A method, device and medium for constructing a molecular ID card for beef cattle breeds

Abstract

The present application discloses an application method, device and medium for constructing a molecular ID card of a beef cattle breed, and relates to the field of data processing. The method comprises: using a training set to train a machine learning model, and using a test set to test the trained machine learning model, until the prediction accuracy of the trained machine learning model reaches a set value, the trained machine learning model is used as a verification model; using a feature filtering strategy, a specific SNP site set is screened in the training set, and the sites in the specific SNP site set verified by the verification model are used as breed characteristic information sites; using a forward search method to search for the minimum SNPs set of breed characteristic information sites, and after verification by the verification model, a molecular ID card of the beef cattle breed is obtained. The present application can obtain a breed molecular ID card that can fully reflect the genetic diversity of beef cattle breeds, and improve the efficiency and accuracy of beef cattle breed detection.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular to an application method, device and medium for constructing a molecular identity card of a beef cattle breed. Background Art

[0002] Germplasm resources are indispensable for the survival and sustainable development of human society, the material basis for original innovation in life sciences and the development of the bio-industry. Their ownership and the degree of research and development and utilization have become one of the important indicators for measuring sustainable development capabilities. Beef cattle have excellent characteristics such as tolerance to roughage, strong disease resistance, docile temperament, good adaptability, stable genetics, and good meat quality. They are important germplasm resources for the development of the beef cattle industry. However, there are many varieties of beef cattle, and the regional situation is complex. The phenomenon of different names for the same species and different species for the same name is common, which brings great challenges to the protection and utilization of beef cattle germplasm resources.

[0003] Since the 20th century, DNA (deoxyribonucleic acid) molecular marker technology [Single Nucleotide Polymorphism (SNP) markers, Simple Sequence Repeats (SSR) markers, Random Amplified Polymorphic DNA markers (RAPD) and other molecular marker technologies] have gradually been applied in germplasm resource-related research, and some studies have begun to use it to construct variety molecular ID cards. Given that SNP molecular marker technology has the characteristics of automation, high throughput and strong genetic stability, it has been widely used in the construction of molecular ID cards.

[0004] With the continuous expansion of biological data, when constructing a molecular identity card of a breed, it is necessary to select appropriate features from a large number of genetic markers (such as SNPs). At the same time, there may be complex interactions between genetic markers, and it is difficult to determine which marker combinations can most effectively represent breed characteristics. Machine learning methods can construct potential biological process information and model predictions in inherently complex situations, and provide importance measures for given features [such as single nucleotide polymorphism sites (SNPs)], which helps to determine which features have the greatest impact on the predicted target. By utilizing these importance measures, machine learning can achieve good predictive performance, that is, given the features, it can accurately predict the value of the target variable. Therefore, this patent application provides an application method, device and medium for constructing a molecular identity card of a beef cattle breed based on machine learning, which provides important technical support for the accurate identification of beef cattle breeds, as well as the preservation and utilization of cattle resources. Summary of the invention

[0005] The purpose of this application is to provide a method, device and medium for constructing and applying a molecular ID card of a beef cattle breed, which can obtain a breed molecular ID card that can comprehensively reflect the genetic diversity of beef cattle breeds and improve the efficiency and accuracy of beef cattle breed detection.

[0006] To achieve the above objectives, this application provides the following solutions:

[0007] In a first aspect, the present application provides a method for constructing a molecular ID card of a beef cattle breed, comprising:

[0008] Acquire sample data; the sample data includes beef cattle breeds and SNP site information corresponding to each beef cattle breed;

[0009] Dividing the sample data into a training set and a test set;

[0010] The training set is used to train the machine learning model, and during the training process, the trained machine learning model is tested with the test set until the prediction accuracy of the trained machine learning model reaches a set value, and the trained machine learning model is used as a verification model;

[0011] By means of a feature filtering strategy, specific SNP sites of different varieties are screened in the training set to obtain a specific SNP site set;

[0012] The validation model is used to validate the specific SNP site set, and the sites in the validated specific SNP site set are used as variety characteristic information sites;

[0013] Using a forward search method to search for the minimum SNPs set for the variety characteristic information loci;

[0014] The verification model is used to verify the minimum SNPs set obtained by the search, and the minimum SNPs set verified by the verification model is used as the molecular identity card of the beef cattle breed.

