SNP primer combination for identifying quiet chicken breeds and application thereof

Abstract

The application provides a SNP primer combination for identifying a static chicken breed and application thereof, and belongs to the technical field of biology. Whether a to-be-detected individual belongs to a static chicken is judged by detecting the genotypes of seven SNP sites of the static chicken. The method is simple in operation, high in accuracy, can accurately identify the static chicken breed, and effectively strikes the fake and inferior products of the static chicken local breed in the market.

Description

Technical Field

[0001] This invention belongs to the field of biotechnology, and in particular relates to an SNP primer combination for identifying Jingyu chicken breeds and its application. Background Technology

[0002] Jingyuan chicken, a dual-purpose (meat and egg) local breed, is mainly distributed in Jingning County, Gansu Province, and Guyuan City, Ningxia Hui Autonomous Region. It is an excellent breed resistant to the cold and arid climate of the Loess Plateau. Jingyuan chickens exhibit distinct local characteristics in their feather color, skin color, and meat quality. Their beaks are mostly gray, their irises are predominantly orange-red, their shanks are gray and robust, and they have a medium build. Adult roosters primarily have red and reddish-black plumage, while adult hens have a more complex plumage pattern, including yellow, black, white, and speckled, with yellow and mottled being the most common. Many counterfeit Jingyuan chickens, randomly crossbred with other breeds, have appeared on the market. These counterfeit breeds closely resemble Jingyuan chickens in appearance, making them difficult for consumers to distinguish, thus harming the interests of many farmers and hindering the preservation and breeding of the original Jingyuan chicken breed. Currently, the main method for identifying genuine Jingyuan chickens relies on visual observation, which is highly subjective, especially when identifying crossbred offspring of Jingyuan chickens and other breeds due to their similar appearance, leading to a high error rate. A method is needed to accurately identify the breed of Jingyuan chicken. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to provide an SNP primer combination for the identification of Jingyuan chicken breeds and its application.

[0004] To achieve the above-mentioned objectives, the present invention provides the following technical solution:

[0005] This invention provides an SNP primer combination for identifying the Jingyu chicken breed, the primer combination comprising primers for identifying any 5 to 7 loci from JY1 to JY7:

[0006] The JY1 locus is located on chromosome NC_006089.5, at locus chr2:88999342, and the bases at this locus are T / G.

[0007] The JY2 locus is located on chromosome NC_006089.5, at locus chr2:137016590, and the bases at this locus are C / T.

[0008] The JY3 locus is located on chromosome NC_006091.5, at locus chr4:22037922, and the bases at this locus are T / C.

[0009] The JY4 locus is located on chromosome NC_006095.5, at locus chr8:13527512, and the bases at this locus are C / A.

[0010] The JY5 locus is located on chromosome NC_006107.5, at locus chr20:9365087, and the bases at this locus are C / G.

[0011] The JY6 locus is located on chromosome NC_006111.5, at locus chr24:150425, and the bases at this locus are A / C.

[0012] The JY7 locus is located on chromosome NC_006114.5 at chr27:6092287, and its bases are A / C.

[0013] Preferably, the sites identified by the primer combination include: JY1+JY2+JY3+JY4+JY5, JY1+JY2+JY3+JY4+JY6, JY1+JY2+JY3+JY5+JY6, JY1+JY2+JY3+JY5+JY7, JY1+JY2+JY3+JY6+JY7, JY1+JY2+JY4+JY5+JY6. JY1+JY2+JY4+JY5+JY7, JY1+JY2+JY4+JY6+JY7, JY1+JY2+JY5+JY6+JY7, JY1+JY3+J Y4+JY5+JY6, JY1+JY3+JY4+JY5+JY7, JY1+JY3+JY4+JY6+JY7, JY1+JY3+JY5+JY6+JY7 , JY1+JY4+JY5+JY6+JY7, JY2+JY3+JY4+JY6+JY7, JY2+JY4+JY5+JY6+JY7, JY3+JY4+ JY5+JY6+JY7, JY1+JY2+JY3+JY4+JY5+JY6, JY1+JY2+JY3+JY4+JY5+JY7, JY1+JY2+J Y3+JY4+JY6+JY7, JY1+JY2+JY3+JY5+JY6+JY7, JY1+JY2+JY4+JY5+JY6+JY7, JY1+JY 3+JY4+JY5+JY6+JY7, JY2+JY3+JY4+JY5+JY6+JY7 or JY1+JY2+JY3+JY4+JY5+JY6+JY7.

