Method for identifying pigeon germplasm resources

Abstract

The invention discloses a method for identifying pigeon germplasm resources. Whether a pigeon individual belongs to one of a white Carneau pigeon, a white feather king pigeon, a mima pigeon, a silver feather king pigeon, a Taishen pigeon, a red Carneau pigeon and a Huangkneau pigeon or not is identified according to 326 SNP sites. According to the method, the pigeons of seven groups can be identified by using relatively few SNP loci, the time consumption is short, the cost is low, the accuracy is high, and the method is beneficial to popularization and application in grass-roots institutions and daily supervision.

Description

Technical Field

[0001] This invention relates to the field of biological identification technology, specifically to a method for identifying pigeon germplasm resources based on SNP site molecular marker combinations. Background Technology

[0002] my country's pigeon industry is expanding rapidly and playing an increasingly important role in animal husbandry. However, in stark contrast to this rapid development, the management and identification of pigeon germplasm resources have long relied on traditional methods. Currently, the registration and market circulation of pigeon breeds are mainly based on experiential judgments of appearance (such as feather color and body shape) and empirical pedigree records, lacking unified, objective, and verifiable molecular standards. This deficiency is becoming increasingly apparent in actual production, especially since many white-feathered breeds with excellent economic traits are highly similar in appearance, making accurate differentiation by the naked eye difficult. Furthermore, frequent crossbreeding, cross-regional introductions, and non-standard breed labeling can easily lead to problems such as mixed use of breed labels, missing pedigree information, and poor genetic consistency between batches. In the core flock construction and breeding stages, insufficient verification of bloodlines and monitoring of inbreeding levels further increase the risk of genetic degradation of strains and maintaining the genetic health of the core flock. The aforementioned factors not only trigger registration and transaction disputes and increase regulatory and compliance costs, but also severely weaken the efficiency of brand building and price discovery of superior strains, ultimately hindering the promotion of superior germplasm resources and the progress of genetic improvement.

[0003] Molecular marker technology, especially single nucleotide polymorphism (SNP) markers, has been widely used in breed identification research due to its significant advantages such as abundant and widely distributed sites, high genetic stability, strong representativeness, and convenient and rapid detection. SNPs refer to variations in a single nucleotide in the genome and have become the preferred tool for constructing molecular identification systems. However, current molecular techniques applied to pigeon germplasm resource identification still have limitations. Existing high-throughput SNP genotyping chips are often designed based on specific pigeon breeds or reference genomes, resulting in a lack of representativeness and specificity. More importantly, these high-throughput chips typically contain tens or even hundreds of thousands of marker sites. Although they can provide comprehensive genetic information, their detection costs are high, computationally complex, and time-consuming, making it difficult to promote and rapidly apply them on a large scale in grassroots institutions and daily market supervision. Summary of the Invention

[0004] The purpose of this invention is to provide a method for identifying pigeon germplasm resources. This method can identify seven groups of pigeons using relatively few SNP loci. It is time-saving, low-cost, and highly accurate, which is conducive to its promotion and application in grassroots institutions and daily supervision.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A method for identifying pigeon germplasm resources, which identifies whether an individual pigeon belongs to one of seven pigeon groups based on 326 SNP loci. The seven pigeon groups are White Cano pigeon, White King pigeon, Mimas pigeon, Silver King pigeon, Tai Shen pigeon, Red Cano pigeon, and Yellow Cano pigeon. The information of the 326 SNP loci is shown in Table 1. The chromosomes and the positions of the SNP loci on the chromosomes in Table 1 are determined based on the pigeon GCA_001887795.1 reference genome.

[0007] Preferably, the above method for identifying pigeon breeds includes the following steps:

[0008] Step 1: Obtain reference set data, which consists of the genotypes of the 326 SNP loci of known pigeon samples from seven groups: White Cano pigeons, White King pigeons, Mimas pigeons, Silver King pigeons, Tai Shen pigeons, Red Cano pigeons, and Yellow Cano pigeons. Each group contains several pigeon samples.

