A high-precision genotype reference panel of cervus nippon and a filling method thereof
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- INST OF SPECIAL ANIMAL & PLANT SCI OF CAAS
- Filing Date
- 2026-01-26
- Publication Date
- 2026-06-05
AI Technical Summary
The accuracy of genotyping for sika deer is low, the filling effect of low-frequency variant sites is poor, and there is a lack of high-quality genotyping reference panels suitable for sika deer, which affects the accuracy and cost-effectiveness of sika deer breeding.
A high-precision genotypic reference panel for sika deer was constructed by collecting blood samples from sika deer and performing 5X resequencing. The panel was self-filled using Beagle software and quality controlled using Plink software. The DR2 filtering threshold was set to 0.95, and the data was filled according to the minor allele frequency range to construct a genotypic data reference panel containing 45,000,000 to 55,000,000 SNP loci.
It significantly improved the genotyping effect of sika deer, with a consistency rate of 92.05%, especially at low-frequency variant sites (MAF<0.1) where the consistency rate reached 95.13%, reducing sequencing costs while maintaining high-precision genotyping capabilities and good population stability.
Smart Images
Figure CN122157776A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of animal genomics technology, and specifically relates to a high-precision genotype reference panel for sika deer and its filling method. Background Technology
[0002] Sika deer (Cervus nippon), a distinctive economic animal in my country, possesses products such as antlers and placenta with extremely high medicinal value in traditional Chinese medicine and shows great development potential in the modern health industry. The Chinese Pharmacopoeia records that antlers have the effects of "strengthening kidney yang, fortifying muscles and bones, regulating the Chong and Ren meridians, and treating carbuncles and boils." With the official inclusion of sika deer in the National Livestock and Poultry Genetic Resources Catalogue in 2020 and the continuous advancement of related industrial development policies, sika deer farming is rapidly developing towards large-scale and intensive operations. However, current sika deer genetic breeding work still faces many challenges, among which efficiently obtaining population genetic information is key to breaking through traditional breeding bottlenecks and achieving high-quality industrial development.
[0003] Whole-genome resequencing technology can comprehensively and accurately detect genomic variations, providing rich genetic information for sika deer breeding research. However, its high cost limits its application in large-scale population studies and routine breeding practices. While liquid-phase microarrays can achieve high-throughput genotyping, they are often limited by probe design, resulting in the loss of many low-frequency variant sites and decreased accuracy in breeding value estimation. In contrast, genotype imputation technology, by utilizing high-density genotype data from a reference population and combining it with low-density marker information from the target population, can infer the genotypes of undetected sites in the target population, thus obtaining near-whole-genome-level genetic data at a lower cost. This technology has achieved significant results in the field of livestock breeding. In Mianyang breeding, when using genotype imputation technology and combining it with imputation data for ssBLUP prediction, the imputation accuracy increased from 0.8875 to 0.9852 as the size and density of the reference panel increased. The accuracy of milk yield prediction by ssBLUP after imputation improved from 0.61 to 0.66 compared to BLUP. In beef cattle breeding, the correlation between the data imputed by QUILT genotype imputation software and the microarray data reached a maximum of 0.97. The accuracy of GS was slightly lower than that of low-depth data. Genotype imputation not only significantly reduced the cost of genotyping but also significantly improved the accuracy and efficiency of genome selection, accelerating the breeding process of superior breeds.
[0004] However, in the field of sika deer research, the application of genotyping technology is still in its early stages. On the one hand, the lack of high-quality genotypic reference panels suitable for sika deer limits the accuracy and reliability of genotyping; on the other hand, the applicability and effectiveness of different genotyping methods in sika deer populations have not yet been systematically evaluated. Constructing a genotypic reference panel specifically for sika deer and verifying its genotyping effectiveness is of great significance for reducing the cost of sika deer genetic research and promoting the application of molecular breeding technology in the sika deer industry. Summary of the Invention
[0005] The purpose of this invention is to provide a high-precision genotype reference panel for sika deer and its filling method, thereby solving the technical problems of low accuracy in filling sika deer genotypes and poor filling effect of low-frequency variant sites in the prior art.