[0014] Preferably, the sequences of each SNP primer are as follows:

[0015] The primer sequences for identifying the JY1 site are shown in SEQ ID NO.1-SEQ ID NO.3;

[0016] The primer sequences for identifying the JY2 site are shown in SEQ ID NO.4-SEQ ID NO.6;

[0017] The primer sequences for identifying the JY3 site are shown in SEQ ID NO.7-SEQ ID NO.9;

[0018] The primer sequences for identifying the JY4 site are shown in SEQ ID NO.10-SEQ ID NO.12;

[0019] The primer sequences for identifying the JY5 site are shown in SEQ ID NO.13-SEQ ID NO.15;

[0020] The primer sequences for identifying the JY6 site are shown in SEQ ID NO.16-SEQ ID NO.18;

[0021] The primer sequences for identifying the JY7 site are shown in SEQ ID NO.19-SEQ ID NO.21.

[0022] This invention provides the application of the SNP primer combination in the identification of Jingyuan chicken breeds.

[0023] Preferably, the identification method includes the following steps:

[0024] (1) Extract genomic DNA from the chickens to be tested;

[0025] (2) Using the chicken genomic DNA to be tested as a template, amplification was performed using the primers in the primer combination to obtain amplification products;

[0026] (3) Sequencing the amplification products and determining the genotypes at different loci;

[0027] (4) Based on the genotype results, determine whether the chicken individual to be tested belongs to the Jingyuan chicken breed;

[0028] The genotypes of the Jingyuan chickens at loci JY1 to JY7 are, in order: TG / GG, CC / CT / TT, TT / TC / CC, CC / CA / AA, CC / CG / GG, AA / AC / CC, AC / CC;

[0029] If the genotype of the selected SNP combination in the amplification product is consistent with that of the Jingyuan chicken, then the chicken individual to be tested is determined to belong to the Jingyuan chicken; if they are inconsistent, then the chicken individual to be tested is determined not to belong to the Jingyuan chicken.

[0030] Preferably, the genomic DNA of the chicken to be tested in step (1) is derived from wing vein blood.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] This invention provides a SNP primer combination for identifying the Jingyu Junglefowl breed. When using this SNP primer combination for Jingyu Junglefowl breed identification, the probability of identifying the test chicken as a Jingyu Junglefowl using the genotype of any single SNP locus is over 86.57%; the probability of identifying the genotype of any two SNP loci is over 83.58%; the probability of identifying the genotype of any three SNP loci is over 83.58%; the probability of identifying the genotype of any four SNP loci is over 91.79%; the probability of identifying the genotype of any five SNP loci is over 94.03%; the probability of identifying the genotype of any six SNP loci is over 96.27%; and the probability of identifying the genotype of all seven SNP loci is 98.51%. ROC curve analysis of the prediction results of this seven-locus combination shows that the accuracy rate for identifying the Jingyu Junglefowl breed is 0.9938. By selecting any 5 SNP loci and determining the genotype of the Jingyuan chicken, the success rate of detection can reach over 94%. This method is simple to operate, highly accurate, and can effectively combat counterfeit Jingyuan chicken products on the market. Attached Figure Description

[0033] Figure 1 A graph was constructed for the SVM algorithm, where A represents the top 20 genome-specific loci of Jingyuan Chicken selected using GBDT, and B represents the ability of 8 machine learning algorithms to identify Jingyuan Chicken at the 20 loci, thus obtaining the best algorithm - SVM.

[0034] Figure 2 To obtain the top 7 identifying loci of Jingyu chicken breeds through SVM algorithm;

[0035] Figure 3 This figure is used to evaluate the accuracy of the SVN model in identifying Jingyu chicken breeds using ROC curves. Detailed Implementation

[0036] This invention provides an SNP primer combination for identifying the Jingyu chicken breed, the primer combination comprising primers for identifying any 5 to 7 loci from JY1 to JY7:

[0037] The JY1 locus is located on chromosome NC_006089.5, at locus chr2:88999342, and the bases at this locus are T / G.

