Method and system for identifying breed lineage of Chinese local cattle, and electronic equipment

Abstract

The invention relates to the technical field of cattle breed bloodline identification scheme design, in particular to a method and a system for identifying Chinese local cattle breed bloodline and electronic equipment. Comprising the following steps: extracting DNA in blood of a cattle to be detected; using a targeted sequencing genotyping technology to obtain genotypes of a preset number of SNP sites, and forming a genotype data set of the cattle to be detected; and comparing the genotype data set of the to-be-detected cattle with the reference SNP data set to obtain the blood line proportion of each variety of the to-be-detected cattle. By applying a convolutional neural network method, the breed lineage proportion of different reference cattle breeds (mainly Chinese local cattle breeds) contained in a to-be-detected sample is directly calculated, and the problem that the lineage proportion of each cattle breed in the Chinese local cattle breeds cannot be accurately identified in the prior art is solved; meanwhile, the intelligent degree, the accuracy and the reliability of identifying the blood lineage of the Chinese local cattle variety are improved to a great extent.

Description

Technical Field

[0001] The present invention relates to the technical field of cattle breed pedigree identification scheme design, and in particular to a method and system and electronic equipment for identifying the pedigree of Chinese local cattle breeds. Background Art

[0002] The purity of cattle's pedigree is directly related to production efficiency and the quality of offspring. For a long time, people have mainly judged cattle's pedigree by visually observing their appearance, body structure and even behavioral habits. Although this method is simple and easy, it is inevitably mixed with subjective assumptions and it is difficult to ensure the accuracy and objectivity of pedigree identification. As time goes by, the limitations of this traditional method have become increasingly prominent and cannot meet the increasingly sophisticated management needs of animal husbandry.

[0003] With the gradual establishment of genetic theory and the rapid development of molecular biology technology, cattle pedigree identification has ushered in a revolutionary change. The rise of DNA analysis technology, especially the application of high-precision genotyping technology such as microsatellite markers, has provided an unprecedented scientific basis for pedigree identification. By extracting DNA samples from cattle, users can deeply analyze their genetic information, and then use DNA fingerprinting technology, microsatellite markers, haplotype analysis and other methods to detect the genetic diversity and kinship between cattle, thereby realizing the identification of cattle pedigree. This process not only gets rid of the subjectivity and limitations of traditional methods, but also has certain advantages.

[0004] Using DNA information for feature extraction is the core computational method for pedigree analysis. Traditional methods are mainly pedigree analysis methods based on the maximum likelihood method (such as the software ADMIXTURE), which has several shortcomings. The first is that the assumptions about the model are relatively strict. The maximum likelihood method assumes that the extreme value in the loss function is the minimum value, which is not always true in practical applications. Second, the least squares method based on the maximum likelihood method is simple to calculate, but in some cases, such as solving nonlinear problems, it requires a huge amount of calculation, which also affects the accuracy of the calculation results. The existing technology is currently unable to accurately identify the proportion of the pedigree of each cattle breed in the local Chinese cattle breeds.

[0005] Therefore, the prior art needs to be further developed. Summary of the invention

[0006] The purpose of the present invention is to overcome the above technical deficiencies and provide a method and system and electronic equipment for identifying the pedigree of Chinese local cattle breeds to solve the problems existing in the prior art.

[0007] To achieve the above technical objectives, according to a first aspect of the present invention, the present invention provides a method for identifying the pedigree of Chinese local cattle breeds, the method comprising: S100, extracting DNA from the blood of the cattle to be tested; S200, using targeted sequencing genotyping technology to obtain genotypes of a preset number of SNP sites to form a genotype data set for the cattle to be tested; S300, using the genotype data set of the cattle to be tested, and comparing it with the reference SNP data set to obtain the pedigree ratio of each breed of the cattle to be tested.

[0008] Specifically, the genotype data set of the cattle to be tested is compared with the reference SNP data set to obtain the pedigree ratio of each breed of the cattle to be tested, including: The convolutional neural network method was used to detect the bloodline ratio of each breed of the tested cattle.

