Method for improving pig complex character genome selection accuracy by using large-scale genome annotation information

Abstract

The invention discloses a method for improving pig complex character genome selection accuracy by using large-scale genome annotation information. The method comprises the following steps: integrating a plurality of function annotations and position information of a reference group pig genome; sNPs data of the multiple function annotation areas are extracted, and a genome genetic relationship matrix A of function annotation is constructed; estimating a variance component of each genome genetic relationship matrix A, sorting the plurality of function annotations based on the variance components, and combining the plurality of function annotations to obtain a function annotation set; deleting repeated SNPs data in the function annotation set, constructing a genome genetic relationship matrix B of the function annotation set, and calculating a random genetic effect value of the function annotation set to obtain an estimated genome breeding value of the pig. According to the method, the genomic breeding value is obtained by extracting the various genomic function annotations of the pig complex characters and fitting the mixed linear model containing the multiple random genetic effects, and the accuracy and timeliness of pig genomic breeding value prediction are improved.

Description

Technical Field

[0001] The present application relates to the field of animal genome breeding, and more specifically, to a method for improving the accuracy of genomic selection of complex traits in pigs using large-scale genome annotation information, and a device and electronic equipment for improving the accuracy of genomic selection of complex traits in pigs using large-scale genome annotation information. Background Art

[0002] Complex traits of individual animals are determined by multiple genes and environmental factors, so they are difficult to accurately predict through traditional single-gene genetic analysis. Genomic selection predicts these complex traits by utilizing the linkage disequilibrium (LD) relationship between a large number of single nucleotide polymorphisms (SNPs) and quantitative trait nucleotides.

[0003] In the field of animal genetic evaluation, genomic selection methods have achieved remarkable success. By constructing prediction models based on whole-genome SNPs data, researchers can more accurately estimate the genetic potential of individuals, improve breeding efficiency, and accelerate the selection of superior breeds. However, genomic selection is facing many challenges in predicting complex traits in pigs. Since LD varies significantly between individuals in different groups or breeds, this means that a prediction model that is effective in one group may not be directly applied to another group, such as pigs in different environments or different breeds. This population-specific LD pattern increases the complexity of cross-population predictions; in addition, the interaction between environmental factors and genotypes also increases the difficulty of predicting complex traits. Therefore, there is a need to improve the current genomic selection methods. Summary of the invention

[0004] In order to solve the above technical problems, the present application is proposed. The embodiments of the present application provide a method for improving the accuracy of genomic selection of complex traits in pigs using large-scale genomic annotation information, a device and an electronic device for improving the accuracy of genomic selection of complex traits in pigs using large-scale genomic annotation information, which integrates multiple functional annotations of different categories on the pig genome, and uses the genomic kinship matrix of the functional annotation set obtained after the merger to calculate the random genetic effect value of the functional annotation set using a mixed linear model containing multiple random genetic effects, thereby improving the accuracy and timeliness of genomic selection.

[0005] Specifically, this application involves the following aspects:

[0006] 1. According to one aspect of the present application, a method for improving the accuracy of genomic selection of complex traits in pigs using large-scale genomic annotation information is provided, comprising:

[0007] Collect multiple functional annotations and position information of the reference pig genome;

[0008] Determine multiple functional annotation regions of the reference group pig genome according to the position information of multiple functional annotations, extract SNPs data of multiple functional annotation regions and construct a genome kinship matrix A of each functional annotation in the multiple functional annotations respectively;

[0009] Estimate the variance components of the genome kinship matrix A, sort multiple functional annotations based on the size of the variance components, and merge multiple functional annotations to obtain a functional annotation set;

[0010] The duplicate SNPs data in the functional annotation set were deleted and the genomic kinship matrix B of the functional annotation set was constructed. Based on the genomic kinship matrix B, a mixed linear model containing multiple random genetic effects was used to calculate the random genetic effect values ​​of the functional annotation set to obtain the estimated genomic breeding values ​​of the target traits of the candidate group pigs.

[0011] 2. A method for improving the accuracy of genomic selection for complex traits in pigs using large-scale genomic annotation information according to item 1, wherein:

[0012] The multiple functional annotation regions of the reference pig genome were previously determined based on the position information of multiple functional annotations:

[0013] A whole-genome association analysis was performed on the genome of the reference group pigs to obtain the significant sites of the target traits of the reference group pigs, and the significant sites of the target traits were merged into multiple functional annotations as a type of functional annotation.

[0014] 3. A method for improving the accuracy of genomic selection for complex traits in pigs using large-scale genomic annotation information according to item 1, wherein:

[0015] Merging multiple functional annotations to obtain a functional annotation set includes:

[0016] Determine the combined interval based on the maximum variance component among the variance components;

[0017] The functional annotations are merged in response to the variance components being within the merge interval to obtain a set of functional annotations.

[0018] 4. The method for improving the accuracy of genomic selection for complex traits of pigs using large-scale genomic annotation information according to item 3, wherein:

[0019] Determining the combined interval based on the maximum variance component among the variance components includes:

[0020] The variance component interval between one tenth of the maximum variance component and the maximum variance component is taken as the combined interval A;

[0021] The variance component interval between one ten-thousandth of the largest variance component and one tenth of the largest variance component is taken as the combined interval B; and

[0022] The variance component interval between zero and one ten-thousandth of the maximum variance component is taken as the combined interval C.