[0015] Optionally, using the verification model to verify the specific SNP site set, and using the sites in the verified specific SNP site set as variety characteristic information sites, comprises:

[0016] Using the beef cattle breed information corresponding to the specific SNP locus set as a sample label, and combining the specific SNP locus set to generate a first sample pair;

[0017] Determining the prediction accuracy of the first sample pair using the verification model, and determining whether the prediction accuracy reaches the set value;

[0018] When the prediction accuracy does not reach the set value, the specific SNP sites of different varieties are re-screened in the training set by means of a feature filtering strategy to obtain a new set of specific SNP sites, and the new set of specific SNP sites is re-validated using the validation model until the prediction accuracy reaches the set value;

[0019] When the prediction accuracy reaches the set value, the sites in the specific SNP site set are used as variety characteristic information sites.

[0020] Optionally, the verification model is used to verify the minimum SNPs set obtained by the search, and the minimum SNPs set verified by the verification model is used as the molecular identity card of the beef cattle breed, including:

[0021] The minimum SNPs set and the sample label form a second sample pair;

[0022] Determining the prediction accuracy of the second sample pair using the verification model, and determining whether the prediction accuracy reaches the set value;

[0023] When the prediction accuracy does not reach the set value, the minimum SNPs set search is performed again on the variety characteristic information site to obtain a new minimum SNPs set, and the new minimum SNPs set is verified again using the verification model until the prediction accuracy reaches the set value;

[0024] When the prediction accuracy reaches the set value, the minimum SNPs set is used as the molecular identity card of the beef cattle breed.

[0025] Optionally, by means of a feature filtering strategy, specific SNP sites of different varieties are screened in the training set to obtain a specific SNP site set, including:

[0026] With the help of feature filtering strategy, the sample data in the training set are feature filtered, and in the feature filtering process, the Delta method is used to screen specific SNP sites of different varieties in the training set to obtain the specific SNP site set.

[0027] Optionally, the Delta method is used to screen specific SNP sites of different varieties in the training set to obtain the specific SNP site set, including:

[0028] Determine the Delta value of each specific SNP site between each beef cattle breed in the sample data;

[0029] Arranging the specific SNP sites of each beef cattle breed in descending order according to the Delta value of the SNP site to obtain the site sequence of each beef cattle breed;

[0030] The top N% of specific SNP sites in the site sequence of each beef cattle breed are retained to obtain the specific SNP site set.

[0031] Optionally, the machine learning model is a support vector machine model.

[0032] In a second aspect, the present application provides a method for applying a molecular ID card of a beef cattle breed, comprising:

[0033] Acquire the SNPs site information of the beef cattle to be tested based on the molecular ID card of the beef cattle breed; the molecular ID card of the beef cattle breed is obtained by the method for constructing the molecular ID card of the beef cattle breed provided above;

[0034] A breed detection model is obtained, and the SNPs site information is input into the breed detection model to obtain the breed information of the beef cattle to be detected.

[0035] Optionally, the variety detection model is constructed based on a support vector machine.

[0036] In a third aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-mentioned method for constructing or applying a molecular identity card for beef cattle breeds.

[0037] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for constructing or applying a molecular identity card for beef cattle breeds provided above.