[0009] Specifically, the method of using a convolutional neural network to detect the pedigree ratio of each breed of the tested cattle includes: A convolutional neural network model for identifying the pedigree of Chinese local cattle breeds is constructed. The convolutional neural network model includes four parts. The first part is the input layer, which accepts the genotype matrix and breed name of a given individual. The matrix size of the genotype matrix is ​​1*P, where P is the number of markers. The second part consists of a convolutional layer and a pooling layer. The convolutional layer is used to filter the features of the input genotype. The output result of the convolutional layer is first activated by a linear rectification function ReLU and then passed to the pooling layer for feature compression. The third part is a fully connected layer. The output result of the third part is passed to the fourth part, which is the output layer. The fourth part uses a linear regression model to output the estimated genomic breeding value.

[0010] Specifically, the convolution layer includes 8 computing nodes, the convolution kernel size of each node is [1*18], and the stride is 1; the pooling layer uses the max pooling method for feature compression, the pooling size is [1*4], and the stride is 1.

[0011] Specifically, the method further includes: A dropout function is set after the pooling layer, and the rate is set to 0.2. The output result of the second part is passed to the fully connected layer through the dropout function.

[0012] Specifically, the fully connected layer includes two layers. The first layer has 32 computing nodes. The result of the first fully connected layer is activated by the ReLU activation function, and a dropout function is set with a rate of 0.1, and then passed to the second fully connected layer. The second fully connected layer has 1 computing node. The result of the second fully connected layer is passed to the output layer of the fourth part through a loss function with a rate of 0.05.

[0013] Specifically, the method further includes: In the process of convolutional neural network model training, for each cross-validation data set, the convolutional neural network model is trained on the training data set and validated on the validation data set. The back propagation algorithm is used to optimize the parameters of the convolutional neural network model. The number of iterations is set to 6000, the learning rate is set to 0.01, the weight momentum parameter is set to 0.5, the weight decay is set to 0.00001, and the decay function uses the mean absolute error: Among them, m represents the number of individuals in the training data set, predictk and obsk represent the predicted breed pedigree ratio value and the actual breed pedigree ratio value of the kth individual, and Mae represents the mean absolute error.

[0014] Specifically, the average standardized discounted cumulative value added was used to evaluate the performance of the convolutional neural network model in identifying the pedigree composition of cattle breeds, specifically: The performance of the convolutional neural network model and the linear model in selecting individuals with high breed ancestry ratio values ​​were compared respectively; the comparison method used the mean standardized discounted cumulative value, i.e., MNV, which evaluates the ability of the genomic selection model to select the highest-ranked k individuals. Assuming there are n individuals, the predicted and observed breed ancestry ratio values ​​form an n×2 score matrix (X, Y). The formula for MNV is as follows: where d(i) = 1 / (log2i + 1) is the monotonically decreasing discount function at position i; y(i, Y) is the i-th breed ancestry proportion value in descending order, i.e., y(1,Y)≥y(2,Y)≥…y(n,Y); y(i,X) is the Y value corresponding to the predicted breed ancestry proportion value X in descending order in (X,Y). When the observed breed ancestry proportion value is greater than 0, the MNV ranges from 0 to 1; the larger the MNV(k,X,Y), the better the breed ancestry proportion value of the high-ranked individuals selected by the genomic selection model. Given α=k / m, 1≤k≤m, 1%≤α≤100%, the entire 10-fold cross-validation experiment was repeated 10 times. Therefore, for each given α, the program produces 10 different MNV values, the average of which is used as the final result.

[0015] According to a second aspect of the present invention, there is provided a system for identifying the pedigree of Chinese local cattle breeds, comprising: An acquisition module is used to extract DNA from the blood of the cattle to be tested; The control module is used to obtain the genotypes of a preset number of SNP sites using targeted sequencing genotyping technology to form a genotype data set of the cattle to be tested; and is used to use the genotype data set of the cattle to be tested to compare with the reference SNP data set to obtain the pedigree ratio of each breed of the cattle to be tested.

[0016] According to a third aspect of the present invention, there is provided an electronic device, comprising: a memory; and a processor, wherein the memory stores computer-readable instructions, and the computer-readable instructions, when executed by the processor, implement the above-mentioned method for identifying the pedigree of Chinese local cattle breeds.