[0023] 5. The method for improving the accuracy of genomic selection of complex traits in pigs using large-scale genomic annotation information according to item 4, wherein:

[0024] Merge interval B includes merge interval B 1 、Merge interval B 2 and merge interval B 3 ;

[0025] Merge interval B, Merge interval B 2 and merge interval B 3 The endpoints of the interval have a geometric relationship in sequence.

[0026] 6. The method for improving the accuracy of genomic selection of complex traits in pigs using large-scale genomic annotation information according to item 1, wherein:

[0027] Based on the genomic kinship matrix B, a mixed linear model containing multiple random genetic effects was used to calculate the random genetic effect values ​​of the functional annotation set to obtain the genomic breeding values ​​of the target traits of the candidate group pigs, including:

[0028] The random genetic effect values ​​of the functional annotation sets were summed up to serve as the genomic breeding values ​​of the target traits in the candidate group of pigs.

[0029] 7. According to another aspect of the present application, a device for improving the accuracy of genomic selection of complex traits of pigs using large-scale genome annotation information is provided, comprising:

[0030] Functional annotation collection unit, which collects multiple functional annotations and their position information of the reference group pig genome;

[0031] A kinship construction unit determines multiple functional annotation regions of the reference group pig genome according to the position information of the multiple functional annotations, extracts SNPs data of the multiple functional annotation regions, and respectively constructs a genome kinship matrix A for each of the multiple functional annotations;

[0032] The functional annotation integration unit estimates the variance components of the genome kinship matrix A, sorts multiple functional annotations based on the size of the variance components, and merges multiple functional annotations to obtain a functional annotation set;

[0033] Output unit, delete duplicate SNPs data in the functional annotation set and construct the genomic kinship matrix B of the functional annotation set, and based on the genomic kinship matrix B, a mixed linear model containing multiple random genetic effects is used to calculate the random genetic effect value of the functional annotation set to obtain the genomic breeding value of the target trait of the candidate group pigs.

[0034] 8. The device for improving the accuracy of genomic selection of complex traits of pigs using large-scale genome annotation information according to item 7, further comprising:

[0035] The functional annotation supplement unit conducts genome-wide association analysis on the reference group pig genome to obtain the significant sites of the target traits of the reference group pigs, and merges the significant sites of the target traits into multiple functional annotations as a type of functional annotation.

[0036] 9. According to another aspect of the present application, there is provided an electronic device, comprising:

[0037] Processor; and

[0038] A memory, in which computer program instructions are stored, and when the computer program instructions are executed by a processor, the processor executes a method for improving the accuracy of genomic selection of complex traits of pigs using large-scale genome annotation information according to any one of items 1-6.

[0039] Beneficial Effects

[0040] 1. The method proposed in this application for improving the accuracy of genomic selection for complex traits of pigs using large-scale genomic annotation information can improve the accuracy of genomic selection by integrating the functional annotation information of the genome, accelerate the genetic progress of the target trait, and shorten the breeding cycle. The method can obtain the genomic kinship matrix of each functional annotation based on the position information of multiple functional annotations of the pig genome and the genome information of the reference group to determine the genetic characteristic relationship of the complex traits of pigs, merge the functional annotations with similar variance components as random genetic effects to predict the genomic breeding values ​​of individual animals with target traits, so that the genomic breeding values ​​of multiple target traits can be used to construct a comprehensive selection index of the target, sort the individuals of the target animals according to the comprehensive selection index, and provide a list of candidate individuals for selection to optimize the breeding strategy.

[0041] 2. The method proposed in the present application for improving the accuracy of genomic selection of complex traits in pigs using large-scale genomic annotation information can extract multiple functional annotations and their position information from the pig genome without limiting the category of the required functional annotations. Up to hundreds of functional annotation information can be extracted from the pig genome and used as multiple random genetic effects. The information dimension is reduced by merging and a mixed linear model is quickly fitted to calculate the individual genomic breeding value. Therefore, while maintaining accuracy, it has higher versatility and popularity in pig breeding evaluation, and can avoid the low adaptability of the model constructed by specific functional annotation information to LD heterogeneity of the pig population, while avoiding the problems of low efficiency and poor real-time performance when constructing models using multiple functional annotations.

[0042] 3. The method proposed in this application for improving the accuracy of genomic selection of complex traits in pigs by using large-scale genome annotation information has increased the prediction accuracy of the breeding value of target pigs by 4.61% compared with the optimal level of breeding prediction achieved by existing genome selection methods, providing pig breeding technicians with a higher selection priority option, further improving the efficiency and level of breeding optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0044] Figure 1 The figure illustrates a flow chart of a method for improving the accuracy of genomic selection for complex traits in pigs using large-scale genome annotation information according to an embodiment of the present application.

[0045] Figure 2 The diagram shows a schematic diagram comparing the genome selection accuracy of a method for improving the accuracy of genome selection for complex traits of pigs using large-scale genome annotation information according to an embodiment of the present application and an existing genome selection method.

