Data processing method and apparatus

Abstract

Embodiments of the present application relate to the technical field of data processing, and disclose a data processing method and apparatus, capable of screening for core SNPs having high genetic characteristics from SNPs of organisms. The method comprises: acquiring a first SNP set; determining a second SNP set from the first SNP set according to first information of SNPs; determining a third SNP set from the second SNP set according to a target feature extraction method and a nonlinear feature extraction method, wherein the first SNP set comprises N SNPs, N is a positive integer, the first information comprises at least one of linkage disequilibrium information or variance, the second SNP set comprises M SNPs, M is a positive integer, M is smaller than N, the target feature extraction method comprises at least one of a filter feature extraction method, a wrapper feature extraction method or an embedded feature extraction method, the third SNP set comprises H SNPs, H is a positive integer, and H is smaller than M.

Description

Data processing method and device

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on June 16, 2023, with application number 202310727647.3 and application name “Data Processing Method and Device”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The embodiments of the present application relate to the field of data processing technology, and in particular to a data processing method and apparatus. Background Art

[0003] Single nucleotide polymorphism (SNP) mainly refers to DNA sequence polymorphism caused by the variation of a single nucleotide at the genomic level. It is the most common type of heritable variation in organisms, accounting for more than 90% of all known polymorphisms. SNPs are widely present in the genome of organisms, with an average of 1 in every 300 base pairs, and the total number is estimated to be 3 million or even more. SNP is a dimorphic marker caused by the conversion or transversion of a single base, or by the insertion or deletion of a base. SNPs can be within the gene sequence or in non-coding sequences outside the gene. Genetic screening requires screening out highly heritable core SNPs from the SNPs of an organism.

[0004] However, among the tens of thousands of SNPs in organisms, only a small number are core SNPs with high genetic characteristics.

[0005] Therefore, how to screen out core SNPs with high genetic characteristics from biological SNPs is one of the problems that technicians in this field urgently need to solve.

[0006] Summary of the Invention

[0007] The present invention provides a data processing method and apparatus, which relates to the field of data processing technology and can screen out core SNPs with high genetic characteristics from biological SNPs. To achieve the above purpose, the present invention adopts the following technical solutions:

[0008] In a first aspect, an embodiment of the present application provides a data processing method, which includes: obtaining a first SNP set. Determining a second SNP set from the above-mentioned first SNP set based on the first information of the SNP. Determining a third SNP set from the above-mentioned second SNP set based on a target feature extraction method and a nonlinear feature extraction method. The above-mentioned first SNP set includes N SNPs, the above-mentioned N is a positive integer, the above-mentioned first information includes at least one of linkage disequilibrium information or variance, the above-mentioned second SNP set includes M SNPs, the above-mentioned M is a positive integer, the above-mentioned M is less than the above-mentioned N, the above-mentioned target feature extraction method includes at least one of a filtering feature extraction method, a wrapper feature extraction method or an embedded feature extraction method, and the above-mentioned third SNP set includes H SNPs, the above-mentioned H is a positive integer, and the above-mentioned H is less than the above-mentioned M.

[0009] It should be noted that among the multiple SNPs in a region, there may be one SNP that can represent the type of most SNPs in the region. The embodiment of the present application determines the representative SNPs in each region from the multiple SNPs obtained through the variance and linkage disequilibrium information of the SNPs, eliminates the remaining redundant SNPs, and thus performs data dimensionality reduction on the obtained SNP data. The target feature extraction method and the nonlinear feature extraction method are then used to perform feature screening on the reduced data, thereby screening out core SNPs with high genetic characteristics from the biological SNPs. In addition, the embodiment of the present application also selects SNPs with strong correlation with the dependent variable (phenotype) from the multiple SNPs obtained through the variance and linkage disequilibrium information of the SNPs.

[0010] Optionally, the linkage disequilibrium information may include a linkage disequilibrium (LD) coefficient.