[0017] Beneficial effects: The present invention proposes a scheme for identifying the pedigree of Chinese local cattle breeds, including extracting DNA from the blood of the cattle to be tested; using targeted sequencing genotyping technology to obtain the genotypes of a preset number of SNP sites to form a genotype data set of the cattle to be tested; using the genotype data set of the cattle to be tested, and comparing it with the reference SNP data set to obtain the pedigree ratio of each breed of the cattle to be tested. By applying the convolutional neural network method, the pedigree ratio of different reference cattle breeds (mainly Chinese local cattle breeds) contained in the sample to be tested can be directly calculated, which solves the problem that the existing technology cannot accurately identify the pedigree ratio of each cattle breed in Chinese local cattle breeds, and at the same time greatly improves the intelligence, accuracy and reliability of identifying the pedigree of Chinese local cattle breeds. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is a flow chart of a method for identifying the pedigree of Chinese local cattle breeds provided in a specific embodiment of the present invention; Figure 2 Schematic diagram of the system composition of the system for identifying the pedigree of Chinese local cattle breeds provided in a specific embodiment of the present invention; Figure 3 It is a schematic diagram of the ancestry analysis results of the test samples provided in the specific embodiments of the present invention. DETAILED DESCRIPTION

[0019] In order to enable those skilled in the art to better understand the technical solution of the present invention, the technical solution of the present invention is clearly and completely described below in conjunction with the accompanying drawings of the present invention. Based on the embodiments in this application, other similar embodiments obtained by ordinary technicians in this field without making creative work should all fall within the scope of protection of this application. In addition, the directional words mentioned in the following embodiments, such as "up", "down", "left", "right", etc., are only reference to the directions of the accompanying drawings. Therefore, the directional words used are used to illustrate rather than limit the invention.

[0020] The present invention will be further described below in conjunction with the accompanying drawings and preferred embodiments.

[0021] See also Figure 1-Figure 3 The present invention provides a method for identifying the pedigree of Chinese local cattle breeds, comprising: S100, extracting DNA from the blood of the cattle to be tested.

[0022] Specifically, the present invention adopts a general method to extract DNA from the blood of the cattle to be tested.

[0023] It should be noted here that China's local cattle breeds include Mongolian cattle, Kazakh cattle, Yanbian cattle, Fuzhou cattle, Luxi cattle, Bohai black cattle, Qinchuan cattle, Leiqiong cattle, and Ji'an cattle.

[0024] S200, using targeted sequencing genotyping technology to obtain the genotypes of a preset number of SNP sites to form a genotype data set for the cattle to be tested.

[0025] Specifically, the genotype file format is shown in Table 1: Table 1 Genotype file format S300, using the genotype data set of the cattle to be tested, and comparing it with the reference SNP data set to obtain the pedigree ratio of each breed of the cattle to be tested.

[0026] Specifically, the pedigree ratios of various breeds of the cattle to be tested are shown in Table 2: Table 2 Proportion of each breed of cattle tested Specifically, the genotype data set of the cattle to be tested is compared with the reference SNP data set to obtain the pedigree ratio of each breed of the cattle to be tested, including: The convolutional neural network method was used to detect the bloodline ratio of each breed of the tested cattle.

[0027] Specifically, the method of using a convolutional neural network to detect the pedigree ratio of each breed of the tested cattle includes: A convolutional neural network model for identifying the pedigree of Chinese local cattle breeds is constructed. The convolutional neural network model includes four parts. The first part is the input layer, which accepts the genotype matrix and breed name of a given individual. The matrix size of the genotype matrix is ​​1*P, where P is the number of markers. The second part consists of a convolutional layer and a pooling layer. The convolutional layer is used to filter the features of the input genotype. The output result of the convolutional layer is first activated by a linear rectification function ReLU and then passed to the pooling layer for feature compression. The third part is a fully connected layer. The output result of the third part is passed to the fourth part, which is the output layer. The fourth part uses a linear regression model to output the estimated genomic breeding value.

[0028] Specifically, the convolution layer includes 8 computing nodes, the convolution kernel size of each node is [1*18], and the stride is 1; the pooling layer uses the max pooling method for feature compression, the pooling size is [1*4], and the stride is 1.

