp genomic DNA was extracted and purified using the standard protocol provided by the DNA extraction kit (TIANGEN, Beijing). In addition to extracting genomic DNA from the reference population, genomic DNA was also extracted from 182 parents of the reference population to validate the method described herein. Illumina paired-end libraries were constructed using the purified genomic DNA and then sequenced using the Illumina HiSeq 2000 sequencing platform. After filtering out low-quality reads, these reads were aligned to the grass carp reference genome using the BWA software. Single-nucleotide polymorphisms (SNPs) were detected using the GATK software. Ultimately, 1,238 individuals from the reference population and 182 individuals from the validation population were successfully sequenced, generating 10 GB of individual sequencing data. A total of 11,097,704 SNP genetic variants were identified, with an average sequencing depth of 11.2X. The variant results are stored in the file "Cl.variances. vcf.gz."

[0041] (2) Screening of high-quality SNP variant sites in grass carp.

[0042] Follow the steps below to filter the above variants, retain high-quality SNP sites, and fill in missing site information. Run the following command in a Linux environment: plink2 --vcf Cl.variances.vcf.gz \ --autosome-num 24 --autosome \ --max-alleles 2 --min-alleles 2 \ --maf 0.05 --geno 0.05 --mind 0.1 --hwe 0.0001 \ --allow-extra-chr --sort-vars \ --make-bpgen –out high-qual.snps plink2 --bpfile high-qual.snps \ --autosome-num 24 \ --export-allele info.alleles.txt \ --export A --out high-qual.geno Linux-R: library(data.table) geno<- fread("high-qual.geno.raw") geno[, c(1: 6):= NULL] M<- as.matrix(geno) save(M, file = “geno4cal.bin”) Specifically it can be described as: Run the plink2 software to filter the original variant file "Cl.variances.vcf.gz". The parameters set include: retaining only autosomes (24), diallelic loci (the number of the smallest allele and the number of the largest allele are both 2), the minimum allele frequency (MAF) ≥ 0.05, the genotype missing rate ≤ 5%, the individual genotype missing rate ≤ 10%, and conforming to Hardy-Weinberg equilibrium ( p =0.0001), the retained variants were sorted after quality inspection, and the filtered high-quality SNP file "high-qual.snps" was output.

[0043] Based on the above output file "high-qual.snps", run the plink2 software again, specify 24 autosomes, export allele information to "info.alleles.txt", and export genotype data in "A" format to "high-qual.geno".

[0044] In the R environment of the Linux system, load the data.table package and read the genotype file "high-qual.geno.raw". Delete the first six columns of non-genotype data in the file and convert the remaining data into a matrix format named "M". Finally, save the matrix "M" to the file "geno4cal.bin" for subsequent analysis.

[0045] After running the above code, 2,177,594 high-quality SNPs remained for subsequent analysis. The genotypes of the SNPs were recoded using 0 / 1 / 2 and stored in the file "geno4cal.bin." To extract the minor allele SNP number and genotype from the file "info.alleles.txt" required for recoding, run the freq command in PLINK2.

[0046] Example 4 This example establishes a prediction model for intensive high-yield traits of grass carp.

[0047] Under high-density experimental conditions, three intensive high-yield phenotypes of grass carp were measured: final body weight at the end of the experiment, total weight gain during the experiment, and daily weight gain. These three phenotypes were fitted using mixed linear models, resulting in six analytical models for predicting the breeding values ​​of candidate parents. The definitions of each model are shown in Table 2: Table 2