[0006] To achieve the above-mentioned objectives, this invention provides a method for filling a high-precision genotype reference panel for sika deer, comprising the following steps: S1. Genotyping data of the original resequencing population: Blood samples were collected from sika deer aged 24 months and older, and 5X resequencing was performed to obtain genotyping data. S2. Use Beagle software to autofill the resequencing data and generate a reference panel; S3. Use Plink software to perform quality control on the reference panel data, including: (1) Individuals with a genotype deletion rate greater than 10% were excluded; (2) Exclude sites with a minor allele frequency of less than 0.05; (3) Remove Hardy-Weinberg equilibrium test p-values less than 1×10 -6 The site; S4, Set DR 2 The filtering threshold is 0.95, and the chip data of the target group is filtered. S5. Divide the loci into five intervals according to the minor allele frequency: [0,0.1), [0.1,0.2), [0.2,0.3), [0.3,0.4), and [0.4,0.5]. Fill the intervals in order.
[0007] Furthermore, in step S1, the number of blood samples collected from 298 sika deer was 45 of which were simultaneously subjected to 40K breeding chip sequencing and genotyping as a validation population.
[0008] Furthermore, in step S4, DR 2 When the filtering threshold is 0.95, the fill consistency reaches its peak.
[0009] Furthermore, in step S5, when filling in intervals in sequence, the sites in the interval [0, 0.1) are filled first, and then the sites in other intervals are filled in sequence.
[0010] Furthermore, the reference population size in the method is 298 individuals.
[0011] Furthermore, the method also includes using vcftools software to separate the validation population from the original resequencing population.
[0012] This invention also provides a sika deer genotyping reference panel, which contains genotyping data for 45,000,000 to 55,000,000 SNP loci, wherein the SNP loci meet the following quality control criteria: genotype deletion rate less than 10%, minor allele frequency greater than or equal to 0.05, and Hardy-Weinberg equilibrium test p-value greater than or equal to 1 × 10⁻⁶. -6 Furthermore, the proportion of SNP sites with minor allele frequencies in the [0, 0.1) range in the reference panel is greater than 60%.
[0013] The beneficial effects of this invention are: This invention sets DR 2 A threshold of 0.95 significantly improved the genotyping effect of sika deer, achieving a consistency of 92.05%. Based on the MAF distribution characteristics of the sika deer population, a strategy of filling in MAF intervals was established, achieving a consistency of 95.13% for low MAF (MAF < 0.1) sites. An optimal reference population size of 298 individuals was determined; further increasing the reference population only improved the consistency by 0.27%, effectively controlling costs. The sika deer-specific genotyping reference panel constructed in this invention exhibits population stability, with a consistency difference of <2% between different age groups. It provides a low-cost and efficient genotyping scheme for sika deer genome breeding, significantly reducing sequencing costs while maintaining high-precision genotyping capabilities. Attached Figure Description
[0014] Figure 1 The filled-in results are for three different age groups (5 years old, 4 years old, and 1 year old) D, E, and H. Detailed Implementation
[0015] To enable those skilled in the art to better understand the technical solutions of this invention, the present application will be further described in detail below with reference to embodiments. Example
[0016] 1. Experimental animal and genotype data All experimental data came from 343 male sika deer (24 months and older) at a deer breeding farm in Changchun City, Jilin Province. Whole blood samples were collected from the sika deer using the jugular vein puncture method and sent to Borui Di Company for 5X resequencing. Among them, 45 samples were additionally genotyped using the 40K SNP chip for sika deer breeding independently developed by our research group. These 45 samples have both resequencing data and chip sequencing data.
[0017] 2. Genotype Data Processing (1) Quality control Plink (V 1.90) was used for quality control of the resequencing data. Filtering criteria included: 1) removing individuals with a genotype deletion rate > 10%; 2) excluding loci with a minor allele frequency (MAF) < 0.05; and 3) removing loci with significant deviations from the Hardy-Weinberg equilibrium test threshold (P < 1 × 10⁻⁶). -6 The site.