[0038] According to the specific embodiments provided in this application, this application has the following technical effects:

[0039] The present application provides an application method, device and medium for constructing a molecular ID card of beef cattle breeds. By using a feature filtering strategy and combining a forward search method, the subtle differences and common characteristics of different beef cattle breeds at the genetic level are deeply mined, and an advanced machine learning algorithm model (i.e., a verification model) is used to efficiently train and screen a large amount of genetic data, so that a molecular ID card of beef cattle breeds that can fully reflect the genetic diversity of beef cattle breeds can be obtained. By using the obtained molecular ID card of beef cattle breeds, combined with a machine learning algorithm model (i.e., a breed detection model), the detection of beef cattle breeds can be achieved quickly and accurately. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0041] Figure 1 A schematic diagram of a process for constructing a molecular ID card for a beef cattle breed provided in one embodiment of the present application;

[0042] Figure 2 This is a diagram of the implementation of a method for constructing a molecular ID card for beef cattle breeds provided in one embodiment of the present application;

[0043] Figure 3 A schematic diagram of a process for applying a molecular ID card for beef cattle breeds provided in one embodiment of the present application;

[0044] Figure 4 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0046] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0047] In an exemplary embodiment, the present application provides a method for constructing a molecular ID card of a beef cattle breed, such as Figure 1 As shown, the method includes:

[0048] Step 100: Obtain sample data. The sample data includes beef cattle breeds and SNP site information corresponding to each beef cattle breed.

[0049] Step 101: Divide the sample data into a training set and a test set.

[0050] Step 102: Use the training set to train the machine learning model, and during the training process, use the test set to test the trained machine learning model until the prediction accuracy of the trained machine learning model reaches a set value, and then use the trained machine learning model as a verification model.

[0051] Step 103: Using a feature filtering strategy, specific SNP sites of different varieties are screened in the training set to obtain a specific SNP site set.

[0052] Step 104: Use the verification model to verify the specific SNP site set, and use the sites in the verified specific SNP site set as variety characteristic information sites.

[0053] Step 105: Use the forward search method to search for the minimum SNPs set for the variety characteristic information loci.

[0054] Step 106: Use the verification model to verify the minimum SNPs set obtained by the search, and use the minimum SNPs set verified by the verification model as the molecular identity card of the beef cattle breed.

[0055] In another exemplary embodiment of the present application, in order to efficiently train and screen massive amounts of genetic data, in this embodiment, the implementation process of the above step 104 may include:

[0056] Step 1041: Use the beef cattle breed information corresponding to the specific SNP locus set as a sample label and combine it with the specific SNP locus set to generate a first sample pair.

[0057] Step 1042: Use the validation model to determine the prediction accuracy of the first sample pair, and determine whether the prediction accuracy reaches a set value (eg, 99%).

[0058] Step 1043: When the prediction accuracy does not reach the set value, the specific SNP sites of different varieties are re-screened in the training set with the help of feature filtering strategy to obtain a new set of specific SNP sites, and the verification model is re-used to verify the new set of specific SNP sites until the prediction accuracy reaches the set value.

[0059] Step 1044: When the prediction accuracy reaches the set value, the sites in the specific SNP site set are used as variety characteristic information sites.

[0060] In another exemplary embodiment of the present application, in order to efficiently train and screen massive amounts of genetic data, in this embodiment, the implementation process of the above step 106 may include:

[0061] Step 1061: The minimum SNPs set and sample labels are used to form a second sample pair.

[0062] Step 1062: Use the verification model to determine the prediction accuracy of the second sample pair, and determine whether the prediction accuracy reaches a set value.

[0063] Step 1063: When the prediction accuracy does not reach the set value, the minimum SNPs set search is performed again on the variety characteristic information sites to obtain a new minimum SNPs set, and the verification model is used again to verify the new minimum SNPs set until the prediction accuracy reaches the set value.

[0064] Step 1064: When the prediction accuracy reaches the set value, the minimum SNPs set is used as the molecular identity card of the beef cattle breed.

[0065] In another exemplary embodiment of the present application, step 103 may be implemented as follows:

[0066] With the help of feature filtering strategy, the sample data in the training set are feature filtered, and in the feature filtering process, the Delta (SHRIVER et al., 1997) method is used to screen the specific SNP sites of different varieties in the training set to obtain a specific SNP site set. Among them:

[0067] (1) Determine the Delta value of each specific SNP locus among various beef cattle breeds in the sample data.