[0038] The JY2 locus is located on chromosome NC_006089.5, at locus chr2:137016590, and the bases at this locus are C / T.

[0039] The JY3 locus is located on chromosome NC_006091.5, at locus chr4:22037922, and the bases at this locus are T / C.

[0040] The JY4 locus is located on chromosome NC_006095.5, at locus chr8:13527512, and the bases at this locus are C / A.

[0041] The JY5 locus is located on chromosome NC_006107.5, at locus chr20:9365087, and the bases at this locus are C / G.

[0042] The JY6 locus is located on chromosome NC_006111.5, at locus chr24:150425, and the bases at this locus are A / C.

[0043] The JY7 locus is located on chromosome NC_006114.5 at chr27:6092287, and its bases are A / C.

[0044] In this invention, the primer combination includes primers that identify any 5 to 7 sites from JY1 to JY7, preferably JY1+JY2+JY3+JY4+JY5, JY1+JY2+JY3+JY4+JY6, JY1+JY2+JY3+JY5+JY6, JY1+JY2+JY3+JY5+JY7, JY1+JY2+JY3+JY6+JY7, JY1+JY2+JY4+JY 5+JY6, JY1+JY2+JY4+JY5+JY7, JY1+JY2+JY4+JY6+JY7, JY1+JY2+JY5+JY6+JY7, JY1+JY3+JY4+ JY5+JY6, JY1+JY3+JY4+JY5+JY7, JY1+JY3+JY4+JY6+JY7, JY1+JY3+JY5+JY6+JY7, JY1+JY4+JY 5+JY6+JY7, JY2+JY3+JY4+JY6+JY7, JY2+JY4+JY5+JY6+JY7, JY3+JY4+JY5+JY6+JY7, JY1+JY2+ JY3+JY4+JY5+JY6, JY1+JY2+JY3+JY4+JY5+JY7, JY1+JY2+JY3+JY4+JY6+JY7, JY1+JY2+JY3+JY 5+JY6+JY7, JY1+JY2+JY4+JY5+JY6+JY7, JY1+JY3+JY4+JY5+JY6+JY7, JY2+JY3+JY4+JY5+JY6+JY7 or JY1+JY2+JY3+JY4+JY5+JY6+JY7, further preferably JY1+JY2+JY4+JY6+JY7 or JY1+JY2+JY5+JY6+JY7.

[0045] In this invention, the primer sequences for each SNP site are as follows:

[0046] The primer sequences for identifying the JY1 site are shown in SEQ ID NO.1-SEQ ID NO.3:

[0047] SEQ ID NO.1: GAAGGTGACCAAGTTCATGCTTTGAATGTTCCAGCTAATCTCACTGA;

[0048] SEQ ID NO.2: GAAGGTCGGAGTCAACGGATTGAATGTTCCAGCTAATCTCACTG;

[0049] SEQ ID NO. 3: AAGGTTATTTATCTTTTTACAGAGAGTGAGG.

[0050] The primer sequences for identifying the JY2 site are shown in SEQ ID NO.4-SEQ ID NO.6:

[0051] SEQ ID NO.4: GAAGGTGACCAAGTTCATGCTAAGTAATTTCTTAACTGAGCCTAGCC;

[0052] SEQ ID NO.5: GAAGGTCGGAGTCAACGGATTTAAGTAATTTCTTAACTGAGCCTAGCT;

[0053] SEQ ID NO. 6: GTGACCTTGCATTGGATTTCTGCTTC.

[0054] The primer sequences for identifying the JY3 site are shown in SEQ ID NO.7-SEQ ID NO.9:

[0055] SEQ ID NO.7: GAAGGTGACCAAGTTCATGCTATAGTTGTTCCACTTCGAAGGTCTTT;

[0056] SEQ ID NO.8: GAAGGTCGGAGTCAACGGATTGTTGTTCCACTTCGAAGGTCTTC;

[0057] SEQ ID NO.9: GAGACACTCTAACAAAAGGAACAACAAG.