[0018] After quality control, the final resequencing data was reduced from 66,656,955 SNP loci to 49,662,047 SNP loci, including 49,997,034 autosomal SNP loci. Meanwhile, the microarray data was reduced from 290,703 SNP loci to 192,819 loci.
[0019] (2) Reference panel and target group construction The resequencing data was self-filled using the default parameters of Beagle (V 5.4) software to create a reference panel. The VCF file of the microarray sequencing genotyping (n = 45) data, after quality control, was used as the target population for subsequent validation. Individuals with the same ID (n = 45) were separated from the original resequencing population using vcftools (V0.1.16) to form the validation population. The remaining individuals (n = 298) formed the reference population, and a genotype-filled reference panel for sika deer was constructed.
[0020] 3. Evaluation of the fill effect of the reference panel (1) Different DR 2 The influence of value To further investigate the impact of chain imbalance on filling effect, different DR settings were used in Plink software. 2 The impact of filter thresholds (0, 0.2, 0.4, 0.6, 0.8, 0.95) on filling accuracy was analyzed. DR 2 The value represents the degree of linkage disequilibrium between loci; a higher DR value indicates a higher degree of linkage disequilibrium. 2 A higher value indicates a stronger linkage between loci.
[0021] (2) The impact of different MAF intervals To investigate the effect of gene frequency on the genotyping effect of sika deer, the loci were divided into different intervals according to MAF ([0, 0.1), [0.1, 0.2), [0.2, 0.3), [0.3, 0.4), [0.4, 0.5]) using Plink software, and the filling accuracy of loci within each interval was analyzed.
[0022] (3) The impact of different target groups The accuracy of filling was statistically analyzed for three different age groups (5 years, 4 years, and 1 year) in D, E, and H, respectively, and the impact of group differences on the filling effect was analyzed.
[0023] (4) The influence of different chromosomes The accuracy of filling each chromosome was statistically analyzed to identify differences between chromosomes.
[0024] (7) Ratio of reference group to target group From the 298 resequencing cases, 270, 225, 180, 135, 90, and 45 samples were randomly selected to construct combinations of reference and target population ratios of 6:1, 5:1, 4:1, 3:1, 2:1, and 1:1, respectively, and the impact of the ratio on the filling accuracy was analyzed.
[0025] 4. Evaluation Indicators The concordance rate (CR) was used to measure the filling effect, and was defined as the ratio of the number of correctly filled sites in the validation population to the total number of filled sites.
[0026] 5. Results Analysis (1) Different DR 2 Impact on filling effect Different DR 2 The specific results of the effect of filter values (0, 0.2, 0.4, 0.6, 0.8, and 0.95) on the fill are shown in Table 1. As can be seen from the table, with DR... 2 As the threshold gradually increases, the number of remaining sites after filtering shows a decreasing trend. Without filtering (DR... 2 = 0) retained nearly 45 million sites, while when the threshold was increased to 0.95, only about 795 k sites were retained (a reduction of 98.23%). The consistency rate showed a pattern of first decreasing and then increasing. Without filtering, the consistency was 88.93%, decreasing to 81.53% at the intermediate threshold (0.2 - 0.6), and further decreasing at higher thresholds (DR). 2 When the coefficient of variation (= 0.95) was reached, the consistency rate significantly rebounded to 92.05%.
[0027] Table 1 Different DR 2 Impact on the filling effect of sika deer
[0028] Note: DR 2 This sets the threshold for chip filtering, such as DR. 2 = 0 indicates no filtering.
[0029] (2) The effect of different MAF intervals on the filling effect Table 2 shows the consistency of sika deer filling in different MAF intervals. The results show that the consistency is 95.13%, 77.02%, 66.50%, 59.70%, and 56.10% in the five MAF frequency gradients from [0, 0.1) to [0.4, 0.5), respectively, and the number of sites decreases sharply from 33.82 million to 1.45 million.
[0030] Table 2. Effects of different MAFs on the filling effect of sika deer.