[0011] In a possible implementation, the target SNP may be deleted from the first SNP set according to the LD coefficient of the SNP to obtain the second SNP set, where the target SNP is a SNP whose LD coefficient is outside a preset LD coefficient range.

[0012] It is understood that if the LD coefficient of a SNP is outside the preset LD coefficient range, the SNP has a high probability of being in a state of linkage disequilibrium, and a SNP in a state of linkage disequilibrium is more likely to have the same genetic characteristics as other SNPs. Therefore, the acquired SNP data can be reduced in dimension by eliminating the target SNP, and then the reduced data can be subjected to feature screening using target feature extraction methods and nonlinear feature extraction methods, thereby screening core SNPs with high genetic characteristics from the biological SNPs.

[0013] In a possible implementation, when there are multiple SNPs with the same variance in the first SNP set, one SNP from the multiple SNPs with the same variance is retained and the other SNPs from the multiple SNPs are deleted to obtain the second SNP set.

[0014] It is understandable that SNPs with the same variance may also have the same genetic characteristics. Therefore, only one SNP among multiple SNPs with the same variance can be retained, and the remaining SNPs among the multiple SNPs with the same variance can be deleted from the acquired SNP data to perform data dimensionality reduction on the acquired SNP data. Feature screening can then be performed on the reduced dimensionality data using target feature extraction methods and nonlinear feature extraction methods, thereby screening core SNPs with high genetic characteristics from the biological SNPs.

[0015] In one possible implementation, a fourth SNP set may be determined from the second SNP set according to the target feature extraction method, the fourth SNP set including K SNPs, where K is a positive integer, less than M and greater than H. A third SNP set may be determined from the fourth SNP set according to the nonlinear feature extraction method.

[0016] It can be seen that the embodiment of the present application determines the representative SNPs in each region from the obtained multiple SNPs through the variance and linkage disequilibrium information of SNPs, eliminates the remaining redundant SNPs, thereby performing data dimensionality reduction on the obtained SNP data, and then performs feature screening on the reduced data through the target feature extraction method and the nonlinear feature extraction method, thereby screening out core SNPs with high genetic characteristics from the biological SNPs.

[0017] In one possible implementation, SNPs can be eliminated from the second SNP set according to the target feature extraction method. If the coefficient of determination of the remaining SNPs in the second SNP set is greater than a first threshold or the first remaining ratio is less than the first ratio threshold, the remaining SNPs in the second SNP set are determined as the fourth SNP set, and the first remaining ratio is the ratio of the number of SNPs remaining in the second SNP set to M.

[0018] It can be seen that the method provided in the embodiment of the present application determines the representative SNPs in each region from the obtained multiple SNPs through the variance and linkage disequilibrium information of the SNPs, eliminates the remaining redundant SNPs, thereby performing data dimensionality reduction on the obtained SNP data, and then further screening the obtained multiple SNPs according to the determination coefficient and number of the remaining SNPs through the target feature extraction method. The further screened SNPs are then feature-screened using nonlinear features, thereby screening core SNPs with high genetic characteristics from the biological SNPs. In addition, by further screening the obtained multiple SNPs according to the determination coefficient and number of the remaining SNPs through the target feature extraction method, SNPs with penalty terms for correlation between features can be obtained.

[0019] In one possible implementation, SNPs can be eliminated from the fourth SNP set according to the nonlinear feature extraction method; when the determination coefficient of the remaining SNPs in the fourth SNP set is greater than the second threshold or the second remaining ratio is less than the second ratio threshold, the remaining SNPs in the fourth SNP set are determined as the third SNP set, and the second remaining ratio is the ratio of the number of remaining SNPs in the fourth SNP set to the M.

[0020] It can be seen that the method provided in the embodiment of the present application determines the representative SNPs in each region from the multiple SNPs obtained through the variance and linkage disequilibrium information of SNPs, eliminates the remaining redundant SNPs, thereby performing data dimensionality reduction on the obtained SNP data, and then further screens the multiple SNPs obtained through the target feature extraction method. The SNPs after further screening are further screened from the multiple SNPs obtained according to the determination coefficient and number of the remaining SNPs through nonlinear characteristics, thereby screening out core SNPs with high genetic characteristics from the SNPs of the organism.