[0058] The primer sequences for identifying the JY4 locus are shown in SEQ ID NO.10-SEQ ID NO.12:

[0059] SEQ ID NO.10: GAAGGTGACCAAGTTCATGCTCAGTCTCTTCTAGCAGCTAAGAC;

[0060] SEQ ID NO.11: GAAGGTCGGAGTCAACGGATTCCAGTCTCTTCTAGCAGCTAAGAA;

[0061] SEQ ID NO. 12: GCAATGAATCTTTCTTGCTGCTGGAAG.

[0062] The primer sequences for identifying the JY5 site are shown in SEQ ID NO.13-SEQ ID NO.15:

[0063] SEQ ID NO.13: GAAGGTGACCAAGTTCATGCTCAGAGACAGACTTCAGCTCTGC;

[0064] SEQ ID NO.14: GAAGGTCGGAGTCAACGGATTCAGAGACAGACTTCAGCTCTGG;

[0065] SEQ ID NO. 15: TACTGCAGGTGAGTTGTCAGTTAG.

[0066] The primer sequences for identifying the JY6 locus are shown in SEQ ID NO.16-SEQ ID NO.18:

[0067] SEQ ID NO.16: GAAGGTGACCAAGTTCATGCTCATCAGAAACAGGATCAGCCATCAA;

[0068] SEQ ID NO.17: GAAGGTCGGAGTCAACGGATTTCAGAAACAGGATCAGCCATCAC;

[0069] SEQ ID NO. 18: CTCCCTGCTGTCGGAACAGTTAC.

[0070] The primer sequences for identifying the JY7 locus are shown in SEQ ID NO.19-SEQ ID NO.21:

[0071] SEQ ID NO.19: GAAGGTGACCAAGTTCATGCTTGTTGTAACCACCAAATTCACTGTGT;

[0072] SEQ ID NO.20: GAAGGTCGGAGTCAACGGATTGTTGTAACCACCAAATTCACTGTGG;

[0073] SEQ ID NO. 21: GACAACGTCCTTACCTTCAGGAATG.

[0074] The three primer sequences for each SNP site are in the following order: forward primer, forward primer and reverse primer, with the italicized portion of the forward primer representing the probe.

[0075] This invention provides the application of the SNP primer combination in the identification of Jingyuan chicken breeds, which can accurately identify Jingyuan chickens.

[0076] In this invention, the identification method includes the following steps:

[0077] (1) Extract genomic DNA from the chickens to be tested;

[0078] (2) Using the chicken genomic DNA to be tested as a template, amplification was performed using the primers in the primer combination to obtain amplification products;

[0079] (3) Sequencing the amplification products and determining the genotypes at different loci;

[0080] (4) Based on the genotype results, determine whether the chicken individual to be tested belongs to the Jingyuan chicken breed;

[0081] The genotypes of the Jingyuan chickens at loci JY1 to JY7 are, in order: TG / GG, CC / CT / TT, TT / TC / CC, CC / CA / AA, CC / CG / GG, AA / AC / CC, AC / CC;

[0082] If the genotype of the selected SNP combination in the amplification product is consistent with that of the Jingyuan chicken, then the chicken individual to be tested is determined to belong to the Jingyuan chicken; if they are inconsistent, then the chicken individual to be tested is determined not to belong to the Jingyuan chicken.

[0083] In this invention, the genomic DNA of the chicken to be tested in step (1) is preferably derived from wing vein blood; the amplification method in step (2) is preferably KASP amplification, and the amplification conditions are as follows: pre-denaturation 94℃, 15 min, cycle number 1; denaturation 94℃, 20 sec, cycle number 10; annealing / extension 61-55℃, 60 sec, cycle number 10; denaturation 94℃, 20 sec, cycle number 26; annealing / extension 55℃, 60 sec, cycle number 26.

[0084] In this invention, after fluorescently labeling the KASP amplification product of the chicken under test with a probe, the genotype is determined and compared with the genotypes corresponding to the selected SNP combinations. If all the genotypes corresponding to the selected SNP combinations match the genotypes of the Silent Red Barrel Chicken, the probability of the individual belonging to the Silent Red Barrel Chicken is determined according to the SVM machine learning algorithm (analyzed in the Python 3.0 environment). The specific steps of the SVM machine learning algorithm are as follows:

[0085] library(e1071)

[0086] library(caret)

[0087] library(dplyr)

[0088] rm(list = ls())

[0089] #The target variable is isJY, indicating whether it is a Jingyuan chicken: 0 indicates no, 1 indicates yes

[0090] snp_data<-read.csv('SNP.txt',row.names=1,sep='\t',check.names=F)

[0091] # Convert SNP sites and target variables to factor types

[0092] `snp_cols <- grep("SNP", names(snp_data), value = TRUE)` # Select columns whose names contain "JY".