[0031] (3) The impact of different groups on the filling effect The results of filling in different groups revealed ( Figure 1 The consistency of the three groups D, E, and H remained basically unchanged, at 88.51%, 88.85%, and 89.59%, respectively.
[0032] (4) The effect of different chromosomes on filling effect Table 3. Influence of different chromosomes of sika deer on the filling effect
[0033] The effects of different chromosomes on the filling effect are shown in Table 3. The filling consistency of the 33 chromosomes ranged from 87.86% to 89.89%. Among them, chromosome 4 had the highest filling consistency, reaching 89.89%. The filling consistency of chromosome 33 (sex chromosome) was relatively low, at 87.86%. The filling consistency of chromosomes 2, 3, 4, 10, and 12 all exceeded 89%. The filling accuracy of chromosomes 1, 6, and 7 ranged from 88.5% to 89%.
[0034] (5) The impact of the ratio of reference group to target group on the filling effect The effects of the ratio of the reference panel to the target population on the genotypic consistency of sika deer are shown in Table 4. As the ratio of the reference population to the target population increased from 1:1 to 6:1, the consistency of filling improved from 88.68% to 88.98%, while the number of loci remained unchanged.
[0035] Table 4. Effect of the ratio of reference population to target population on sika deer genotype filling.
[0036] In summary, the method of this invention fills 40K chip data to near the resequencing density level, achieving an average filling consistency of 89%. Particularly in filling low-frequency variant sites (MAF < 0.1), the filling consistency reaches 95.13%, significantly outperforming existing technologies. Furthermore, this invention determines an optimal reference population size of 298 heads, avoiding the resource waste caused by blindly expanding the reference population size.
Claims
1. A method for filling a high-precision genotype reference panel for sika deer, characterized in that, Includes the following steps: S1. Genotyping data of the original resequencing population: Blood samples were collected from sika deer aged 24 months and older, and 5X resequencing was performed to obtain genotyping data. S2. Use Beagle software to autofill the resequencing data and generate a reference panel; S3. Use Plink software to perform quality control on the reference panel data, including: (1) Individuals with a genotype deletion rate greater than 10% were excluded; (2) Exclude sites with a minor allele frequency of less than 0.05; (3) Remove Hardy-Weinberg equilibrium test p-values less than 1×10 -6 The site; S4, Set DR 2 The filtering threshold is 0.95, and the chip data of the target group is filtered. S5. Divide the loci into five intervals according to the minor allele frequency: [0,0.1), [0.1,0.2), [0.2,0.3), [0.3,0.4), and [0.4,0.5]. Fill the intervals in order.
2. The method for filling the high-precision genotype reference panel for sika deer according to claim 1, characterized in that, In step S1, 298 blood samples were collected from sika deer, of which 45 samples were simultaneously subjected to 40K breeding chip sequencing and genotyping as a validation population.
3. The method for filling the high-precision genotype reference panel for sika deer according to claim 1, characterized in that, In step S4, DR 2 When the filtering threshold is 0.95, the fill consistency reaches its peak.
4. The method for filling the high-precision genotype reference panel for sika deer according to claim 1, characterized in that, In step S5, when filling in interval order, the sites in the interval [0, 0.1) are filled first, and then the sites in other intervals are filled in sequence.
5. The method for filling the high-precision genotype reference panel for sika deer according to claim 1, characterized in that, The reference population size in the method is 298 individuals.
6. The method for filling the high-precision genotype reference panel for sika deer according to claim 2, characterized in that, The method also includes using vcftools software to separate the validation population from the original resequencing population.
7. A sika deer genotype reference panel, characterized in that, The reference panel contains genotypic data for 45,000,000 to 55,000,000 SNP loci, which meet the following quality control criteria: genotype deletion rate less than 10%, minor allele frequency greater than or equal to 0.05, and Hardy-Weinberg equilibrium p-value greater than or equal to 1 × 10⁻⁶. -6 Furthermore, the proportion of SNP sites with minor allele frequencies in the [0, 0.1) range in the reference panel is greater than 60%.