[0048] The distribution of individual random effects in the model is ,in is the individual additive genetic variance; G The genomic kinship matrix constructed using the high-quality SNPs obtained in Example 2, and the specific implementation method of the genomic kinship matrix and its inverse matrix are as follows (Linux-R): library(data.table) library(parallel) load("geno4cal.bin") ginv<- function(invG){ Ginv<- data.frame(row = rep(1 : nrow(invG), nrow(invG)), column = rep(1 : nrow(invG), each = nrow(invG)), value = as.numeric(invG), lower.mat = as.logical(lower.tri(invG, diag = T))) Ginv<- Ginv[Ginv$lower.mat == T, c("row", "column", "value")] Ginv<- Ginv[order(Ginv$row, Ginv$column), ] return(Ginv) } iid<- fread("high-qual.snps.fam")$V2 A<- diag(1, length(iid)) pi<- round(colSums(M) / (2 * nrow(M)), 3) P<- matrix((2 * pi), byrow = T, nrow = nrow(M), ncol = ncol(M)) Z<- as.matrix(M - P) Zt<- t(Z) ZZt<- do.call('rbind', mclapply(1: nrow(Z), FUN = function(x) {Z[x, ]%*% Zt}, mc.cores = 10)) sum2pq<- 2 * (sum(pi * (1 - pi))) G<- (ZZt / sum2pq) w<- 0.95 G<- (w * G) + ((1 - w) * A) invG<- solve(G) Ginv<- ginv(invG) attr(Ginv, "rowNames")<- iid attr(Ginv, "INVERSE")<- TRUE save(Ginv, file = "Ginv.RefHD.Can.bin") After running, the genomic kinship matrix information required for model operation is stored in the file "Ginv.RefHD.Can.bin". The specific steps of the above code can be described as follows: (1) Data preparation Load genotype data (geno4cal.bin) and individual IDs (high-qual.snps.fam).

[0049] Create the identity matrix A as the base affinity matrix.

[0050] (2) Calculation of allele frequencies The minor allele frequency (pi) of each SNP site is calculated, that is, the proportion of the minor allele at this site in the population.

[0051] (3) Construction of standardized genotype matrix Create an allele frequency matrix P and normalize the genotype of each individual by allele frequency.

[0052] Calculate the standardized genotype matrix Z (Z=MP), where M is the original genotype matrix.

[0053] (4) Parallel calculation of genome kinship matrix The ZZt matrix (the product of Z and its transpose) was generated by parallel computing (10 cores) to capture the genetic similarity between individuals.

[0054] Calculate the normalization factor sum2pq (the sum of 2pq for all SNP sites, where p and q are allele frequencies).

[0055] Generate the initial G matrix: G=ZZt / sum2pq.

[0056] (5) Integrating genomic and pedigree information The weight w=0.95 is used to fuse the genome matrix G and the pedigree unit matrix A, with the formula: G=0.95G+0.05A. This step balances the accuracy of the genome data and the stability of the pedigree data.

[0057] (6) Matrix inversion and formatting After calculation fusion G The inverse matrix invG.

[0058] Convert the inverse matrix to a triplet format (row, column, value), sort it by row and column, and compress the lower triangular part.

[0059] Add individual IDs and matrix attributes and save as a binary file (Ginv.RefHD.Can.bin).

[0060] In R, using the built G The matrix is ​​used to fit the intensive high-yield phenotype of grass carp using different models. The variables for final weight, total weight gain, and daily weight gain are named bw, bw.gains, and bw.daily; age and weight at marking are represented by Days and Avg.bw, respectively; and the variables for maternal effect and individual additive effect are named Dam and IID, respectively. Taking model M1 as an example, the specific script is as follows: library(asreml) load("Ginv.RefHD.Can.bin") #Phenotype file is stored in "Pheno.RefHD.csv" pheno<- asreml.read.table("Pheno.RefHD.csv", sep = ",", header = T) #Fitting with M1 model M1<- asreml(bw ~ 1 + Avg.bw, random = ~vm(IID, Ginv), dense = ~vm(IID, Ginv), maxiter = 50, data = pheno) #Calculate heritability vpredict(M1, h2 ~ V1 / (V1 + V2)) #Calculate the candidate parent's intensive high-yield GEBV gebv_M1<- coef(M1)$random write.table(gebv_M1, "Gebv.M1.csv", sep = ",", quote = F) After running the script, the intensive high-yield GEBVs of the candidate parents are saved in the file "Gebv.M1.csv." In the above script, bw corresponds to final body weight. To analyze other phenotypes, replace bw with bw.gains or bw.daily. The variable to the right of the phenotypic "~" is a fixed effect and can be replaced with a different fixed effect depending on the model. 1 represents the population mean. The maternal effect is Dam. If this effect is included in the model, the term "random" should be changed to random = ~vm(IID, Ginv) + Dam.

[0061] The following is an explanation of the above code: (1) Data preparation: Load the precomputed inverse genomic relationship matrix (Ginv.RefHD.Can.bin).