[0046] Figure 3 A block diagram of an apparatus for improving the accuracy of genomic selection for complex traits in pigs using large-scale genome annotation information according to an embodiment of the present application is illustrated.

[0047] Figure 4 A block diagram of an electronic device according to an embodiment of the present application is illustrated. DETAILED DESCRIPTION

[0048] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0049] Application Overview

[0050] As mentioned above, genomic selection has achieved remarkable results in animal genetic breeding predictions, but it still faces many challenges in predicting complex traits in pigs. LD varies significantly between different groups or breeds, which means that a prediction model that is effective in a single group or target group may not be directly applicable to another group or groups. This population-specific LD pattern increases the complexity of cross-population predictions in pigs. Secondly, the interaction between environmental factors and partial genotypes also increases the difficulty of predicting complex traits using existing genomic selection-based models. For populations growing in different environments, the cross-environment evaluation results of a single genomic selection model may also be difficult to apply universally.

[0051] Therefore, designing a method that can process a large amount of functional annotation information to improve the accuracy of pig genomic selection, in order to improve the availability and real-time performance of breeding value prediction, is an important research direction in the field of genomic selection in pig breeding.

[0052] Based on the above technical problems, the basic concept of the present application is to propose a method for improving the accuracy of genomic selection for complex traits in pigs by using large-scale genomic annotation information, integrating multiple functional annotations of different categories on the pig genome, merging multiple functional annotations using the genomic kinship matrix of the functional annotations, and using the genomic kinship matrix of the functional annotation set obtained after the merger to calculate the random genetic effect value of the functional annotation set using a mixed linear model containing multiple random genetic effects, and adding the random genetic effect values ​​to provide fast and accurate support for the genomic selection of complex traits in pigs.

[0053] Specifically, Figure 1As shown, multiple functional annotation information on the reference group pig genome is integrated to obtain the position information of the functional annotation on the genome; at the same time, a whole genome association analysis is performed on the reference group pig genome to obtain the significant sites of the target traits of the pigs, and the significant sites of the target traits are merged into multiple functional annotation information as a type of functional annotation information; SNPs located in the functional annotation area on the reference group genome are extracted according to the position information of multiple functional annotations, and a functionally annotated genomic kinship matrix A is constructed based on SNPs; the variance components of the genomic kinship matrix A of each functional annotation are estimated, and multiple functional annotations are sorted according to the size of the variance component values; the interval between one ten-thousandth of the maximum variance component and the maximum variance component is taken as the merged interval A, the interval between one ten-thousandth of the maximum variance component and one ten-thousandth of the maximum variance component is taken as the merged interval B, and the interval between zero and one ten-thousandth of the maximum variance component is taken as the merged interval C, and the merged interval B includes the merged interval B 1 , B 2 , B 3 , whose interval endpoints have a geometric relationship in sequence; multiple functional annotations are merged based on the merged interval to obtain a functional annotation set; repeated SNPs in the functional annotation set are deleted to construct a genomic kinship matrix B, and the random genetic effect values ​​of the functional annotation set are calculated based on the genomic kinship matrix A. The genomic breeding values ​​of the target traits of pigs can be obtained by accumulating the random genetic effect values ​​of the functional annotation set.

[0054] In this way, since the contribution of a set of multiple functional annotations with similar variance components to the genetic effect of the target trait is taken into account in the process of obtaining the genomic breeding value of the target trait of pigs, the described method can process a large amount of functional annotation information, and good breeding prediction results can be achieved when facing different groups of pigs, making this optimized pig genomic selection method more accurate and practical.

[0055] After introducing the basic principles of the present application, various non-limiting embodiments of the present application will be described in detail with reference to the accompanying drawings.

[0056] Exemplary Methods

[0057] According to an embodiment of the present application, a method for improving the accuracy of genomic selection of complex traits in pigs by using large-scale genomic annotation information includes the following steps: S110, collecting multiple functional annotations and their position information of the genome of a reference group of pigs; S120, determining multiple functional annotation regions of the genome of a reference group of pigs based on the position information of the multiple functional annotations, extracting SNPs data of the multiple functional annotation regions and respectively constructing a genomic kinship matrix A for each of the multiple functional annotations; S130, estimating the variance components of the genomic kinship matrix A, sorting the multiple functional annotations based on the size of the variance components, and merging the multiple functional annotations to obtain a functional annotation set; and S140, deleting duplicate SNPs data in the functional annotation set and constructing a genomic kinship matrix B of the functional annotation set, and calculating the random genetic effect values ​​of the functional annotation set based on the genomic kinship matrix B using a mixed linear model containing multiple random genetic effects to obtain the genomic breeding values ​​of the target traits of the candidate group of pigs.