[0021] In a second aspect, an embodiment of the present application provides a data processing device, which includes: a transceiver unit and a processing unit. The transceiver unit is used to obtain a first SNP set, wherein the first SNP set includes N SNPs, and N is a positive integer. The processing unit is used to determine a second SNP set from the first SNP set based on the first information of the SNP, wherein the first information includes at least one of linkage disequilibrium information or variance, and the second SNP set includes M SNPs, wherein M is a positive integer, and M is less than N. The processing unit is also used to determine a third SNP set from the second SNP set based on a target feature extraction method and a nonlinear feature extraction method, wherein the target feature extraction method includes at least one of a filtering feature extraction method, a wrapper feature extraction method, or an embedded feature extraction method, and the third SNP set includes H SNPs, wherein H is a positive integer, and H is less than M.

[0022] In one possible implementation, the first information includes linkage disequilibrium information, the linkage disequilibrium information includes the linkage disequilibrium LD coefficient, and the processing unit is specifically used to: delete the target SNP from the first SNP set according to the LD coefficient of the SNP to obtain the second SNP set, and the target SNP is a SNP whose LD coefficient is outside the preset LD coefficient range.

[0023] In one possible implementation, the first information includes variance, and the processing unit is specifically used to: when there are multiple SNPs with the same variance in the first SNP set, retain one SNP among the multiple SNPs with the same variance and delete the other SNPs among the multiple SNPs to obtain the second SNP set.

[0024] In one possible implementation, the processing unit is specifically used to: determine a fourth SNP set from the second SNP set according to the target feature extraction method, the fourth SNP set including K SNPs, K being a positive integer, K being less than M and greater than H; and determine a third SNP set from the fourth SNP set according to the nonlinear feature extraction method.

[0025] In one possible implementation, the processing unit is specifically used to: eliminate SNPs from the second SNP set according to the target feature extraction method; when the determination coefficient of the remaining SNPs in the second SNP set is greater than the first threshold or the first remaining ratio is less than the first ratio threshold, the remaining SNPs in the second SNP set are determined as the fourth SNP set, and the first remaining ratio is the ratio of the number of remaining SNPs in the second SNP set to the M.

[0026] In a possible implementation, the processing unit is specifically used to: eliminate SNPs from the fourth SNP set according to the nonlinear feature extraction method; when the determination coefficient of the remaining SNPs in the fourth SNP set is greater than the second threshold or the second remaining ratio is less than the second ratio threshold, determine the remaining SNPs in the fourth SNP set as the third SNP set, and the second remaining ratio is the ratio of the number of remaining SNPs in the fourth SNP set to the M.

[0027] In a third aspect, an embodiment of the present application further provides a data processing device, which includes: at least one processor, which, when the at least one processor executes program code or instructions, implements the method described in the above first aspect or any possible implementation thereof.

[0028] Optionally, the data processing device may further include at least one memory, and the at least one memory is used to store the program code or instruction.

[0029] In a fourth aspect, embodiments of the present application further provide a chip comprising: an input interface, an output interface, and at least one processor. Optionally, the chip further comprises a memory. The at least one processor is configured to execute code in the memory. When the at least one processor executes the code, the chip implements the method described in the first aspect or any possible implementation thereof.

[0030] Optionally, the chip may also be an integrated circuit.

[0031] In a fifth aspect, an embodiment of the present application further provides a computer-readable storage medium for storing a computer program, which includes methods for implementing the above-mentioned first aspect or any possible implementation thereof.

[0032] In a sixth aspect, an embodiment of the present application further provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to implement the method described in the first aspect or any possible implementation thereof.