[0093] snp_data[,snp_cols]<-lapply(snp_data[,snp_cols],as.factor)

[0094] snp_data$isJY<-as.factor(snp_data$isJY)

[0095] #Split the data into training and test sets

[0096] set.seed(123) # Set a random seed to ensure repeatability

[0097] trainIndex<-createDataPartition(snp_data$isJY,p=.8,list=FALSE)

[0098] trainData<-snp_data[trainIndex,]

[0099] testData<-snp_data[-trainIndex,]

[0100] #Function: Train an SVM model and calculate accuracy

[0101] train_and_evaluate<-function(trainData,testData,features){

[0102] #Training the SVM model

[0103] formula<-as.formula(paste("isJY~",paste(features,collapse="+")))

[0104] svm_model<-svm(formula,data=trainData,kernel='radial')

[0105] #Making Predictions

[0106] predictions<-predict(svm_model,newdata=testData)

[0107] #Calculate accuracy

[0108] accuracy<-mean(predictions==testData$isJY)

[0109] return (accuracy)

[0110] }

[0111] # Initialize a data frame to store the results

[0112] results <- data.frame(

[0113] num_features = integer(),

[0114] features = character(),

[0115] accuracy = numeric(),

[0116] stringsAsFactors=FALSE )

[0118] #Cycle with different numbers of feature combinations; there are 7 in total, so the ratio is 1:7.

[0119] for(num_features in 1:7){

[0120] feature_combinations<-combn(snp_cols,num_features,simplify=FALSE)

[0121] for(combo in feature_combinations){

[0122] features<-paste(combo,collapse="+")

[0123] accuracy<-train_and_evaluate(trainData,testData,combo)

[0124] results<-rbind(results,data.frame(

[0125] num_features = num_features,

[0126] features = features

[0127] accuracy = accuracy

[0128] stringsAsFactors=FALSE ))

[0130] }}

[0131] #View results

[0132] print(results)

[0133] write.csv(results,'results.csv',row.names=F)

[0134] Experimental materials:

[0135] The poultry germplasm resource genetic material bank of the local chicken germplasm resource conservation and utilization team of Henan Agricultural University preserves the following chickens: Jingyuan chicken (JY), Albert Ikka broiler (AA), Gushi chicken X Anka broiler F2 generation (F2), Gushi chicken (GS), Hy-Line Brown laying hen (HL), Xichuan black-bone chicken (XC), and Hubbard broiler (HBD).

[0136] The technical solutions provided by the present invention will be described in detail below with reference to the embodiments, but they should not be construed as limiting the scope of protection of the present invention.

[0137] Experiment Example 1: Identification of Gene Loci in Jingyu Chicken and Primer Design

[0138] Genetic differentiation index analysis was performed using resequencing data from 60 Jingyuan chickens, resequencing data from 3536 chickens publicly available in databases, and 600K microarray sequencing data. Specific breed and sequencing information can be found in "Genetic Structure Analysis of Local Chicken Populations in Henan Province and Development and Utilization of a Molecular Identification System for Chicken Genetic Resources" (Zhi Yihao, Henan Agricultural University, June 1, 2023). The top 20 loci of the genetic differentiation index of the Jingyuan chicken population were obtained for machine learning model building. A Gradient Boosting Decision Tree Machine (GBDT) model was used to extract important features. Then, eight machine learning classification methods were used to construct training models, including multiple linear regression (MLR), support vector machine (SVM), K-nearest neighbors (KNN), Naive Bayes (NB), Decision Trees (DT), random forest (RF), BackPropagation Neural Network (BPNN), and GBDT. The optimal hyperparameter combination was found using GridSearchCV. The classification model was built using Python 3.6. Breed-specific SNP loci obtained from the optimal training model were extracted for validation. As the number of SNPs increases, the accuracy of identifying the Jingyuan chicken breed also improves. The accuracy of identification using the SVM model peaks at 7 SNPs. Figure 1 The SVM model specifically identified seven molecular markers unique to the Jingyuan Chicken: chr2:88999342, chr2:137016590, chr4:22037922, chr8:13527512, chr20:9365087, chr24:150425, and chr27:6092287. Figure 2 The information for the 7 SNP sites is shown in Table 1 below.