[0058] In step S110, in order to integrate and obtain effective functional annotation information, the selected reference group pig genome may include abundant pig samples and multiple target traits, and the genome functional annotation information to be integrated may come from the genome, including, for example, chromatin open regions, nucleosome-free regions, transcription factor footprint regions, transcription factor motifs, active promoter regions, and active enhancer regions. The types and locations of functional annotation information may not be limited. For the method for improving the accuracy of pig complex trait genome selection using large-scale genome annotation information according to the embodiment of the present application, any functional annotation on the reference group pig genome that may affect the target trait may be integrated and obtained, thereby increasing the breadth of the selected functional annotation information. In the case where complex traits in different groups have different genetic characteristics, the effects of more widely distributed annotation information in the genome on traits may be considered, rather than the effects of specific annotation information under specific conditions.

[0059] In one example, according to the method for improving the accuracy of genomic selection for complex traits of pigs using large-scale genomic annotation information according to an embodiment of the present application, a set of public pig growth trait data sets was selected as the reference group pig genome, and the pig growth trait data set included 2796 pig samples and 7 growth traits. According to the method for improving the accuracy of genomic selection for complex traits of pigs using large-scale genomic annotation information according to an embodiment of the present application, functional annotation information of 10 pig genomes was integrated and collected, including: chromatin open regions, nucleosome-free regions, transcription factor footprint regions, 3 transcription factor motifs, 2 active promoter regions and 2 active enhancer regions, so as to extract SNPs data using the 10 functional annotations.

[0060] Before determining multiple functional annotation regions of the reference group pig genome according to the position information of multiple functional annotations and extracting SNPs data, the method for improving the accuracy of pig complex trait genome selection using large-scale genome annotation information according to the embodiment of the present application also includes step S111: performing genome-wide association analysis on the reference group pig genome to obtain significant sites of the target trait of the reference group pig, and merging the significant sites of the target trait into multiple functional annotations as a type of functional annotation. Considering that the target trait information that may be contained in the multiple functional annotation information integrated by step S110 is not sufficient, further extracting significant sites of the target trait on the reference group pig genome as a type of supplementary functional annotation is a feasible supplementary solution.

[0061] In one example, a genome-wide association study (GWAS) was performed based on the genomic information of a reference group of pigs to obtain significant sites for the target trait. In the GWAS process, there are a variety of factors that may lead to false positives, such as population structure, kinship, etc. If these factors are not properly controlled, the reliability of the GWAS results will be seriously affected. The mixed linear model can effectively control the interference of these factors on the association analysis by considering both fixed effects and random effects, thereby improving the accuracy of GWAS. Therefore, according to the method for improving the accuracy of genomic selection of complex traits of pigs using large-scale genomic annotation information according to the embodiment of the present application, a mixed linear model is selected to obtain significant sites for the target trait to improve the robustness of functional annotation information.

[0062] In step S120, SNPs data within the functional annotation region is extracted on the reference group pig genome based on the location information of the functional annotation. Those skilled in the art will appreciate that once the functional annotation region on the reference genome is determined, a variety of sequencing and alignment technology tools can be used to extract the SNPs therein, and through data processing, such as quality control, a high-quality SNPs data set can be further extracted.

[0063] Next, the genome kinship matrix A can be constructed based on the SNPs data of multiple functional annotation information; in one example, the genome kinship matrix A can be obtained by the following formula:

[0064]

[0065] Among them, M k is the additive marker covariate matrix of the kth functional annotation among multiple functional annotations, with genotypes AA, AB, and BB coded as 0, 1, and 2, respectively, and p i is the minor allele frequency of site i in SNPs. The genetic similarity of pigs can be obtained by constructing the genomic kinship matrix A to obtain the random genetic effect of each functional annotation.

[0066] In step S130, according to the method for improving the accuracy of pig complex trait genome selection using large-scale genome annotation information according to the embodiment of the present application, multiple functional annotations are selectively merged based on the size of the variance component of each functional annotation genome kinship matrix A to obtain a functional annotation set with a smaller number. Therefore, step S130 can specifically include the following sub-steps:

[0067] Step S1301, for the target trait, fit a mixed linear model to obtain the variance components of each random genetic effect. Specifically, each functionally annotated genomic kinship matrix A is taken as a random genetic effect of the target trait, and the variance components of each random effect are estimated. Here, it can be understood by those skilled in the art that the phenotypic values ​​of the target traits of the reference group of pigs are used, that is, the random genetic effect values ​​are fitted with the known phenotypic values ​​of the complex traits. According to the method of using large-scale genomic annotation information to improve the accuracy of genomic selection of complex traits of pigs in an embodiment of the present application, a mixed linear model containing multiple random genetic effects is selected to calculate the random genetic effect values. The mixed linear model can be used to simultaneously consider the characteristics of multiple fixed effects and random effects to effectively control the interference of factors such as population structure and kinship on the prediction, thereby improving the accuracy of the analysis.

[0068] In one example, according to the method for improving the accuracy of genomic selection of complex traits in pigs using large-scale genomic annotation information according to an embodiment of the present application, a mixed linear model is fitted to the seven growth traits of the pig growth trait data set illustrated in steps S110 and S111, and the average information restricted maximum likelihood method can be used to estimate the variance components of the genomic kinship matrix A of each of the 11 functional annotations.