[0033] The data processing device, computer storage medium, computer program product and chip provided in this embodiment are all used to execute the method provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the method provided above and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following is a brief introduction to the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0035] FIG1 is a schematic diagram of the structure of a data processing system provided in an embodiment of the present application;

[0036] FIG2 is a flow chart of a data processing method provided in an embodiment of the present application;

[0037] FIG3 is a schematic structural diagram of a data processing device provided in an embodiment of the present application;

[0038] FIG4 is a schematic diagram of the structure of a chip provided in an embodiment of the present application;

[0039] FIG5 is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application;

[0040] FIG6 is a schematic structural diagram of another data processing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the embodiments of this application.

[0042] The term "and / or" in this article is merely a description of the association relationship between associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone.

[0043] The terms "first" and "second" and so on in the description and drawings of the embodiments of this application are used to distinguish different objects, or to distinguish different processing of the same object, rather than to describe a specific order of objects.

[0044] Furthermore, the terms "including," "having," and any variations thereof, mentioned in the description of the embodiments of the present application are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or units is not limited to the listed steps or units but may optionally include other steps or units not listed, or may optionally include other steps or units inherent to the process, method, product, or apparatus.

[0045] It should be noted that in the description of the embodiments of this application, words such as "exemplarily" or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "exemplarily" or "for example" in the embodiments of this application should not be interpreted as having priority or advantage over other embodiments or designs. Rather, the use of words such as "exemplarily" or "for example" is intended to present the relevant concepts in a concrete manner.

[0046] First, the terms involved in the embodiments of the present application are explained.

[0047] High-dimensional small sample data: refers to data with a very high data dimension, a very small absolute number of samples, or a number of samples far smaller than the data dimension.

[0048] Linkage disequilibrium refers to the probability that alleles belonging to two or more gene loci co-occur on a single chromosome, which is higher than the chance of random occurrence. Simply put, as long as two genes are not inherited completely independently, they will exhibit some degree of linkage. This condition is called linkage disequilibrium. Linkage disequilibrium can occur in different regions of the same chromosome or on different chromosomes.

[0049] An allele generally refers to a pair of genes that control opposing traits at the same position on a pair of homologous chromosomes. It can occur in one of two or more genes at a specific locus on a chromosome. If a gene at a locus exists in two or more states, it is called a "multiple allele." If both alleles in a pair are identical, the individual is homozygous for the trait; if the two alleles are different, the individual is heterozygous for the trait. In heterozygous pairs, the dominant allele prevents the recessive allele from expressing the trait.

[0050] Linkage disequilibrium (LD) decays. Generally speaking, the closer two SNPs (SNP sites) are on the genome, the stronger the correlation and the larger the LD coefficient. Conversely, the smaller the LD coefficient. In other words, as the distance between sites increases, the LD coefficient usually decreases slowly. This law is usually presented using an LD decay graph. The decay rate of the LD coefficient varies in different subpopulations. Different populations of the same species have very different LD decay rates due to their different genetic backgrounds. Domestication selection will lead to a decrease in population genetic diversity and an increase in the correlation (linkage degree) between sites. Therefore, the higher the degree of domestication and the greater the intensity of selection, the slower the LD decay rate. For example, cultivated rice usually has a larger LD decay distance than wild rice. Similarly, a decrease in population genetic diversity caused by natural selection and genetic drift will also slow down the rate of LD decay.

[0051] The D value is the basic unit of LD, measuring the deviation of the observed haplotype frequency from the expected frequency under equilibrium. Although D can well express the basic meaning of LD, it is not suitable for expressing the actual LD ​​strength because it is strictly dependent on allele frequency, especially for comparing LD values ​​from different studies. Among the several commonly used symbols for measuring LD, the most important are D' and r 2 , both are based on D, each has its own characteristics and uses.

[0052] Filtering feature extraction methods involve first selecting features from a dataset and then training a learner. The feature selection process is independent of the subsequent learner, effectively filtering the initial features before training the model. Filtering feature extraction methods include those that remove low-variance features and those that select univariate features.

