KR20180049486A - A breeding pig economic trait prediction using SNP chip and related business model - Google Patents

Abstract

The present invention establishes a method for predicting the economic trait of a pig using Yorkshire genome data, which is a maternal breeding pig, confirms the industrial applicability of indicators regarding bearing and raising traits (daily weight gain, back fat thickness, age at 90 kg, fresh meat rate, sirloin thickness) and indicators regarding a breeding trait (total litter size, actual litter size), verifies the effects thereof, estimates the effect of marking all genomic variation within the chip data by using the same, and proposes the business model related thereto on the basis of a technique enabling a breeding pig genome breeding value service when a pig breeding farm has only genome data for the above traits. Based on the marker effect information estimated from reference genome data, a desired pig breeding farm uses genome breeding values more accurate than conventional breeding values in order to select pigs having fundamentally improved breeding and bearing and raising traits and improve the whole ability of the pig breeding farm.

Description

[0001] The present invention relates to a method for predicting economic traits of pigs using a Yorkshire SNP chip data analysis and a business model using the same,

The present invention relates to a method for predicting economic traits of pigs using genomic data and a business model related thereto. For this purpose, reference genomic data is prepared, a predictive mathematical model is set up, and effects are verified. The present invention relates to a business model for producing genomic data in a hog house by producing information of a genetic breeder in the field and estimating a genetic breeder price for a pig economic trait of interest.

Since 2001, a genome selection method using SNP genome information for pure gold has been developed using multi-step method. Genomic selection has been established through a reference group and genomic estimated breeding value (GEBV) It is used for selection. (Meuwissen, 2001)

Since the method of predicting the economic characteristics of pigs by selecting some SNPs is too low to be applied to industrial applications, a genome selection method has been developed to estimate breeding price using whole SNPs. The most common method for estimating the effect of SNP markers and calculating the breeding price using this method and the two methodologies for constructing the genetic relational matrix and replacing it with the existing coefficient matrix is the most advanced.

In contrast to dairy cows, high selection intensity and a relatively short generation interval have led to many questions about the genetic selection effect in many animal husbandry varieties, but despite the low accuracy of EBV in pigs, , And multinational breeders have applied it to business (Muir, 2007).

Forni et al. (2011) estimate the genetic breeding value for the number of pigs by using the genetic relation matrix. Currently Forni et al. (2011) are working on application of genome selection technology in the parent company PIC. It is known that pig selection is proceeding. As a result of estimating the genome-related matrix breeding price of land race species in 2013 in Korea according to the global dairy flow, the genetic relation matrix was higher than the breeding price using the pedigree information, and there was a lack of phenotype data in the population or Mendelian distribution Using genomic information is expected to greatly improve the accuracy of genetic evaluation when predicting the breeding value of young piglet piglets that are not considered at all.

In 2013, Akanno and colleagues found that genomic selection is advantageous to genetic improvement faster than traditional selection methods in the absence of pedigree data. Based on this research, And genetic data related to sow productivity, such as age, sex, and gestation period, and examined genetic engineering techniques suitable for sows in swine genomics by studying the Yorkshire genome selection technique using penumbrae reduction estimation techniques.

As a policy for the development of domestic pig breeding industry and swine industry, we have established a nationwide pig improvement net- work to build a national genetic capacity evaluation system by raising the size of the breeding pig breeding group, and establish the appropriate size of the participating companies and government policy consistency (Hayes, 2009; Loberg and Durr, 2009), the genetic screening has been widely used in cows with very long generations, (Brune, 2011).

Recently, Misztal et al. (2013) found that the method based on the genomic relationship matrix used for genome selection, when combined with phenotype data containing no genotypes using BLUP, involves several steps in the method of combining information, (SsGBLUP) has been reported to be useful in order to simplify this. In addition, there is an attempt to apply this method to the evaluation of domestic domesticated genomes if they are most widely used as US small genome evaluation methods. There are diverse databases available for estimating the breeding price, such as economic traits such as cattle and sheep, economic traits favorable to racing horses, possible genetic disease incidence, and so on. However, based on genetic information, Very few databases exist.

It is an object of the present invention to overcome the problem of selecting excellent markers by selecting several markers and using the whole genome data of pig commercialization chips which are currently used most in the world, The genetic breeding cost is estimated by estimating the genetic breeding cost and the whole marker effect is estimated by using the reference data of the reference point to estimate the genetic breeding cost. We propose a business model that provides information.

In order to achieve the above goal, 2,128 yorkshire's Illumina 60K chips, which are used as annual pigs from 2007 to 2016, are used to produce genomic data and provide them as a reference genome group for selection of breed genome and can be continuously updated.

In the present study, the traits of interest in selection of breeding stocks were classified into the raising traits and breeding traits. In the case of raising traits, daily gain, backing thickness, reaching age of 90kg, The phenotypic information above is used to standardize the accumulated data for 1997 ~ 2016, and it can be continuously updated in the business model.

The BLUPF90 program, which can be analyzed by the ssBLUP method developed by Professor Misztal of the University of Georgia, USA, is used for estimating the breeding price and estimating the marker effect. In the case of ssBLUP, unlike the G-BLUP method, which uses only the entities having existing genomic data, it uses a method of combining entities having genomic data and entities not having genomic data. It is possible. Based on this, the genetic breeding price is estimated, and based on this, the total marker effect is estimated. Thus, the genetic breeding price can be estimated based on the genetic data alone.

After the reference data are prepared and the standardized phenotype information is prepared, it is proved that the breeding price evaluation is superior and the marker effect is estimated. The internal system capable of producing the genome chip data for the veterinary breeding price estimation, We propose a business model that provides genetic service by estimating genome breeding price using genomic data.