[0139] Table 1 Information on 7 SNP sites

[0140] Serial Number chromosome site Reference genomic sites mutation site JY1 NC_006089.5 chr2:88999342 T G JY2 NC_006089.5 chr2:137016590 C T JY3 NC_006091.5 chr4:22037922 T C JY4 NC_006095.5 chr8:13527512 C A JY5 NC_006107.5 chr20:9365087 C G JY6 NC_006111.5 chr24:150425 A C JY7 NC_006114.5 chr27:6092287 A C

[0141] The genotypes of the SNP loci in Jingyuan chickens and other breeds are shown in Table 2 below (in the SNP column, 0 indicates wild-type homozygous, 1 indicates mutant heterozygous, and 2 indicates mutant homozygous; in the "Is it JY?" column, 0 indicates no and 1 indicates yes):

[0142] Table 2 Genotypes of SNP loci in Jingyuan chickens and other breeds

[0143]

[0144]

[0145]

[0146]

[0147]

[0148]

[0149]

[0150]

[0151]

[0152]

[0153]

[0154]

[0155]

[0156] Note: Letters are abbreviations of breeds, and numbers are the sample numbers of each breed. The specific breed abbreviations and names are (AA-broiler chicken, CS-Changshun green-shelled laying hen, DX-Dongxiang black-feathered green-shelled laying hen, F2-F2 generation of cross between Anka broiler chicken and Gushi chicken from Ziyuan Farm, GS-Gushi chicken, HBD-Hubbard broiler chicken, HL-Hy-Line Brown laying hen, JY-Jingyuan chicken, KB-Cobb broiler chicken, RM-Lohmann laying hen, XC-Xichuan black-bone chicken).

[0157] Based on the physical locations of the seven SNP loci, and using the red junglefowl reference genome sequence (Galgal6), the DNA sequences of the seven SNP loci were obtained through the NCBI website (https: / / www.ncbi.nlm.nih.gov / ). Primers were then designed according to the KASP detection rules. The primer sequences for each SNP are shown in Table 3 below.

[0158] Table 3. Primer sequences for each SNP

[0159]

[0160]

[0161] The three primers for each SNP site are, in order, a forward primer, a reverse primer, and the italicized portion of the forward primer is the probe.

[0162] Based on the typing results, SVM model calculations were performed to obtain the probability of identifying an individual as a Silent Red Barnyard Grasshopper after different numbers and combinations of 7 SNP loci. The specific steps of the SVM machine learning algorithm to determine the probability that an individual belongs to a Silent Red Barnyard Grasshopper are as follows:

[0163] library(e1071)

[0164] library(caret)

[0165] library(dplyr)

[0166] rm(list = ls())

[0167] #The target variable is isJY, indicating whether it is a Jingyuan chicken: 0 indicates no, 1 indicates yes

[0168] snp_data<-read.csv('SNP.txt',row.names=1,sep='\t',check.names=F)

[0169] # Convert SNP sites and target variables to factor types

[0170] `snp_cols <- grep("SNP", names(snp_data), value = TRUE)` # Select columns whose names contain "JY".

[0171] snp_data[,snp_cols]<-lapply(snp_data[,snp_cols],as.factor)

[0172] snp_data$isJY<-as.factor(snp_data$isJY)

[0173] #Split the data into training and test sets

[0174] set.seed(123) # Set a random seed to ensure repeatability

[0175] trainIndex<-createDataPartition(snp_data$isJY,p=.8,list=FALSE)

[0176] trainData<-snp_data[trainIndex,]

[0177] testData<-snp_data[-trainIndex,]

[0178] #Function: Train an SVM model and calculate accuracy

[0179] train_and_evaluate<-function(trainData,testData,features){

[0180] #Training the SVM model

[0181] formula<-as.formula(paste("isJY~",paste(features,collapse="+")))

[0182] svm_model<-svm(formula,data=trainData,kernel='radial')

[0183] #Making Predictions

[0184] predictions<-predict(svm_model,newdata=testData)

[0185] #Calculate accuracy

[0186] accuracy<-mean(predictions==testData$isJY)

[0187] return (accuracy)

[0188] }

[0189] # Initialize a data frame to store the results

[0190] results <- data.frame(

[0191] num_features = integer(),

[0192] features = character(),

[0193] accuracy = numeric(),

[0194] stringsAsFactors=FALSE )

[0196] #Cycle with different numbers of feature combinations; there are 7 in total, so the ratio is 1:7.