[0053] Filtering feature extraction method to remove low variance: Assume that the eigenvalues ​​of a certain feature are only 0 and 1, and in all input samples, 95% of the instances have the value of 1, then it can be considered that this feature is not very useful. If 100% are 1, then this feature is meaningless. This method can only be used when the eigenvalues ​​are all discrete variables. If they are continuous variables, the continuous variables need to be discretized before they can be used. In reality, there are generally not many features that have a certain value more than 95%, so although this method is simple, it is not very easy to use. It can be used as a preprocessing for feature selection, first remove those features with small value changes, and then select the appropriate feature selection method mentioned below for further feature selection.

[0054] Univariate feature selection (univariate feature selection) is a filtering feature extraction method. The principle of univariate feature selection is to calculate a statistical indicator for each variable separately, determine which indicators are important based on this indicator, and eliminate those that are not. Specifically, univariate feature selection can use the chi-square test for classification problems. For regression problems, univariate feature selection can use the Pearson correlation coefficient.

[0055] The Pearson correlation coefficient is the simplest method that can help understand the relationship between features and response variables. This method measures the linear correlation between variables. The result range is [-1, 1], where -1 indicates a complete negative correlation, +1 indicates a complete positive correlation, and 0 indicates no linear correlation.

[0056] Wraparound feature extraction methods involve continuously selecting feature subsets from an initial feature set, training a learner, and evaluating the subsets based on the learner's performance until the optimal subset is selected. Within wraparound feature extraction methods, the most readily apparent approach to searching for feature subsets is exhaustive search. Alternatively, a randomized subset search strategy (Las Vegas rapper, LVW) can be used within the Las Vegas method framework. However, because the LVW algorithm uses a randomized strategy for feature subset search, each feature subset evaluation requires training the learner, which is computationally expensive. If the initial number of features is large, the algorithm may run for a long time before reaching the stopping condition. If there is a runtime limit, it may not be able to provide a solution. Therefore, greedy algorithms can also be used, such as forward search (which gradually adds features to the optimal subset until adding features no longer improves model performance), backward search, and bidirectional search (which combines forward and backward search).

[0057] Embedded feature extraction methods: In filtering and wrapper feature selection methods, the feature selection process is distinct from the learner training process. Embedded feature selection, however, automatically performs feature selection during the learner training process. The most commonly used embedded feature extraction methods are L1 and L2 regularization. The larger the regularization term, the simpler the model and the smaller the coefficient. When the regularization term increases to a certain level, all feature coefficients tend to zero. During this process, the coefficients of some features will initially become zero, thus completing the feature selection process. Logistic regression, linear regression, and decision trees can all be used as base learners for regularized feature selection. Only algorithms that can obtain feature coefficients or feature importance can be used as base learners for embedded selection.

[0058] The Hilbert-Schmidt Independence Criterion (HSIC) regression model (Lasso) feature extraction method is a kernel-based feature selection algorithm that selects important features to explain the HSIC Lasso prediction based on the weights of different features in the HSIC Lasso. The form of the HSIC Lasso is as follows:

[0059] Where α=[α1,α2,…,α d ] T is the feature weight, λ is the regularization parameter, is the kth eigenvalue in the data, and y is the true value.

[0060] Ridge regression is a biased estimation regression method specially used for collinear data analysis. In essence, it is a modified least squares estimation method. By abandoning the unbiasedness of the least squares method, it obtains a more realistic and reliable regression coefficient at the cost of losing some information and reducing precision. It fits pathological data better than the least squares method.