[0062] Read the CSV file containing phenotypic data (Pheno.RefHD.csv), which includes phenotypes such as final weight (bw), total weight gain (bw.gains), daily weight gain (bw.daily), and covariates such as age (Days) and weight at marking (Avg.bw).

[0063] (2) Model construction: Taking model M1 as an example, use the asreml() function to fit the mixed linear model: Fixed effects: intercept term (1) and body weight at the time of labeling (Avg.bw).

[0064] Random effects: Individual additive genetic effects (IID), parameterized by the genomic relationship matrix (Ginv).

[0065] You can analyze different traits by replacing phenotypic variables (such as bw.gains or bw.daily), or adjusting for fixed effects (such as adding Days).

[0066] If the maternal effect is considered, the Dam term needs to be added to the random parameter.

[0067] (3) Heritability estimation: Use vpredict() function to calculate heritability ( h ²), the formula is: h ² = additive genetic variance / (additive genetic variance + residual variance).

[0068] (4) GEBV prediction and output: Random effect coefficients (i.e., individual GEBVs) were extracted from the model results.

[0069] Save the predicted GEBV as a CSV file (such as Gebv.M1.csv) for subsequent parent screening.

[0070] Example 5 This embodiment provides the relevant process of selecting high-yield grass carp suitable for intensive farming.

[0071] (1) Estimation of heritability of intensive high-yield The model established in Example 4 and the software ASReml-R were used to calculate the heritability of intensive high-yield traits of grass carp under high-density culture experimental conditions.

[0072] The results showed that the heritability of intensive high-yield traits in grass carp ranged from (0.347±0.068) to (0.494±0.080), indicating medium-to-high heritability. High genetic progress can be achieved with a well-planned breeding program. Furthermore, the maternal effect was not significant in models M1 and M2, but was significant in models M3 and M5. Therefore, models M1, M4, and M6 were used to predict the breeding values ​​of candidate grass carp parents. The genetic parameters of the high-density group are shown in Table 3.

[0073] Table 3

[0074] (2) Summary of family parent resequencing According to pedigree records, there are 180 grass carp pedigrees, of which 182 parents were successfully sequenced. The sequencing status of the parents of each family is shown in Table 4.

[0075] Table 4

[0076] (3) Analysis of Estimated Breeding Values ​​(GEBV) of Reference Populations for Intensive High-Yield Traits a) Using body weight as the phenotype, the above analysis model calculated the average GEBV of the families to be -1.045. There were 37 families with a GEBV greater than the average, of which the top 20 were F24008, F24035, F24093, F24101, F24106, F24084, F24089, F24012, F24063, F24121, F24019, F24081, F24011, F24002, F24022, F24085, F24029, F24100, F24086, and F24108. The prediction results using body weight as the phenotype are shown in Table 5.

[0077] Table 5

[0078] b) Using weight gain as the phenotype, the above analysis model calculated the average GEBV of the families to be -0.812. There were 44 families with a GEBV greater than the average, of which the top 20 families were F24008, F24116, F24035, F24101, F24084, F24103, F24093, F24106, F24081, F24012, F24117, F24120, F24107, F24089, F24110, F24090, F24080, F24018, F24109, and F24092. The prediction results using weight gain as the phenotype are shown in Table 6.

[0079] Table 6

[0080] c) Using daily weight gain as the phenotype, the average GEBV of the families calculated by the above analysis model was -0.001. There were 42 families with a GEBV greater than the average value, of which the top 20 families were F24008, F24035, F24084, F24012, F24081, F24088, F24020, F24017, F24018, F24021, F24009, F24101, F24019, F24085, F24029, F24024, F24080, F24078, F24063, and F24116. The GEBVs of the families using daily weight gain as the phenotype are shown in Table 7.

[0081] Table 7

[0082] Based on the above results, the intersection of the top 20 families from the three phenotypic prediction results was taken, resulting in a total of 6 families: F24008, F24035, F24101, F24084, F24012, and F24081. These 6 families were selected as key families for intensive high-yield production. Detailed information is shown in Table 8.