[0068] For any two species i and j The calculation formula of the Delta value of each specific SNP site is expressed as:

[0069] .

[0070] In the formula, and Allele A in the variety i and varieties j The gene frequency in . It is a variety i and varieties j The Delta value for each specific SNP site.

[0071] (2) The specific SNP loci of each beef cattle breed were arranged in descending order according to the Delta value of the SNP loci to obtain the locus sequence of each beef cattle breed.

[0072] (3) The top N% of specific SNP sites in the site sequence of each beef cattle breed are retained to obtain a specific SNP site set.

[0073] For example, the Delta value of each specific SNP locus between different beef cattle breeds was calculated using R language , sorted from large to small, selected the SNP sites ranked TOP 1% ~ TOP 10% among each beef cattle breed (a total of 10 groups were selected), removed the duplicate sites, and the retained sites were used as the breed characteristic information sites (Most breed-informative SNPs, MBISNPs) to obtain a specific SNP site set.

[0074] In another exemplary embodiment of the present application, the data quality of the core germplasm is crucial for constructing a molecular identity card. If the core germplasm is not properly selected, some individuals or breeds with important genetic characteristics may be missed, resulting in the constructed molecular identity card failing to fully reflect the genetic diversity of the entire beef cattle breed. In order to overcome this problem, in this embodiment, 939 beef cattle from 29 breeds in 15 provinces in China were selected as research objects, and the implementation process of the method for constructing a molecular identity card for beef cattle breeds provided above in this application is described. Based on this, in this embodiment, the process of constructing a molecular identity card for 939 beef cattle from 29 breeds in 15 provinces in China is as follows: Figure 2 As shown, including:

[0075] S1: Beef cattle sample collection and genotype data quality control.

[0076] (1) Collection of beef cattle samples.

[0077] The selected research subjects included 939 beef cattle of 29 breeds as shown in Table 1 .

[0078]

[0079] (2) Genotype data quality control.

[0080] Blood or frozen semen from all breeds of beef cattle were collected and extracted according to the instructions of the DNA extraction kit. The obtained sample DNA was tested for purity and concentration using an ultraviolet spectrophotometer and agarose gel electrophoresis. The DNA concentration was greater than 50 ng / μL as a qualified condition, and the A260 / A280 ratio should be between 1.9 and 2.1. Then, gene chips were used for typing to obtain the genotype information of genomic SNPs.

[0081] Prior to analysis, the genotype data of 939 cattle from 29 beef cattle breeds were quality controlled using PLINK v 1.9 software:

[0082] First, all the genotype data of varieties were merged and quality controlled. The running code was: plink --bfilefilename1 --merge filename2 --make-bed --out filename, plink --cow --bfilefilename --geno 0.1 --make-bed --out filename (the genotype data of varieties in file 1 and file 2 were merged and re-encoded and output, and the sites with typing rates less than 90% were filtered out).

[0083] Secondly, quality control was performed again for each variety, and the running code was: plink --cow --bfile filename --geno 0.1 --mind 0.1 --maf 0.05 --hwe 0.000001 --make-bed --out filename (retain loci with a minimum allele frequency greater than 0.05, a typing rate greater than 90%, and a Hardy-Weinberg equilibrium of 0.000001, and retain individuals with a typing rate greater than 90%).

[0084] After quality control, Beagle 4.0 software was used to fill in the missing SNP sites according to the varieties. The running command of Beagle 4.0 software was: java -Xmx1000m -jar unphased=file.bgl out=output niterations=100 (phased self-filling).

[0085] The genotype data of each variety were merged again for quality control, and SNPs with a detection rate less than 0.9 were removed. The running code was: plink --bfile filename1 --merge filename2 --make-bed --outfilename, plink --cow --bfile filename --geno 0.1 --make-bed --out filename. Finally, 22,529 autosomal common SNP sites of 939 individuals were obtained for subsequent analysis.