[0029] Specifically, the method further includes: A dropout function is set after the pooling layer, and the rate is set to 0.2. The output result of the second part is passed to the fully connected layer through the dropout function.

[0030] Specifically, the fully connected layer includes two layers. The first layer has 32 computing nodes. The result of the first fully connected layer is activated by the ReLU activation function, and a dropout function is set with a rate of 0.1, and then passed to the second fully connected layer. The second fully connected layer has 1 computing node. The result of the second fully connected layer is passed to the output layer of the fourth part through a loss function with a rate of 0.05.

[0031] Specifically, the method further includes: In the process of convolutional neural network model training, for each cross-validation data set, the convolutional neural network model is trained on the training data set and validated on the validation data set. The back propagation algorithm is used to optimize the parameters of the convolutional neural network model. The number of epochs is set to 6000, the learning rate is set to 0.01, the weight momentum parameter is set to 0.5, the weight decay is set to 0.00001, and the loss function uses the mean absolute error (Mae): Among them, m represents the number of individuals in the training data set, predictk and obsk represent the predicted breed pedigree ratio value and the actual breed pedigree ratio value of the kth individual, and the back propagation algorithm is stopped when the mean absolute error between the predicted breed pedigree ratio and the observed breed pedigree ratio is stable, and Mae represents the mean absolute error.

[0032] It should be noted here that, in the present invention, when the mean absolute value error no longer changes, the mean absolute value error is considered stable.

[0033] It should be noted that the present invention uses the deep learning framework MXNet (version 0.9.3, https: / / github.com / dmlc / mxnet) based on the graphics processing unit (GPU) to implement the convolutional neural network model to be constructed in this study. It runs on a GPU server equipped with an Nvidia Tesla P4 graphics card. The accuracy of variety identification is evaluated by reliability and unbiasedness.

[0034] Specifically, the average standardized discounted cumulative value added was used to evaluate the performance of the convolutional neural network model in identifying the pedigree composition of cattle breeds, specifically: The performance of the convolutional neural network model and the linear model in selecting individuals with high breed ancestry ratio values ​​were compared respectively; the comparison method used the mean normalized discounted cumulative gain value (MNV), that is, MNV, which evaluates the ability of the genomic selection model in selecting the highest-ranked k individuals. Assuming there are n individuals, the predicted and observed breed ancestry ratio values ​​form an n×2 score matrix (X, Y). The formula of MNV is as follows: where d(i) = 1 / (log2i + 1) is the monotonically decreasing discount function at position i; y(i, Y) is the i-th breed ancestry proportion value in descending order, i.e., y(1,Y)≥y(2,Y)≥…y(n,Y); y(i,X) is the Y value corresponding to the predicted breed ancestry proportion value X in descending order in (X,Y). When the observed breed ancestry proportion value is greater than 0, the MNV ranges from 0 to 1; the larger the MNV(k,X,Y), the better the breed ancestry proportion value of the high-ranked individuals selected by the genomic selection model. Given α=k / m, 1≤k≤m, 1%≤α≤100%, the entire 10-fold cross-validation experiment was repeated 10 times. Therefore, for each given α, the program produces 10 different MNV values, the average of which is used as the final result.

[0035] The present invention will be further described below by means of specific examples: First, the SNP genotype data of existing purebred beef cattle are used to form a reference SNP data set, which is integrated with the SNP genotype data of the cattle to be tested, and the pedigree ratio of each breed of the cattle to be tested is detected based on the convolutional neural network method.

[0036] 2. Sample Information A total of 6 blood samples were received for this experiment, and the sample information is as follows: 3. Pedigree Analysis Results Using Angus, Charolais, Hereford, Holstein, Limousin, Simmental, and Wagyu as reference groups, the pedigree of the test sample (indicated by "this sample" in the figure) was analyzed, and the results of two test cases were added for comparison. Figure 3 As shown: Figure 3 Different colors represent different cattle breeds and bloodlines. Purple represents Angus cattle breed, blue represents Charolais cattle breed, green and black represent Hereford cattle breed, pink represents Holstein cattle breed, green represents Limousin cattle breed, khaki represents Simmental cattle breed, and brown represents Wagyu cattle breed.