[0083] Table 8

[0084] Example 6 This example provides the relevant process of cultivating high-yield grass carp adapted to intensive farming and determining its characteristics. (1) Prediction and application of GEBV in candidate populations for intensive high-yield traits of grass carp In November 2024, all broodstock were tagged and archived, and fin rays were collected, totaling 848 individuals. Genotypes for the candidate population were obtained using whole-genome resequencing, and the genotype information is stored in the file "candidates.geno.vcf.gz." Using the phenotypic and genotypic data of the constructed reference population, the GEBV of the candidate population was estimated using the G-BLUP method, with an overall average breeding value of -0.148. According to the breeding plan, 317 grass carp parent individuals were selected for breeding and subsequent validation of the predictions. Among these, 101 parent individuals had high breeding values, with an average individual breeding value of 12.995; 216 parent individuals had low breeding values, with an average individual breeding value of -6.873.

[0085] (2) Determination of offspring phenotype To further demonstrate that individuals with high GEBV for intensive high-yield production have strong tolerance to high-density farming, the present invention uses a high-density stress experiment to determine the high-density farming tolerance of the progeny of a candidate population. Progeny individuals that grow fast in a high-density farming environment are identified as tolerant to high-density farming, while progeny individuals that grow slowly in a high-density farming environment are identified as intolerant to high-density farming. The specific implementation details are as follows: The broodstock with high and low breeding values ​​for intensive high-yield traits were bred, and 6 million water splashes were left for each offspring. They were cultured in two 8-mu ponds at the Zhongshan base. The growth traits were measured every 7 days for preliminary comparison. A total of 5 measurements were made. The first 4 measurements were made on randomly selected individuals, and the fifth measurement was made on the first fry. The growth of the offspring of broodstock with high and low breeding values ​​for intensive high-yield traits was compared. Figure 4 As shown in the figure, Group 1 represents the offspring of bred fish with high breeding values ​​for intensive high-yield traits, and Group 2 represents the offspring of bred fish with low breeding values ​​for intensive high-yield traits. Body lengths were compared five times, once every seven days. It can be seen that the offspring of bred fish with high breeding values ​​for intensive high-yield traits significantly outperformed the offspring of bred fish with low breeding values ​​for intensive high-yield traits in terms of growth data.

Claims

1. A method for intensive high-yield trait genomic selection breeding of grass carp, characterized in that: Including the establishment of a prediction model for intensive high-yield traits of grass carp; The grass carp intensive high-yield trait prediction model is established and comprises the following steps: (1) Obtaining single nucleotide polymorphism sites of candidate grass carp broodstock; (2) Construct an analytical model to predict the breeding values ​​of candidate parents; (3) Using the obtained single nucleotide polymorphism sites to construct a molecular kinship matrix, combined with the analysis model in step (2) to predict the genomic estimated breeding value of the candidate population; The method for constructing the kinship matrix is ​​as follows: S1. Construct and load the recoded single nucleotide polymorphism genotype matrix M and individual identifier iid to generate the pedigree-based identity matrix A; S2. Calculate the minor allele frequency pi of each SNP site, and construct the minor allele matrix P and the normalization matrix Z=MP in turn; S3. Parallel calculation of the product ZZt of the row vector of the normalization matrix Z and the transposed matrix, and calculation of the normalization factor sum2pq = 2 × (sum(pi × (1-pi))), and the initial genome kinship matrix G = ZZt / sum2pq is obtained by the normalization factor sum2pq; S4. Matrix G is fused with matrix A using weight w, using the formula G = wG + (1-w) A, to obtain a weighted genomic kinship matrix. S5. Calculate the inverse matrix of G and format and store it for use in predicting breeding values ​​for intensive high-yield traits of grass carp.

2. The method according to claim 1, characterized in that The screening of the single nucleotide polymorphism sites in step (1) meets the following conditions: the minimum allele frequency ≥ 0.05, the genotype missing rate ≤ 5%, the individual genotype missing rate ≤ 10% and conformity to the Hardy-Weinberg equilibrium ( p =0.0001).

3. The method according to claim 1, characterized in that The analytical model for predicting the breeding values ​​of candidate parents in step (2) satisfies the following conditions: phenotype = fixed effect + random effect + residual.

4. A grass carp variety bred using the method according to any one of claims 1 to 3, characterized in that: The grass carp variety has an intensive and high-yield phenotype.

Images