[0086] S2: Division of training set and test set: 80% of beef cattle samples in each breed were randomly selected as training set, and the remaining 20% ​​of beef cattle samples were used as test set. Feature filtering, machine learning, minimum subset search and other operations were performed in the training set, and the accuracy of the model was evaluated in the test set. Beef cattle samples included beef cattle breeds and 22,529 autosomal public SNP loci.

[0087] S3: Use machine learning combined with feature filtering to train using a training set.

[0088] (1) Screening of variety-characteristic SNP loci.

[0089] By means of the feature filtering strategy, the specific sites of different breeds are screened within the obtained training set. In the feature filtering process, the Delta method is used to select the SNP sites ranked TOP 1% to TOP 10% among the beef cattle breeds (a total of 10 groups are selected) to determine the breed-specific single nucleotide polymorphism (SNP) sites. For example, in this embodiment, the feature filtering obtains a total of 10 groups of specific site sets, the numbers of which are: 203, 340, 462, 593, 702, 798, 886, 983, 1068, and 1163, respectively.

[0090] (2) Machine learning model training.

[0091] PLINK v 1.9 was used to recode each specific locus set obtained. The running code was: plink--bfile filename --recode A --out filename (recoding) to obtain the raw format SNPs genotype data file of 939 beef cattle. For example, the support vector machine (SVM) method was selected for machine learning. At this time, the input data was the SNP locus information and sample labels (i.e., breed information) extracted from the raw format file of 939 beef cattle. The training set was trained using the SVM method using R language e1071. The running code was: model<- svm(x= x-train, y = y-train, kernel = "rbf", cost = 10, gamma = 0.1).

[0092] Use the trained machine learning model to predict the test set. The running code is: y_pred<- predict(model, x_test), and the accuracy of the prediction is evaluated through five-fold cross validation. The accuracy is determined by:

[0093] .

[0094] In the formula, is the number of individuals that can be accurately judged, is the number of individuals with incorrect judgments, For accuracy.

[0095] In this embodiment, the prediction results show that when the set selection threshold is TOP 2%, that is, the number of SNPs is 340, the accuracy of breed identification using SVM is greater than 99%, and it is used as the characteristic information site of the breed (Most breed-informative SNPs, MBI SNPs), and finally 340 MBI SNPs of 939 beef cattle are obtained.

[0096] S4: Construction of Molecular ID Cards for Beef Cattle Breeds.

[0097] In order to further reduce the number of MBI SNPs sites, a set of sites with the most characteristic information content was extracted as much as possible to construct a molecular ID card for beef cattle breeds. Based on this, the R language e1071 was used to search for the minimum SNPs set using the sequential forward selection (SFS) method, the maximum search step was set to 140, the parameter was selected as svmLinear, and the accuracy threshold was set to greater than 99% as the evaluation condition for the search. Finally, a set of 121 SNPs was constructed as the molecular ID card for the 29 beef cattle breeds selected in this embodiment, and the breed identification accuracy rate could reach 99%.

[0098] In an exemplary embodiment, a method for applying a molecular ID card of a beef cattle breed is provided, such as Figure 3 As shown, the method includes:

[0099] Step 200: Acquire the SNPs locus information of the beef cattle to be tested based on the molecular ID card of the beef cattle breed. The molecular ID card of the beef cattle breed is obtained using the method for constructing the molecular ID card of the beef cattle breed provided above.

[0100] Step 201: Obtain a breed detection model, and input SNPs site information into the breed detection model to obtain breed information of beef cattle to be detected.

[0101] As an optional implementation, the variety detection model is constructed based on a support vector machine.

[0102] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 4As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the construction or application data of the molecular identity card of beef cattle breeds. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for constructing or applying a molecular identity card of a beef cattle breed is implemented.