[0037] from Figure 3 It can be seen that all the tested cattle have Wagyu pedigree, generally more than 50%, and other pedigrees are also detected in different cattle. The proportion of cattle breed pedigree contained in each tested cattle is shown in Table 3.

[0038] Table 3 Proportion of breed pedigrees in the tested cattle It is understandable that in today's rapidly developing livestock industry, pedigree identification, as a core technology, plays an irreplaceable role in ensuring the genetic quality of cattle, promoting breed improvement, increasing market value and preventing genetic diseases. It is not only the cornerstone of genetic breeding work, but also an important driving force for the sustainable development of the entire industry.

[0039] Pedigree identification uses scientific means to trace the genetic information of cattle ancestors and accurately evaluate the genetic potential of their offspring. This process not only helps to identify and retain individuals with excellent genetic characteristics, but also provides a scientific basis for livestock managers to pay more attention to the stability and consistency of genetic quality during the breeding process.

[0040] In the process of cattle breed improvement, pedigree identification plays a vital role. By accurately dividing the pedigree of cattle, the genetic differences and connections between breeds can be clearly identified, providing data support for the formulation of scientific breeding plans. For example, when formulating a hybrid breeding plan, the clear blood relationship enables breeders to reasonably select paired individuals, avoid genetic decline caused by inbreeding, and promote the inheritance and spread of excellent traits. This process not only improves the production performance of cattle, but also promotes the optimization and upgrading of the entire livestock breed structure.

[0041] It should be further explained that the fundamental basis of pedigree identification lies in the transmission and expression mechanism of genetic information in genetics. Each beef cattle breed carries its own unique genetic information, which is passed from generation to generation through the breeding process, shaping the genetic characteristics of the offspring. Therefore, by analyzing the genetic information of cattle, the genetic background of their ancestors can be traced, thereby clarifying their pedigree.

[0042] It is understandable that the present invention proposes a scheme for identifying the pedigree of Chinese local cattle breeds, including extracting DNA from the blood of the cattle to be tested; using targeted sequencing genotyping technology to obtain the genotypes of a preset number of SNP sites to form a genotype data set of the cattle to be tested; using the genotype data set of the cattle to be tested, and comparing it with the reference SNP data set to obtain the pedigree ratio of each breed of the cattle to be tested. By applying the convolutional neural network method, the pedigree ratio of different reference cattle breeds (mainly Chinese local cattle breeds) contained in the sample to be tested can be directly calculated, which solves the problem that the existing technology cannot accurately identify the pedigree ratio of each cattle breed in the Chinese local cattle breeds, and at the same time greatly improves the intelligence, accuracy and reliability of identifying the pedigree of Chinese local cattle breeds.

[0043] See also Figure 2 The present invention provides another embodiment, which provides a system for identifying the pedigree of Chinese local cattle breeds, and the system for identifying the pedigree of Chinese local cattle breeds comprises: An acquisition module 100 is used to extract DNA from the blood of the cattle to be tested; The control module 200 is used to obtain the genotypes of a preset number of SNP sites using targeted sequencing genotyping technology to form a genotype data set of the cattle to be tested; and is used to use the genotype data set of the cattle to be tested to compare with the reference SNP data set to obtain the pedigree ratio of each breed of the cattle to be tested.

[0044] It should be noted that the present invention proposes a scheme for identifying the pedigree of Chinese local cattle breeds, including extracting DNA from the blood of the cattle to be tested; using targeted sequencing genotyping technology to obtain the genotypes of a preset number of SNP sites to form a genotype data set of the cattle to be tested; using the genotype data set of the cattle to be tested, and comparing it with the reference SNP data set to obtain the pedigree ratio of each breed of the cattle to be tested. By applying the convolutional neural network method, the pedigree ratio of different reference cattle breeds (mainly Chinese local cattle breeds) contained in the sample to be tested can be directly calculated, which solves the problem that the existing technology cannot accurately identify the pedigree ratio of each cattle breed in Chinese local cattle breeds, and at the same time greatly improves the intelligence, accuracy and reliability of identifying the pedigree of Chinese local cattle breeds.