[0197] for(num_features in 1:7){

[0198] feature_combinations<-combn(snp_cols,num_features,simplify=FALSE)

[0199] for(combo in feature_combinations){

[0200] features<-paste(combo,collapse="+")

[0201] accuracy<-train_and_evaluate(trainData,testData,combo)

[0202] results<-rbind(results,data.frame(

[0203] num_features = num_features,

[0204] features = features

[0205] accuracy = accuracy

[0206] stringsAsFactors=FALSE ))

[0208] }}

[0209] #View results

[0210] print(results)

[0211] write.csv(results,'results.csv',row.names=F)

[0212] The calculation results are shown in Table 4 below:

[0213] Table 4. Probability of identifying *Gnaphalium affine* with different numbers and combinations of the 7 SNP loci.

[0214]

[0215]

[0216]

[0217]

[0218]

[0219] It can be seen that the probability of identifying Jingyuan chicken as a single genotype based on any 5 SNP loci is over 94.03%.

[0220] The samples in Table 2, containing multiple breeds, were divided into a training set (80%) and a test set (20%) using the `sample.split` function. The training set was used for model training and parameter optimization, while the test set was used to validate model performance. An SVM model was built on the training set, and parameters were tuned to obtain the optimal model configuration. The trained model was then used on the test set for prediction, generating a confusion matrix and ROC curve. The AUC value was calculated. Analysis of the prediction results for this combination of seven loci using ROC curves and AUC values ​​revealed that the breed identification accuracy for Jingyuan chicken was 0.9938. See details in [link to results]. Figure 3 .

[0221] Example 1

[0222] Twenty-five Jingyuan chickens, 25 Albert Iberia broilers, 25 Gushi chickens x Anka broiler F2 generation chickens, 25 Gushi chickens, 25 Hy-Line Brown chickens, 25 Xichuan Silkie chickens, and 25 Hubbard broilers were randomly selected from the poultry germplasm resource genetic material bank of the Henan Agricultural University Local Chicken Germplasm Resource Conservation and Utilization Team Laboratory. Blood was collected from the wing veins, and genomic DNA was extracted using the phenol-chloroform method. Primers for five SNP loci (JY1, JY2, JY4, JY6, and JY7) were selected for KASP typing.

[0223] Based on the typing results, all 25 Jingyuan chicken individuals were identified as Jingyuan chickens, while the remaining chickens were not of the Jingyuan chicken breed.

[0224] Example 2

[0225] Other conditions are the same as in Example 1, except that primers for the six SNP loci corresponding to JY1, JY2, JY3, JY5, JY6, and JY7 are used for KASP amplification. Based on the genotyping results, an SVM model is used for prediction. The probability of the combined genotype of the six loci being *Gnaphalium affine* is no less than 98.5%, indicating that a 98.5% probability of identifying all 25 individuals as *Gnaphalium affine* can be achieved using these six loci. If the probability obtained by the algorithm is less than 98.5%, the individual is classified as not being *Gnaphalium affine*.

[0226] Example 3

[0227] Other conditions are the same as in Example 1, except that primers for the seven SNP loci corresponding to JY1 to JY7 are used for KASP amplification. Based on the genotyping results, an SVM model is used for prediction. The probability of the combined genotype of the seven loci being *Gnaphalium affine* is no less than 98.5%, indicating that a 98.5% probability of identifying all 25 individuals as *Gnaphalium affine* can be achieved using these seven loci. If the probability obtained by the algorithm is less than 98.5%, the individual is classified as not being *Gnaphalium affine*.