[0009] Step 2: Extract genomic DNA from the pigeons to be tested and obtain the genotypes of the 326 SNP loci in the genomic DNA of the pigeons to be tested;

[0010] Step 3: Train the support vector machine model using the reference set data, then input the genotypes of the 326 SNP loci in the genomic DNA of the pigeon individual to be tested into the optimized support vector machine model, and determine the germplasm resources of the pigeon individual to be tested based on the output of the support vector machine model.

[0011] Preferably, in step 1, each group contains 10 pigeon samples; in step 3, the support vector machine model is trained using the reference set data in a Python 3.12.2 environment, and the optimal hyperparameters are determined through grid search and cross-validation. The optimal parameters are the penalty parameter C=1 and the kernel function parameter γ=0.001; then, the genotypes of the 326 SNP loci in the genomic DNA of the pigeon individual to be tested are input into the optimized support vector machine model. The output of the support vector machine model is the most likely breed label, the probability or confidence of each breed, and the germplasm resources of the pigeon individual to be tested are determined based on the output.

[0012] The present invention has the following beneficial effects: (1) The genotype data of the above 326 SNP loci can accurately identify whether the individual pigeon to be tested belongs to one of the following: White Cano pigeon, White King pigeon, Mimas pigeon, Silver King pigeon, Tai Shen pigeon, Red Cano pigeon or Yellow Cano pigeon. The method is simple and easy to operate, and is easy to promote and popularize; (2) A molecular marker combination of pigeon germplasm resources is established using 326 SNP loci, which can help to understand and analyze the specificity of pigeon breeds more deeply, which is conducive to the construction of the core pigeon population and the smooth and high-quality development of breeding pigeons; (3) It helps to reduce disputes in pigeon market registration and trading, reduce regulatory and compliance costs, effectively improve the efficiency of brand building and price discovery of superior strains, and ultimately benefit the promotion and genetic improvement of superior germplasm resources. Attached Figure Description

[0013] Figure 1 This is a flowchart of the screening process for 326 SNP sites in Embodiment 1 of the present invention;

[0014] Figure 2 This is a distribution map of the 326 SNP sites in the chromosome in Example 1 of the present invention;

[0015] Figure 3 This is a confusion matrix diagram for the identification of 70 pigeon germplasm resources in Example 1 of the present invention;

[0016] Figure 4 This is a diagram showing the results of phylogenetic analysis of 70 pigeons based on 326 SNP loci in Embodiment 1 of the present invention.

[0017] Figure 5 This is a diagram showing the results of the identification of germplasm resources of 46 pigeons in Example 2 of the present invention. Detailed Implementation

[0018] The present invention will be further described below through specific embodiments:

[0019] Example 1: Screening and Validation of SNP Molecular Marker Combinations for Identifying Pigeon Germplasm Resources

[0020] (1) Sequencing data analysis

[0021] Venous blood was collected from 10 White Cano pigeons, 10 White King pigeons, 10 Mimas pigeons, 10 Silver King pigeons, 10 Tai Shen pigeons, 10 Red Cano pigeons, and 10 Yellow Cano pigeons. DNA was extracted and genome sequencing was performed, with sequencing depths greater than 10×. The sequencing data were filtered using FASTP software and aligned to the pigeon reference genome (GCA_001887795.1) using BWA software. Variation detection and quality control were performed using GATK and Plink software. The filtering and quality control criteria were: filtering out variations with a QD lower than 2; filtering out variations with a FS greater than 60; filtering out variations with an MQ less than 40; retaining only loci on chromosomes 1-15, 4A, and 17-28, loci with a minimum allele frequency greater than 0.01, and a 100% detection rate. After these filtering and quality control steps, a VCF file containing 5,796,462 loci from 70 samples was generated.