[0045] In a preferred embodiment, the present application further provides an electronic device, the electronic device comprising: A memory; and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the method for identifying the pedigree of Chinese local cattle breeds is implemented. The computer device can be broadly a server, a terminal, or any other electronic device with necessary computing and / or processing capabilities. In one embodiment, the computer device may include a processor, a memory, a network interface, a communication interface, etc. connected via a system bus. The processor of the computer device can be used to provide necessary computing, processing and / or control capabilities. The memory of the computer device may include a non-volatile storage medium and an internal memory. An operating system, a computer program, etc. may be stored in or on the non-volatile storage medium. The internal memory can provide an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface and the communication interface of the computer device can be used to connect and communicate with external devices via a network. When the computer program is executed by the processor, the steps of the method of the present invention are executed.

[0046] The present invention may be implemented as a computer-readable storage medium having a computer program stored thereon, which causes the steps of the method of an embodiment of the present invention to be executed when executed by a processor. In one embodiment, the computer program is distributed on a plurality of computer devices or processors coupled to a network so that the computer program is stored, accessed and executed in a distributed manner by one or more computer devices or processors. A single method step / operation, or two or more method steps / operations, may be performed by a single computer device or processor or by two or more computer devices or processors. One or more method steps / operations may be performed by one or more computer devices or processors, and one or more other method steps / operations may be performed by one or more other computer devices or processors. One or more computer devices or processors may perform a single method step / operation, or perform two or more method steps / operations.

[0047] It will be appreciated by a person skilled in the art that the method steps of the present invention may be performed by instructing related hardware such as a computer device or a processor through a computer program, and the computer program may be stored in a non-temporary computer-readable storage medium, which causes the steps of the present invention to be performed when the computer program is executed. Depending on the circumstances, any reference to memory, storage, database, or other media herein may include non-volatile and / or volatile memory. Examples of non-volatile memory include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, magnetic tape, floppy disk, magneto-optical data storage device, optical data storage device, hard disk, solid state disk, etc. Examples of volatile memory include random access memory (RAM), external cache memory, etc.

[0048] It is understandable that the present invention proposes a scheme for identifying the pedigree of Chinese local cattle breeds, including extracting DNA from the blood of the cattle to be tested; using targeted sequencing genotyping technology to obtain the genotypes of a preset number of SNP sites to form a genotype data set of the cattle to be tested; using the genotype data set of the cattle to be tested, and comparing it with the reference SNP data set to obtain the pedigree ratio of each breed of the cattle to be tested. By applying the convolutional neural network method, the pedigree ratio of different reference cattle breeds (mainly Chinese local cattle breeds) contained in the sample to be tested can be directly calculated, which solves the problem that the existing technology cannot accurately identify the pedigree ratio of each cattle breed in the Chinese local cattle breeds, and at the same time greatly improves the intelligence, accuracy and reliability of identifying the pedigree of Chinese local cattle breeds.

[0049] The various technical features described above can be combined arbitrarily. Although all possible combinations of these technical features are not described, any combination of these technical features should be considered to be covered by this specification as long as there is no contradiction in such combination.

[0050] The specific implementation of the present invention described above does not constitute a limitation on the protection scope of the present invention. Any other corresponding changes and modifications made based on the technical concept of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A method for identifying the pedigree of Chinese local cattle breeds, characterized in that: The method comprises: S100, extracting DNA from the blood of the cattle to be tested; S200, using targeted sequencing genotyping technology to obtain genotypes of a preset number of SNP sites to form a genotype data set for the cattle to be tested; S300, using the genotype data set of the cattle to be tested, and comparing it with the reference SNP data set to obtain the pedigree ratio of each breed of the cattle to be tested.

2. The method for identifying the pedigree of Chinese local cattle breeds according to claim 1, characterized in that: The genotype data set of the cattle to be tested is compared with the reference SNP data set to obtain the pedigree ratio of each breed of the cattle to be tested, including: The convolutional neural network method was used to detect the bloodline ratio of each breed of the tested cattle.