[0069] Step S1302, based on the variance component values, optimize the number of random genetic effects in the mixed linear model to reduce the dimension of the functional annotations in the model. Specifically, according to the method for improving the accuracy of genomic selection of complex traits in pigs using large-scale genomic annotation information according to the embodiment of the present application, multiple functional annotations can be sorted based on the size of the variance components, and the maximum variance component among the multiple variance components can be used to specify the merging method of the corresponding multiple functional annotations. In one example, a plurality of sub-intervals of different numbers are set in the interval composed of the variance component values ​​of all functional annotations by different interval division methods; specifically, the interval is divided into 2-5 sub-intervals of different numbers by using methods such as uniform partitioning, regular partitioning, recursive partitioning or probability distribution partitioning, and the influence of the interval division method and the number of sub-intervals on the accuracy of the estimation of genomic breeding values ​​on the functional annotation set obtained based on the sub-intervals is evaluated.

[0070] In this way, the optimal solution is obtained: for the overall variance component interval, the variance component interval from one tenth of the maximum variance component value to the maximum variance component value is used as the merged interval A; the variance component interval from one ten-thousandth of the maximum variance component value to one ten-thousandth of the maximum variance component value is used as the merged interval B; and the variance component interval from 0 to one ten-thousandth of the maximum variance component value is used as the merged interval C. After clustering and merging multiple functional annotations based on the merged intervals where the variance components of their genomic kinship matrix A are located, three functional annotation sets can be obtained, and the average accuracy of the genomic breeding values ​​output by the mixed linear model is the highest; and under this interval division method, the number of random genetic effects of the functional annotations that need to be fitted by the mixed linear model can be significantly reduced, and the model calculation efficiency is improved.

[0071] In one example, according to the method of improving the accuracy of genomic selection of complex traits of pigs using large-scale genomic annotation information in an embodiment of the present application, the merged interval B is further divided to test the influence of the functional annotation set obtained based on more merged intervals on the accuracy of genomic breeding value estimation and model performance; specifically, the partitioning method implemented on the merged interval B may include uniform partitioning, regular partitioning, recursive partitioning or probability distribution partitioning, etc., and the range of the intervals obtained by division can be 1-5. After testing, when the partitions of the merged interval B are 3 or more, the genomic breeding value obtained based on the current partition is better, and the more interval ranges, the more random genetic effect values ​​the mixed linear model needs to fit, and the longer the fitting time of the mixed linear model. Therefore, the number of partitions of the merged interval B can be set to preferably 3 to ensure the quality and fitting speed of the obtained functional annotation set at the same time.

[0072] More preferably, the combined interval B is partitioned based on multiple multiples of the series rule and the obtained functional annotation set is tested for its effect on the accuracy of the genome breeding value estimation. It is found that the three functional annotation sets obtained by partitioning the combined interval B based on the 10-fold rule partition have the highest average genome breeding value, that is, the combined interval B is obtained. 1 、Merge interval B 2 and merge interval B 3 :Among them, set the merge interval B 1 Set the merge interval B to be the variance component interval from 1% of the maximum variance component value to 1% of the maximum variance component value. 2 Set the merge interval B to be the variance component interval from one thousandth of the maximum variance component value to one hundredth of the maximum variance component value 3 is the variance component interval from one ten-thousandth of the maximum variance component value to one thousandth of the maximum variance component value, B 1 , B 2 and B 3The endpoints are in geometric relationship, the ratio is one tenth, and the interval range is reduced. Therefore, this pollen method can more effectively merge functional annotations with similar variance components and similar dispersion, and improve the data convergence of each functional annotation set obtained by merging.

[0073] In this way, by preferentially merging the functional annotations whose variance components of multiple intervals differ by less than ten times, the mixed linear model contains at most five functional annotation sets as new random genetic effects, achieving effective optimization of the functional annotation dimension. The optimization process is shown in the following formula:

[0074]

[0075] Among them, Set N is the N (N≤5) functional annotation set included in the optimized mixed linear model, is the estimate of the variance component of the functional annotation in each binned interval, is the maximum variance component estimate among all functional annotation variance components. Those skilled in the art will appreciate that the maximum variance component estimate can be used to estimate the variance component. It is appropriate to adjust the number of merged intervals to adjust the number of functional annotation sets, thereby further optimizing the performance of the mixed linear model. For example, according to the method for improving the accuracy of genomic selection of complex traits of pigs using large-scale genomic annotation information in an embodiment of the present application, three merged intervals, namely merged interval A, merged interval B and merged interval C, are set. In this way, fewer functional annotation sets can be obtained, further improving the convergence speed of solving the random genetic effect value of the mixed linear model to meet the needs of actual breeding work.