[0022] (2) Site screening

[0023] This invention designs and implements a multi-stage, high-precision SNP locus screening process integrating population genetics and machine learning algorithms. The aim is to screen for the smallest core locus combinations for precise identification of pigeon germplasm resources at the whole-genome scale. Specifically, the process involves: First, using Plink software to remove redundancy using linkage disequilibrium LD (standard: window 50Kb, step size 10Kb, removing r...). 2 After identifying loci with an importance greater than 0.1, 237,617 loci remained. Then, pairwise FST analysis was performed on seven pigeon populations: White Cano, White King, Mimas, Silver King, Taishen, Red Cano, and Yellow Cano. Loci with the top 5% FST values ​​in each comparison combination were selected, and after removing duplicates, 2,139 candidate SNPs with high population discrimination potential were obtained. Finally, the feature importance of the 2,139 SNP loci was evaluated using the Gini coefficient (criterion=gini) and average impurity reduction (criterion=entropy) methods in a random forest model using Python 3.12.2. Loci with an importance of 0 were removed, and the intersection of the remaining loci was used to select those with a cumulative importance threshold of 0.9. The final set of loci selected thus explained more than 90% of the model's discrimination ability, thereby maximizing the reduction of the number of loci while retaining most of the key discrimination information.

[0024] The above screening process yielded 326 loci, detailed in Table 1. The chromosomes and their locations on the chromosomes in Table 1 were determined based on the pigeon GCA_001887795.1 reference genome. The distribution and number of these SNP loci on each chromosome are shown below. Figure 2 As shown.