3. The method for identifying the pedigree of Chinese local cattle breeds according to claim 2, characterized in that: The method of using a convolutional neural network to detect the pedigree ratio of each breed of the tested cattle includes: A convolutional neural network model for identifying the pedigree of Chinese local cattle breeds is constructed. The convolutional neural network model includes four parts. The first part is the input layer, which accepts the genotype matrix and breed name of a given individual. The matrix size of the genotype matrix is ​​1*P, where P is the number of markers. The second part consists of a convolutional layer and a pooling layer. The convolutional layer is used to filter the features of the input genotype. The output result of the convolutional layer is first activated by a linear rectification function ReLU and then passed to the pooling layer for feature compression. The third part is a fully connected layer. The output result of the third part is passed to the fourth part, which is the output layer. The fourth part uses a linear regression model to output the estimated genomic breeding value.

4. The method for identifying the pedigree of Chinese local cattle breeds according to claim 3, characterized in that: The convolution layer includes 8 computing nodes, the convolution kernel size of each node is [1*18], and the stride is 1; the pooling layer uses the maxpooling method for feature compression, the pooling size is [1*4], and the stride is 1.

5. The method for identifying the pedigree of Chinese local cattle breeds according to claim 3, characterized in that: The method further comprises: A dropout function is set after the pooling layer, and the rate is set to 0.

2. The output result of the second part is passed to the fully connected layer through the dropout function.

6. The method for identifying the pedigree of Chinese local cattle breeds according to claim 3, characterized in that: The fully connected layer consists of two layers. The first layer has 32 computing nodes. The result of the first fully connected layer is activated by the ReLU activation function, and a dropout function is set with a rate of 0.1, and then passed to the second fully connected layer. The second fully connected layer has 1 computing node. The result of the second fully connected layer is passed to the output layer of the fourth part through a loss function with a rate of 0.

05.

7. The method for identifying the pedigree of Chinese local cattle breeds according to claim 3, characterized in that: The method further comprises: In the process of convolutional neural network model training, for each cross-validation data set, the convolutional neural network model is trained on the training data set and validated on the validation data set. The back propagation algorithm is used to optimize the parameters of the convolutional neural network model. The number of iterations is set to 6000, the learning rate is set to 0.01, the weight momentum parameter is set to 0.5, the weight decay is set to 0.00001, and the decay function uses the mean absolute error: Where m represents the number of individuals in the training data set, predictk and obsk represent the predicted breed pedigree ratio and the actual breed pedigree ratio of the kth individual, and Mae represents the mean absolute error.

8. The method for identifying the pedigree of Chinese local cattle breeds according to claim 7, characterized in that: The average standardized discounted cumulative value added was used to evaluate the performance of the convolutional neural network model in identifying the pedigree composition of cattle breeds, specifically: The performance of the convolutional neural network model and the linear model in identifying the breed pedigree ratio were compared respectively; the comparison method used the mean standardized discounted cumulative value, i.e., MNV. MNV evaluates the ability of the convolutional neural network model in identifying the breed pedigree ratio. Assuming that there are n individuals, the predicted and observed breed pedigree ratio values ​​form an n×2 score matrix (X, Y). The formula of MNV is as follows: where d(i) = 1 / (log2i + 1) is the monotonically decreasing discount function at position i; y(i, Y) is the i-th breed pedigree ratio value in descending order, i.e., y(1,Y)≥y(2,Y)≥…y(n,Y); y(i,X) is the Y value corresponding to the breed pedigree ratio value X in descending order in (X,Y). When the observed breed pedigree ratio value is greater than 0, the MNV ranges from 0 to 1; the larger the MNV(k,X,Y), the higher the breed pedigree ratio content of the highly ranked individuals selected by the convolutional neural network model. Given α=k / m, 1≤k≤m, 1%≤α≤100%, the entire 10-fold cross-validation experiment was repeated 10 times. Therefore, for each given α, the program produces 10 different MNV values, the average of which is used as the final result.

9. A system for identifying the pedigree of Chinese local cattle breeds, characterized in that: include: An acquisition module is used to extract DNA from the blood of the cattle to be tested; The control module is used to obtain the genotypes of a preset number of SNP sites using targeted sequencing genotyping technology to form a genotype data set of the cattle to be tested; and is used to use the genotype data set of the cattle to be tested to compare with the reference SNP data set to obtain the pedigree ratio of each breed of the cattle to be tested.

10. An electronic device, characterized in that: include: Memory; and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the method for identifying the pedigree of Chinese local cattle breeds according to any one of claims 1 to 8 is implemented.

Images