[0076] In step S140, in each optimized functional annotation set, the repeated SNPs data are deleted and the genomic kinship matrix B of each functional annotation set is reconstructed, and the genomic kinship matrix B of each functional annotation set is used as the optimized random genetic effect to refit the mixed linear model. Specifically, according to the method for improving the accuracy of genomic selection of complex traits of pigs using large-scale genomic annotation information according to an embodiment of the present application, multiple genomic kinship matrices B are used to solve the mixed linear model equation group to calculate the random genetic effect value of each functional annotation set. In one example, the solved mixed linear model is as follows:

[0077]

[0078] Among them, y is the phenotypic value vector of the target trait of the reference group pigs, b is the estimated value of the fixed effect, X is the design matrix corresponding to the fixed effect, μ k is the random genetic effect value of the kth functional annotation set among multiple functional annotation sets, which follows a normal distribution: Among them, Gk is the genomic kinship matrix B of the kth functional annotation set, is the variance component corresponding to the functional annotation set, Z k is the design matrix corresponding to the random effect, and e is the residual. By solving the mixed linear model equations containing multiple random genetic effects, we can obtain the random genetic effect values, the estimated values ​​of the fixed effects, and the estimated values ​​of the model residuals for multiple functional annotation sets.

[0079] After obtaining the random genetic effect values ​​of multiple functional annotation sets, the genomic breeding values ​​of the candidate group of pigs can be further obtained. In one example, according to the method for improving the accuracy of genomic selection of complex traits of pigs using large-scale genomic annotation information according to an embodiment of the present application, multiple random genetic effect values ​​can be accumulated and the summed result can be used as the estimated genomic breeding value of the candidate group of pigs.

[0080] In order to verify the selection accuracy of the method for improving the accuracy of genomic selection of complex traits of pigs using large-scale genomic annotation information according to the embodiment of the present application, the Pearson correlation coefficient value of the estimated genomic breeding value and the phenotypic value is used as a measure of the accuracy of genomic selection. Specifically, a conventional genomic selection model (hereinafter referred to as GBLUP) is selected to estimate the genomic estimated breeding values ​​of the seven growth traits contained in the reference group of pigs in the pig growth trait data set used in the example of step S110-step S140, and a genomic kinship matrix is ​​constructed using all SNPs data on the genome of the reference group of pigs, and the genomic breeding values ​​corresponding to the genomic kinship matrix are calculated by GBLUP, and the Pearson correlation coefficient of the estimated genomic breeding value and the phenotypic value is calculated.

[0081] Furthermore, the Pearson correlation coefficient between the genomic estimated breeding value and the phenotypic value of GBLUP was compared with the Pearson correlation coefficient between the genomic estimated breeding value and the phenotypic value obtained by the method for improving the accuracy of genomic selection of complex traits of pigs using large-scale genomic annotation information according to the embodiment of the present application, and the performance improvement effect of the method for improving the accuracy of genomic selection of complex traits of pigs using large-scale genomic annotation information according to the embodiment of the present application relative to the conventional model was calculated. The comparison results are shown in FIG. Figure 2 As shown, under the 7 growth traits of pigs in the reference group, model B is the genomic selection model used in the method for improving the accuracy of genomic selection for complex traits of pigs using large-scale genomic annotation information according to the embodiment of the present application, and its genomic selection accuracy is higher than the selection accuracy provided by model A. Model A is GBLUP, and the average accuracy under 7 growth traits is improved by 4.61%, which provides more accurate genomic estimated breeding values ​​than conventional models.

[0082] By setting the number of merged intervals and interval endpoints multiple times and comparing them with the GBLUP prediction performance, it is preferred that the number of merged intervals is five and the interval endpoints are in geometric proportion. In this way, five functional annotation sets can be obtained, and a better genomic estimated breeding value obtained by adding random genetic effect values ​​can be obtained by fitting a mixed linear model, thereby improving the accuracy of genomic selection.

[0083] According to the method for improving the accuracy of genomic selection for complex traits of pigs using large-scale genomic annotation information according to the embodiment of the present application, in practical applications, the accuracy of selection can be improved by integrating and utilizing genomic functional annotation information, accelerating the genetic progress of target traits and shortening the breeding cycle. The application of the method mainly includes the following steps: extracting SNPs in each annotation region based on reference group data and pig genome annotation position information, constructing a genomic kinship matrix of each functional annotation using SNPs data, merging functional annotations with similar variance component values ​​into a functional annotation set, and using a genomic selection model containing an appropriate number of random genetic effects to predict the genomic breeding values ​​of target traits of individual pigs. In this way, a comprehensive selection index can be further constructed through the genomic breeding values ​​of multiple complex target traits, and pig individuals can be ranked according to the comprehensive selection index to provide a list of candidate individuals for selection. Therefore, the present application provides a method for genomic selection of complex traits of pigs with a wide range of applications and higher accuracy.

[0084] Exemplary Devices

[0085] Figure 3 A block diagram of an apparatus for improving the accuracy of genomic selection for complex traits in pigs using large-scale genome annotation information according to an embodiment of the present application is illustrated.

[0086] like Figure 3 As shown, according to an embodiment of the present application, the device 200 for improving the accuracy of genomic selection of complex traits in pigs using large-scale genome annotation information includes: a functional annotation collection unit 210, which collects multiple functional annotations and their position information of the genome of the reference group pigs; a kinship construction unit 220, which determines multiple functional annotation regions of the genome of the reference group pigs according to the position information of the multiple functional annotations, extracts SNPs data of the multiple functional annotation regions and respectively constructs a genomic kinship matrix A for each functional annotation in the multiple functional annotations; a functional annotation integration unit 230, which estimates the variance components of the genomic kinship matrix A, sorts the multiple functional annotations based on the size of the variance components, and merges the multiple functional annotations to obtain a functional annotation set; and an output unit 240, which deletes duplicate SNPs data in the functional annotation set and constructs a genomic kinship matrix B of the functional annotation set, and calculates the random genetic effect value of the functional annotation set based on the genomic kinship matrix B to obtain the genomic breeding value of the target trait of the candidate group pigs.