[0025] Table 1. 326 SNP loci for identifying pigeon seed resources

[0026] Chromosome numbering Location of SNP sites on chromosomes Reference base of SNP site Mutant bases at SNP sites 1 2152672 C T 1 2606684 T C 1 6230367 T A 1 12404376 T C 1 12548806 C G 1 15857222 A G 1 17678137 C T 1 18930279 C A 1 20020038 T C 1 21811082 G C 1 22021779 G A 1 22047028 T C 1 22079534 C G 1 24349952 C T 1 27482488 G A 1 28080451 T G 1 33488907 C T 1 35009678 A C 1 38873688 A G 1 42327545 C T 1 46026191 A G 1 47549772 A G 1 52955336 T G 1 52994510 T C 1 53016477 C T 1 53049694 G A 1 53236127 C T 1 54648889 T C 1 54651713 A G 1 59603474 C T 1 60538581 A G 1 69146225 A C 1 74024369 G A 1 77315530 G A 1 82553843 A G 1 84689676 T C 1 89097556 C G 1 93262966 T C 1 101556274 C T 1 104340912 A G 1 105401145 T G 1 106361433 G A 1 106437940 T C 1 106558165 C T 1 107729282 A G 1 107739521 C T 1 108317293 C G 1 108403912 T C 1 108860985 A G 1 108876400 G A 1 108883516 A C 1 108980604 T C 1 108988565 A G 1 109059272 G A 1 109074013 G A 1 109119556 G A 1 109127410 G A 1 109142979 T G 1 109162480 T A 1 109168572 A G 1 109177298 T C 1 109197220 C A 1 109202651 T C 1 109208992 A G 1 109213361 A T 1 109217736 A G 1 109228432 T C 1 109232087 T C 1 109234972 G T 1 109238980 C T 1 109243971 G A 1 109251570 C T 1 109256763 T C 1 109261404 C T 1 109267208 T C 1 109282764 C G 1 109736467 A C 1 110332601 A C 1 110872513 G T 1 110924054 A G 1 111318089 T A 1 114910346 G A 1 115015445 A G 1 115219965 C T 1 116160864 C T 1 116181996 C T 1 117043913 G A 1 119729992 T G 1 128350096 G A 1 134991105 T C 1 140399740 C T 1 140859701 A G 1 141490282 T C 1 146051899 T C 1 149774610 A G 1 149941276 T C 1 149959100 T A 1 150095428 T G 1 151185852 T C 1 153115615 A G 1 161452112 G A 1 167052820 G A 1 168461185 C G 1 174579815 T C 1 174675965 T C 1 176954738 G T 1 177578591 T C 1 177648356 T C 1 201888912 G C 2 1269607 G A 2 1352862 A G 2 1395535 G A 2 2312364 G A 2 2825932 C T 2 4744342 G C 2 5053103 G A 2 15471353 A C 2 22216878 C T 2 23803268 T C 2 34016874 C T 2 37790707 A G 2 46113673 C T 2 46351550 G T 2 46951238 G A 2 52690065 T C 2 53743861 C T 2 55038281 A T 2 55080151 A C 2 55184521 G A 2 63068713 T C 2 63658608 G A 2 63664994 G A 2 63669207 C T 2 63677061 T G 2 63903720 T C 2 64149019 A G 2 64671387 G A 2 64683326 G A 2 67646387 T G 2 67689936 G A 2 67743256 C A 2 73014057 C T 2 74147719 T C 2 75064829 A C 2 76038581 C T 2 79132850 T G 2 81215711 C T 2 81276892 C A 2 86722021 T C 2 86779036 T C 2 87685208 G A 2 88635746 G C 2 88839012 T C 2 88870955 T A 2 89101467 G A 2 100719754 A G 2 129304354 G A 2 132162725 T G 2 135237924 A G 3 889164 C T 3 996233 T G 3 1615840 C T 3 3122376 G C 3 6376963 A G 3 6412035 T A 3 9294873 A G 3 24236811 A G 3 24593322 A G 3 27077462 T C 3 27505868 C T 3 34275401 C T 3 40427538 A C 3 46280337 A G 3 46304907 A G 3 59524265 A G 3 60175822 G A 3 60667537 T C 3 60971211 A G 3 62105203 T G 3 63777938 A G 3 67175424 G T 3 76045285 G A 3 76347842 T C 3 77357180 G T 3 81953730 A G 3 84877055 C G 3 96237595 T C 3 102143604 G C 3 109733241 G A 4 9295813 T A 4 10075031 G A 4 11877663 C A 4 14844751 T C 4 15080999 C T 4 16376616 C A 4 22144844 G A 4 22979500 G T 4 23827734 G A 4 31828773 G A 4 33568830 G A 4 33691332 G A 4 33839030 G A 4 47138671 G T 4 47770366 G A 4 47816491 G A 4 47873316 C A 4 47913613 T C 4 58807199 T C 5 3692497 C T 5 3776856 A G 5 6997054 A G 5 12682704 G A 5 18050735 C A 5 26509429 A C 5 29866281 A C 5 30864388 T C 5 36722747 C T 5 37626124 C T 5 40417477 G A 5 42500252 C T 5 42923723 A G 5 46312586 A G 5 46755641 C T 6 5506697 G A 6 5534735 C T 6 5806189 A G 6 6140247 T G 6 9088704 A G 6 13893134 C T 6 14657486 A G 6 16790137 T C 6 19432803 G T 6 25243921 T C 6 25851562 C T 6 26172152 G A 7 2168074 C T 7 16299724 T C 7 19228771 A G 7 20183446 A G 7 22048215 A G 7 23995271 T C 7 24302954 A G 8 8338474 C G 8 13010296 A G 8 16302089 G C 8 17994646 C G 8 24248944 T C 8 28171205 A G 8 28425214 T C 8 28839391 A G 9 2040981 G C 9 2668945 T A 9 3145194 T A 9 3943951 G C 9 4269368 T C 9 5963830 C A 9 8391198 G A 9 8418565 C G 9 8611908 G A 9 8894979 C A 9 14821485 A G 9 16739257 A G 9 20010759 A G 10 40883 T C 10 117275 C G 10 181193 C T 10 438515 T C 10 4130283 A G 10 9130516 T C 10 12914179 T C 10 13038965 G T 10 14018713 A G 10 15518195 G A 10 17585070 T C 11 1428674 C G 11 4703044 G C 11 5912716 C G 11 8038267 T C 11 9000333 C A 11 9393078 T C 11 16997718 C G 12 7291328 G A 13 6111761 C T 13 11882514 G A 13 13450403 A G 14 154207 T C 14 3609339 T C 14 6187683 C T 14 6444568 T C 15 10135788 G C 15 10196733 G T 15 10320512 C T 17 1996863 C T 17 5210142 T C 17 5219514 G A 17 5643461 A G 17 8166626 T A 17 10271740 A G 17 10633908 A C 18 716665 T C 18 921966 T C 18 5026211 G T 18 7676458 G A 20 4198358 A T 20 4208570 G A 20 6711545 C A 20 7233115 G A 20 7854962 C T 20 9251865 T C 21 2124842 A G 22 1624829 T C 23 461929 C T 23 869522 C G 23 2391300 G A 23 2641543 C G 23 5048185 C T 24 2286340 G A 24 2770626 T G 24 3852571 T G 28 763587 C T 4A 254882 T G 4A 276082 T C 4A 282622 A G 4A 310053 C T 4A 550260 A G 4A 3845425 A G