[0228] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. An SNP primer combination for identifying Jingyu chicken breeds, characterized in that, The primer combination includes primers that identify any 5 to 7 sites from JY1 to JY7 as follows: The JY1 locus is located on chromosome NC_006089.5, at locus chr2:88999342, and the bases at this locus are T / G. The JY2 locus is located on chromosome NC_006089.5, at locus chr2:137016590, and the bases at this locus are C / T. The JY3 locus is located on chromosome NC_006091.5, at locus chr4:22037922, and the bases at this locus are T / C. The JY4 locus is located on chromosome NC_006095.5, at locus chr8:13527512, and the bases at this locus are C / A. The JY5 locus is located on chromosome NC_006107.5, at locus chr20:9365087, and the bases at this locus are C / G. The JY6 locus is located on chromosome NC_006111.5, at locus chr24:150425, and the bases at this locus are A / C. The JY7 locus is located on chromosome NC_006114.5, at locus chr27:6092287, and the bases at this locus are A / C. The sequences of each SNP primer are as follows: The primer sequences for identifying the JY1 site are shown in SEQ ID NO.1-SEQ ID NO.3; The primer sequences for identifying the JY2 site are shown in SEQ ID NO.4-SEQ ID NO.6; The primer sequences for identifying the JY3 site are shown in SEQ ID NO.7-SEQ ID NO.9; The primer sequences for identifying the JY4 site are shown in SEQ ID NO.10-SEQ ID NO.12; The primer sequences for identifying the JY5 site are shown in SEQ ID NO.13-SEQ ID NO.15; The primer sequences for identifying the JY6 site are shown in SEQ ID NO.16-SEQ ID NO.18; The primer sequences for identifying the JY7 site are shown in SEQ ID NO.19-SEQ ID NO.

21.

2. The SNP primer combination according to claim 1, characterized in that, The sites identified by the primer combinations include: JY1+JY2+JY3+JY4+JY5, JY1+JY2+JY3+JY4+JY6, JY1+JY2+JY3+JY5+JY6, JY1+JY2+JY3+JY5+JY7, JY1+JY2+JY3+JY6+JY7, JY1+JY2+JY4+JY5+JY6, JY1 +JY2+JY4+JY5+JY7、JY1+JY2+JY4+JY6+JY7、JY1+JY2+JY5+JY6+JY7、JY1+JY3+JY4+ JY5+JY6, JY1+JY3+JY4+JY5+JY7, JY1+JY3+JY4+JY6+JY7, JY1+JY3+JY5+JY6+JY7, J Y1+JY4+JY5+JY6+JY7, JY2+JY3+JY4+JY6+JY7, JY2+JY4+JY5+JY6+JY7, JY3+JY4+J Y5+JY6+JY7, JY1+JY2+JY3+JY4+JY5+JY6, JY1+JY2+JY3+JY4+JY5+JY7, JY1+JY2+JY 3+JY4+JY6+JY7, JY1+JY2+JY3+JY5+JY6+JY7, JY1+JY2+JY4+JY5+JY6+JY7, JY1+JY3 +JY4+JY5+JY6+JY7, JY2+JY3+JY4+JY5+JY6+JY7 or JY1+JY2+JY3+JY4+JY5+JY6+JY7.

3. The application of the SNP primer combination described in claim 1 or 2 in the identification of Jingyuan chicken breeds.

4. The application according to claim 3, characterized in that, The identification method includes the following steps: (1) Extract genomic DNA from the chicken to be tested; (2) Using the chicken genomic DNA to be tested as a template, amplification is performed using the primers in the primer combination described in claim 1 to obtain amplification products; (3) Sequencing the amplification products and determining the genotypes at different loci; (4) Based on the genotype results, determine whether the chicken individual to be tested belongs to the Jingyuan chicken breed; The genotypes of the Jingyuan chickens at the JY1 to JY7 loci are, in order, TG / GG, CC / CT / TT, TT / TC / CC, CC / CA / AA, CC / CG / GG, AA / AC / CC, and AC / CC; If the genotype of the selected SNP combination in the amplification product is consistent with that of the Jingyuan chicken, then the chicken individual to be tested is determined to belong to the Jingyuan chicken. If there is no discrepancy, the tested chicken individual is determined not to belong to the Jingyuan chicken family.

5. The application according to claim 4, characterized in that, In step (1), the genomic DNA of the chicken to be tested was obtained from the wing vein blood.

Images