[0087] In such Figure 3 On the basis of the shown embodiment, the device 200 for improving the accuracy of genomic selection of complex traits in pigs using large-scale genomic annotation information according to the embodiment of the present application further includes: a functional annotation supplement unit 250, which is used to perform a whole-genome association analysis on the genome of a reference group of pigs to obtain significant sites of the target trait, and treat the significant sites of the target trait as a type of functional annotation and merge them into multiple functional annotations of the functional annotation extraction unit.

[0088] Here, those skilled in the art can understand that the specific functions and operations of each unit in the above-mentioned device 200 for improving the accuracy of pig complex trait genome selection using large-scale genome annotation information have been described in the above reference. Figure 1 The description of the method of using large-scale genome annotation information to improve the accuracy of genomic selection for complex traits in pigs has been introduced in detail, and therefore, its repeated description will be omitted.

[0089] As described above, the device 200 for improving the accuracy of genomic selection of complex traits of pigs using large-scale genomic annotation information according to the embodiment of the present application can be implemented in various terminal devices, such as servers for performing object recognition tasks, etc. In one example, the device 200 for improving the accuracy of genomic selection of complex traits of pigs using large-scale genomic annotation information according to the embodiment of the present application can be integrated into the terminal device as a software module and / or a hardware module. For example, the device 200 for improving the accuracy of genomic selection of complex traits of pigs using large-scale genomic annotation information can be a software module in the operating system of the terminal device, or it can be an application developed for the terminal device; of course, the device 200 for improving the accuracy of genomic selection of complex traits of pigs using large-scale genomic annotation information can also be one of the many hardware modules of the terminal device.

[0090] Alternatively, in another example, the apparatus 200 for improving the accuracy of pig complex trait genome selection using large-scale genome annotation information and the terminal device may also be separate devices, and the apparatus 200 for improving the accuracy of pig complex trait genome selection using large-scale genome annotation information may be connected to the terminal device via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0091] Exemplary Electronic Devices

[0092] Below, reference Figure 4 To describe an electronic device according to an embodiment of the present application.

[0093] Figure 4 A block diagram of an electronic device according to an embodiment of the present application is illustrated.

[0094] like Figure 4 As shown, the electronic device 10 includes one or more processors 11 and a memory 12 .

[0095] The processor 13 may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device 10 to perform desired functions.

[0096] The memory 12 may include one or more computer program products, and the computer program product may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may, for example, include random access memory (RAM) and / or cache memory (cache), etc. The non-volatile memory may, for example, include read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 11 may run the program instructions to implement the method for improving the accuracy of pig complex trait genome selection using large-scale genome annotation information in each embodiment of the present application described above and / or other desired functions. Various contents such as genome kinship matrix, functional annotation set, etc. may also be stored in the computer-readable storage medium.

[0097] In one example, the electronic device 10 may further include: an input device 13 and an output device 14, and these components are interconnected via a bus system and / or other forms of connection mechanisms (not shown).

[0098] The input device 13 may include, for example, a keyboard, a mouse, etc.

[0099] The output device 14 can output various information to the outside, including the target pig genome breeding value output by the mixed linear model, etc. The output device 14 can include, for example, a display, a speaker, a printer, a communication network and a remote output device connected thereto, etc.

[0100] Of course, to simplify, Figure 4 Only some of the components related to the present application in the electronic device 10 are shown, and components such as a bus, an input / output interface, etc. are omitted. In addition, according to specific application situations, the electronic device 10 may also include any other appropriate components.

[0101] Exemplary computer program products and computer-readable storage media

[0102] In addition to the above-mentioned methods and devices, an embodiment of the present application may also be a computer program product, which includes computer program instructions, which, when executed by a processor, enable the processor to execute the steps of the method for improving the accuracy of genomic selection of complex traits of pigs using large-scale genomic annotation information according to various embodiments of the present application as described in the above "Exemplary Methods" section of this specification.

[0103] The computer program product may be written in any combination of one or more programming languages ​​to write program codes for performing the operations of the embodiments of the present application, including object-oriented programming languages, such as Java, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as an independent software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0104] In addition, an embodiment of the present application may also be a computer-readable storage medium having computer program instructions stored thereon, which, when executed by a processor, causes the processor to execute the steps of the method for improving the accuracy of genomic selection of complex traits of pigs using large-scale genomic annotation information according to various embodiments of the present application as described in the above “Exemplary Method” section of this specification.

[0105] The computer readable storage medium can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can include, for example, but is not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0106] The basic principles of the present application are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present application are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present application. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, not for limitation, and the above details do not limit the present application to being implemented by adopting the above specific details.