[0103] Those skilled in the art will understand that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components. In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0104] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0105] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0106] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0107] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0108] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0109] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0110] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for constructing a molecular ID card of a beef cattle breed, characterized in that: The method for constructing the molecular ID card of beef cattle breeds comprises: Step 100: Obtain sample data; the sample data includes beef cattle breeds and SNP site information corresponding to each beef cattle breed; Step 101: Divide the sample data into a training set and a test set; Step 102: using the training set to train the machine learning model, and during the training process, using the test set to test the trained machine learning model until the prediction accuracy of the trained machine learning model reaches a set value, and then using the trained machine learning model as a verification model; wherein the machine learning model is a support vector machine model; and the set value is 99%; Step 103: Using a feature filtering strategy, the specific SNP sites of different varieties are screened in the training set to obtain a specific SNP site set, including: using a feature filtering strategy, feature filtering the sample data in the training set, and in the feature filtering process, using a Delta method to screen the specific SNP sites of different varieties in the training set to obtain the specific SNP site set; wherein: (1) Determine the Delta value of each specific SNP site between each beef cattle breed in the sample data; (2) arranging the specific SNP loci of each beef cattle breed in descending order according to the Delta value of the SNP loci to obtain the locus sequence of each beef cattle breed; (3) retaining the top 1% to 10% of the specific SNP sites in the site sequence of each beef cattle breed to obtain the specific SNP site set; Step 104: using the verification model to verify the specific SNP site set, and using the sites in the verified specific SNP site set as variety characteristic information sites, including: Step 1041: using the beef cattle breed information corresponding to the specific SNP locus set as a sample label, and combining the specific SNP locus set to generate a first sample pair; Step 1042: using the verification model to determine the prediction accuracy of the first sample pair, and determining whether the prediction accuracy reaches the set value; Step 1043: when the prediction accuracy does not reach the set value, using a feature filtering strategy, re-screening specific SNP sites of different varieties in the training set to obtain a new set of specific SNP sites, and re-using the verification model to verify the new set of specific SNP sites until the prediction accuracy reaches the set value; Step 1044: when the prediction accuracy reaches the set value, the loci in the specific SNP loci set are used as variety characteristic information loci; Step 105: using a forward search method to search for a minimum SNPs set for the variety characteristic information loci; Step 106: using the verification model to verify the minimum SNPs set obtained by the search, and using the minimum SNPs set verified by the verification model as the molecular identity card of the beef cattle breed, including: Step 1061: Form a second sample pair with the minimum SNPs set and the sample label; Step 1062: using the verification model to determine the prediction accuracy of the second sample pair, and determining whether the prediction accuracy reaches the set value; Step 1063: when the prediction accuracy does not reach the set value, re-search the minimum SNPs set for the variety characteristic information loci to obtain a new minimum SNPs set, and re-use the verification model to verify the new minimum SNPs set until the prediction accuracy reaches the set value; Step 1064: When the prediction accuracy reaches the set value, the minimum SNPs set is used as the molecular identity card of the beef cattle breed.

2. A method for applying a molecular ID card of a beef cattle breed, characterized in that: The method for applying the molecular ID card of beef cattle breeds comprises: Based on the molecular ID card of the beef cattle breed, the SNPs site information of the beef cattle to be tested is obtained; the molecular ID card of the beef cattle breed is obtained by the method for constructing the molecular ID card of the beef cattle breed as claimed in claim 1; A breed detection model is obtained, and the SNPs site information is input into the breed detection model to obtain the breed information of the beef cattle to be detected; wherein the breed detection model is constructed based on a support vector machine.

3. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for constructing a molecular ID card of a beef cattle breed as described in claim 1, or the method for applying a molecular ID card of a beef cattle breed as described in claim 2.

4. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, it implements the method for constructing a molecular ID card of a beef cattle breed as described in claim 1, or the method for applying a molecular ID card of a beef cattle breed as described in claim 2.

Images