As described above, based on the marker effect information estimated from the reference genome data, the desired hatchlings are more suitable for selection of pigs having excellent breeding and raising traits and using the correct genetic breeding price than existing breeding stocks and improving the overall ability of the hatchlings Lt; / RTI >

Figure 1. Dielectric Selection System
2. Diagram of genome breeding price estimation using reference data
2. Diagram of genetic breeding price estimation using estimated marker effect
Figure 4. Genetic breeding calculation using estimated marker effect and genomic data
Figure 5. Business model using genome breeding price estimation system

Hereinafter, the present invention will be described in more detail by way of specific examples. It will be apparent to those skilled in the art that the present invention is not limited by the following embodiments, but may be modified or modified in various ways within the scope of the present invention.

Example  One. Reference dielectric  collecting data

Genomic data were generated for four representative domestic hoglings using the Illumina porcine60 SNP chip (version 2) for reference genomic data collection for genome selection. A total of 2,128 genomic data were produced from 2007 to 2016 Respectively. The phenotypic data accumulated during 1997-2016, and 2,128 strains carrying genomic data confirmed phenotype and pedigree data.

Farm Porcinesnp60v2 porcine70 Total A 143 - 143 B 350 - 350 C 1,090 - 1,090 D 545 - 545 Total 2,128 - 2,128

Example  2. On breeding traits  Verification of superiority of selection of Korean genome

The genome selection for breeding traits was verified by using 3,443 data, including 210 genotypes of total phenotypes and genomic data. The statistical model for reproductive traits in this study is as follows.

In the above model, Y contains phenotypes (total number and number of live births) and parity, herd-year-week (farm-birth-week) and permanent environment for breeding traits. Using the above model, the analysis using the pedigree data (BLUP) and the whole data and the pedigree data (ssBLUP) were performed according to the existing method.

Breeder's
accuracy all
(3,443) Genome analysis
(210 pairs) with record
(198 pairs) Without record
(Twelve) BLUP 0.380 + 0.111 0.265 0.080 0.274 + 0.072 0.104 + 0.039 ssBLUP 0.382 + 0.108 0.300 0.074 0.304 + 0.072 0.213 + 0.048

Breeder's
accuracy all
(3,443) Genome analysis
(210 pairs) with record
(198 pairs) Without record
(Twelve) BLUP 0.370 + - 0.109 0.257 ± 0.079 0.266 + 0.070 0.100 + 0.037 ssBLUP 0.373 + - 0.107 0.290 + 0.073 0.296 + 0.071 0.207 + 0.048

Based on the above results, it can be concluded that the genome selection result is more accurate than the analysis using the existing lineage data, regardless of the phenotype of the whole data.

Example  3. On the traits  Verification of superiority of selection of Korean genome

The genome selection of the female traits was verified by using 24,236 data including 2,065 genotypes of total phenotypes and genomic data. In this study, the statistical model of the trait is as follows.

In the above model, Y contains the phenotype (gross body weight, backfat thickness, reaching age of 90 kg, meat size, sirloin depth) and parity, sex, and herd-year-week (farm-birth year-birth week) . Using the above model, BLUP was used for analysis of blood flow data, and ssBLUP was used for both genetic data and blood flow data.

Comparison of BLUP and ssBLUP estimated breeding values for individuals without phenotypic phenotypes correlation BLUP ssBLUP daily_wt (daily gain) 0.204 0.271 bf_avg (backfill thickness) 0.210 0.328 meat_per 0.258 0.347 day_90 (age reached 90kg) 0.201 0.247 bf_dep (fillet depth) 0.150 0.166

As shown in Table 4, as a result of comparing the accuracy of obtaining genome data from genotype data from individuals having no phenotype, the accuracy of genome breeding genome was higher than that of breeding genome according to conventional methods, We can make a more precise selection about traits.

Example  4. Marker effect  Estimates and Dielectric Properties for New Dielectric Data Breeder  Calculation

When estimating the genetic breeding value, an estimated marker effect value is required to calculate a relational matrix using about 40,000 SNP data and directly calculate genetic breeding value for new genome data. After estimating the genomic breeding value, the marker effect is estimated for about 40,000 SNPs using PostGSF90 in the BLUPF90 program.

After understanding the flowchart shown in FIG. 3, if the marker effect is estimated for about 40,000 SNPs by the method shown in FIG. 4, and then the marker effect is estimated as the base information, if the genetic breeding price estimation is desired in the sows, You can generate the data and obtain the genetic breeding price using the calculation method as shown below.

Example  5. Dielectric Breeder  Business model using estimation system

After having the reference genome population and proving the effect of the genome selection, if a large amount of marker effect is estimated and used to produce the genome data in a hog house, each hog will have a marker effect only from the reference genome, Can be estimated and service can be performed. In this case, it is judged that selection can be made more precisely than selecting the breeder using the existing lineage information. A series of these processes have already been carried out in the laboratory and the effectiveness has been thoroughly checked, and it is judged that it is possible to propose a business model as a technology applicable to the field.

Claims (3)

Prediction of breeding traits and characteristics of pig breeding pigs using genetic data and standardized phenotypes
(Total body weight, total body weight, body weight, body weight, body weight, 90kg reaching age, meat weight ratio, and sirloin depth) and reproductive traits In order to estimate the genetic breeding value using new genomic data,
The business model for a series of processes that produce genetic data for a variety of traits using genetic data production in a hog house (Figure -5) using the calculated marker effect to estimate genetic breeding values using reference genetic data,

Images