[0107] The block diagrams of the devices, apparatuses, equipment, and systems involved in this application are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagram. As will be appreciated by those skilled in the art, these devices, apparatuses, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including", "comprising", "having", etc. are open words, referring to "including but not limited to", and can be used interchangeably with them. The words "or" and "and" used here refer to the words "and / or" and can be used interchangeably with them, unless the context clearly indicates otherwise. The words "such as" used here refer to the phrase "such as but not limited to", and can be used interchangeably with them.

[0108] It should also be noted that in the methods, devices and apparatuses of the present application, each component or each step can be decomposed and / or recombined, and these decompositions and / or recombinations should be regarded as equivalent solutions of the present application.

[0109] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present application. Therefore, the present application is not intended to be limited to the aspects shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

[0110] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present application to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.

Claims

1. Methods for improving the accuracy of genomic selection for complex traits in pigs using large-scale genome annotation information, including: Collect multiple functional annotations and position information of the reference pig genome; Determine multiple functional annotation regions of the reference group pig genome according to the position information of the multiple functional annotations, extract SNPs data of the multiple functional annotation regions and construct a genome kinship matrix A of each functional annotation in the multiple functional annotations respectively; estimating variance components of the genome kinship matrix A, sorting the multiple functional annotations based on the sizes of the variance components, and merging the multiple functional annotations to obtain a functional annotation set; The repeated SNPs data in the functional annotation set are deleted and a genomic kinship matrix B of the functional annotation set is constructed. Based on the genomic kinship matrix B, a mixed linear model containing multiple random genetic effects is used to calculate the random genetic effect values ​​of the functional annotation set to obtain the estimated genomic breeding values ​​of the target traits of the candidate group pigs.

2. The method for improving the accuracy of pig complex trait genome selection using large-scale genome annotation information according to claim 1, wherein: Determining multiple functional annotation regions of the reference group pig genome according to the position information of the multiple functional annotations includes: A genome-wide association analysis is performed on the genome of the reference group of pigs to obtain significant sites of the target trait of the reference group of pigs, and the significant sites of the target trait are merged into the multiple functional annotations as a type of functional annotation.

3. The method for improving the accuracy of pig complex trait genome selection using large-scale genome annotation information according to claim 1, wherein: The merging of the plurality of functional annotations to obtain a functional annotation set comprises: determining a combined interval based on a maximum variance component among the variance components; The functional annotations are merged in response to the variance component being within the merge interval to obtain a functional annotation set.

4. The method for improving the accuracy of pig complex trait genome selection using large-scale genome annotation information according to claim 3, wherein: Determining the merging interval based on the maximum variance component among the variance components comprises: The variance component interval between one tenth of the maximum variance component and the maximum variance component is taken as the combined interval A; The variance component interval between one ten-thousandth of the maximum variance component and one tenth of the maximum variance component is taken as the combined interval B; and The variance component interval between zero and one ten-thousandth of the maximum variance component is taken as the combined interval C.

5. The method for improving the accuracy of pig complex trait genome selection using large-scale genome annotation information according to claim 4, wherein: The merging interval B includes merging interval B1, merging interval B2 and merging interval B3; The interval endpoints of the merged interval B, the merged interval B2 and the merged interval B3 are in a geometric relationship in sequence.

6. The method for improving the accuracy of pig complex trait genome selection using large-scale genome annotation information according to claim 1, wherein: The method of calculating the random genetic effect value of the functional annotation set based on the genomic kinship matrix B using a mixed linear model containing multiple random genetic effects to obtain the genomic breeding value of the target trait of the candidate group pigs includes: The random genetic effect values ​​of the functional annotation set are accumulated to serve as the genomic breeding value of the target trait of the candidate group of pigs.

7. A device for improving the accuracy of genomic selection for complex traits in pigs using large-scale genomic annotation information, comprising: Functional annotation collection unit, which collects multiple functional annotations and their position information of the reference group pig genome; A kinship construction unit is configured to determine a plurality of functional annotation regions of the reference group pig genome according to the position information of the plurality of functional annotations, extract SNPs data of the plurality of functional annotation regions, and respectively construct a genomic kinship matrix A of each of the plurality of functional annotations; A functional annotation integration unit estimates the variance components of the genome kinship matrix A, sorts the multiple functional annotations based on the size of the variance components, and merges the multiple functional annotations to obtain a functional annotation set; The output unit deletes the repeated SNPs data in the functional annotation set and constructs a genomic kinship matrix B of the functional annotation set, and based on the genomic kinship matrix B, a mixed linear model containing multiple random genetic effects is used to calculate the random genetic effect value of the functional annotation set to obtain the genomic breeding value of the target trait of the candidate group pigs.

8. The device for improving the accuracy of pig complex trait genome selection using large-scale genome annotation information according to claim 7, further comprising: The functional annotation supplement unit performs genome-wide association analysis on the reference group pig genome to obtain significant sites of the target trait of the reference group pigs, and merges the significant sites of the target trait into the multiple functional annotations as a type of functional annotation.

9. An electronic device, comprising: processor; as well as A memory, in which computer program instructions are stored, and when the computer program instructions are executed by the processor, the processor executes the method for improving the accuracy of pig complex trait genome selection using large-scale genome annotation information according to any one of claims 1-6.

Images