[0027] (3) Identification effect of 326 SNP loci in 70 pigeons

[0028] To systematically evaluate the discriminative power of the selected SNP loci, a rigorous 10-fold cross-validation process combined with a support vector machine algorithm was employed. During the data preparation phase, a complete dataset was constructed from 70 individual samples (10 samples from each of the seven groups) and their corresponding genotype data for 326 SNP loci, representing seven different populations: White Cano pigeons, White King pigeons, Mimas pigeons, Silver King pigeons, Taishen pigeons, Red Cano pigeons, and Yellow Cano pigeons.

[0029] The 10-fold cross-validation loop works as follows: First, the entire dataset (containing all 70 individual samples) is randomly divided into 10 roughly equal subsets, using stratified sampling to ensure that the proportion of samples from each group is approximately the same in each subset. In each round of validation, one subset is alternately designated as the test set, and the remaining nine subsets are combined as the training set. Then, the support vector machine model is trained using the training set data from the current round. This process essentially solves a constrained optimization problem to find the optimal classification hyperplane and determine the key hyperparameters of the support vectors (penalty parameter C and kernel function parameter γ).

[0030] After the model training is complete, its performance is evaluated using the test set of the current round. The genotype data of 326 SNPs for each individual in the test set (after the same preprocessing as the training set) are input into the trained support vector machine model, which outputs a predicted variety label for each individual. The classification accuracy on the test set for this round is calculated by comparing the predicted label with the true label. Finally, the average of the accuracies obtained from 10 rounds of validation is used as the overall evaluation of the model's ability to distinguish combinations of these 326 SNP loci.

[0031] The results showed that the final average accuracy of 10-fold cross-validation reached 100%. Figure 3 The results show that all 70 individuals were correctly classified without any misclassification, which strongly proves that the selected SNP locus combination has extremely high information content and can very accurately distinguish and identify these seven pigeon groups.

[0032] (4) Population structure analysis of 70 pigeons based on 326 SNP loci

[0033] Phylogenetic analysis was performed on 70 pigeons based on 326 SNP loci. The analysis involved using the `vcf2phylip.py` script to convert the genotype VCF files into .phy files, then using MEGA software to convert the .phy files into .meg files. A phylogenetic tree was then constructed from the .meg files using the neighbor-joining method and saved as a .nwk file. The results were visualized using the iTOL online website to analyze the phylogenetic relationships between different breeds. The phylogenetic tree results based on 326 SNP loci are shown below. Figure 4 As shown, the seven groups of pigeons can be accurately distinguished.

[0034] Example 2: Application of the SNP molecular marker combinations obtained in Example 1

[0035] This invention was applied to the identification of 20 white Cano pigeons and 26 silver king pigeons:

[0036] First, a total of 70 known individual samples (10 samples from each group) from the populations of White Cano pigeons, White King pigeons, Mimas pigeons, Silver King pigeons, Tai Shen pigeons, Red Cano pigeons, and Yellow Cano pigeons, along with their corresponding genotypic data of 326 SNP loci in Table 1, were used as the reference set data.

[0037] Then, 20 white Cano pigeons and 26 silver king pigeons were randomly selected as test samples (the group to which these 46 pigeons belonged was known to the sampling personnel of this experiment, but unknown to the testing personnel. The testing personnel compared the test results with those of the sampling personnel to verify the results). The genomic DNA of the test samples was obtained and sequenced. 326 SNP sites were extracted from the VCF file after sequencing data analysis, which are the 326 SNP sites of the test samples.

[0038] Then, using Python 3.12.2, the support vector machine model was trained with reference set data. The optimal hyperparameters were determined through grid search and cross-validation, with the penalty parameter C=1 and the kernel function parameter γ=0.001. The genotypic features of 326 SNP loci from 46 individual pigeons were then input into the optimized support vector machine model. The model then outputs the most probable breed label and provides the probability or confidence score for each breed, thus determining the population to which the individual pigeons belong.

[0039] Ultimately, the identification results were as follows: Figure 5 As shown, by Figure 5 As can be seen, all 46 pigeons tested were correctly distinguished, with an identification accuracy rate of 100%. This indicates that the molecular marker combination composed of the 326 SNP sites in Table 1 obtained in Example 1 has high accuracy and stability for identification.

[0040] The above embodiments are merely illustrative of the concept and implementation of the present invention and are not intended to limit it. Under the concept of the present invention, technical solutions without substantial changes are still within the scope of protection.

Claims

1. A method for identifying pigeon germplasm, characterized by According to 326 SNP sites, whether a pigeon individual belongs to one of seven pigeon populations, namely white cardan pigeon, white feather king pigeon, Mimas pigeon, silver feather king pigeon, Thai deep pigeon, red cardan pigeon and yellow cardan pigeon, is identified, and information of the 326 SNP sites is shown in the following table: The chromosome and the location of the SNP site on the chromosome are determined based on the pigeon GCA_001887795.1 reference genome.

2. The method of claim 1, wherein the method is characterized by The method comprises the following steps: Step 1: Obtain reference set data, which is the genotype of the 326 SNP sites of known pigeon samples of seven populations, namely white cardan pigeon, white feather king pigeon, Mimas pigeon, silver feather king pigeon, Thai deep pigeon, red cardan pigeon and yellow cardan pigeon, each population containing a number of pigeon samples; Step 2: Extract the genomic DNA of the pigeon individual to be tested, and obtain the genotype of the 326 SNP sites in the genomic DNA of the pigeon individual to be tested; Step 3: Train the support vector machine model using the reference set data, then input the genotype of the 326 SNP sites in the genomic DNA of the pigeon individual to be tested into the optimized support vector machine model, and determine the germplasm resource of the pigeon individual to be tested according to the output of the support vector machine model.

3. The method of claim 1, wherein the method is characterized by: In step 1, each population contains 10 pigeon samples; In step 3, the support vector machine model is trained based on the Python 3.12.2 environment using the reference set data, the optimal hyperparameters are locked through grid search and cross-validation, the optimal parameters are penalty parameter C=1 and kernel function parameter γ=0.001; then the genotype of the 326 SNP sites in the genomic DNA of the pigeon individual to be tested is input into the optimized support vector machine model, the output of the support vector machine model is the most possible breed label, the probability or confidence of each breed, and the germplasm resource of the pigeon individual to be tested is determined according to the output.

Images