[0061] The characteristic of lasso regression is that when establishing a generalized linear model, the generalized linear model includes one-dimensional continuous dependent variables, multidimensional continuous dependent variables, non-negative frequency dependent variables, binary discrete dependent variables, and multivariate discrete dependent variables. In addition, whether the dependent variable is continuous or discrete, lasso can handle it. In general, lasso has extremely low requirements for data, so it is widely used; in addition, lasso can also screen variables and reduce the complexity of the model. Variable screening here means not putting all variables into the model for fitting, but selectively putting variables into the model to obtain better performance parameters. Complexity adjustment refers to controlling the complexity of the model through a series of parameters to avoid overfitting. For linear models, complexity is directly related to the number of variables in the model. The more variables there are, the higher the model complexity. More variables can often give a seemingly better model when fitting, but at the same time they also face the risk of overfitting.

[0062] Elastic net regularization is a method that combines L1 regularization and L2 regularization. It can control the complexity and sparsity of the model and avoid the instability that L1 regularization may cause in some cases.

[0063] L1 regularization is a widely used regularization method that penalizes the size of the parameters by introducing the L1 norm of the model parameters in the loss function. This regularization method can be used for feature selection because it tends to set some parameters to zero, making the model more sparse.

[0064] L2 regularization is another common regularization method that penalizes the size of the parameters by introducing the L2 norm of the model parameters in the loss function. Unlike L1 regularization, L2 regularization does not tend to set the parameters to zero, but instead controls the complexity of the model by reducing the size of the parameters.

[0065] Support vector regression (SVR) is a supervised learning algorithm used to predict reunion values. It uses a similar principle to support vector machines (SVMs). The fundamental idea behind SVR is to find the best-fit line. In SVR, the best-fit line is the hypercube with the largest number of points.

[0066] Light gradient boosting machine (light GBM) is a type of boosting method. Boosting involves using a series of linear combinations of models to complete a model task. In boosting learning, each model is gradually determined, and each sub-model is superimposed into a composite model. This process ensures that the loss function gradually decreases as the number of sub-models increases.

[0067] Random Forest is an algorithm that integrates multiple trees using the concept of ensemble learning. Its basic unit is the decision tree, and its essence belongs to the ensemble learning method, a major branch of machine learning. The name Random Forest contains two key words: "random" and "forest." The word "forest" is easy to understand: if one tree is called a tree, then hundreds or thousands of trees can be called a forest. This metaphor is quite apt and embodies the key concept of Random Forest: ensemble learning. Intuitively, each decision tree is a classifier (assuming the problem is classification). For an input sample, N trees will produce N classification results. Random Forest, on the other hand, integrates all the classification votes and assigns the category with the most votes as the final output.

[0068] The Pearson correlation coefficient (PCC) is a method to measure the degree of correlation between two variables. It is a value between 1 and -1, where 1 indicates a perfect positive correlation between the variables, 0 indicates no linear correlation, and -1 indicates a perfect negative correlation. The PCC calculation formula is as follows:

[0069] Where ρ is PPC, is the mean of the variable x, It is the mean of the variable y. The value range of PCC is [-1, 1]. If it is close to 0, there is no correlation. The closer it is to 1 (-1), the stronger the positive correlation (negative correlation) between the two variables.

[0070] The Spearman correlation coefficient (SCC) is used to measure the correlation between two variables. The Spearman correlation coefficient indicates the direction of the correlation between X (independent variable) and Y (dependent variable). If Y tends to increase when X increases, the Spearman correlation coefficient is positive. If Y tends to decrease when X increases, the Spearman correlation coefficient is negative. The Spearman correlation coefficient ranges from -1 to 1, where -1 indicates a perfect negative correlation, 0 indicates no correlation, and 1 indicates a perfect positive correlation. The SCC calculation formula is a variation of the PC C calculation formula. The SCC calculation formula is as follows:

[0071] Among them, ρ s For SCC, R i and S i The observed values ​​x i and y i SCC evaluates the monotonic relationship between two variables and determines whether the variables show a trend of changing together.

[0072] The mean absolute error (MAE) reflects the absolute cumulative deviation between the predicted values ​​and the true values ​​of multiple samples. The mean absolute error is the average of the absolute deviations of all individual observations from the arithmetic mean. The mean absolute error can avoid the problem of errors canceling each other out, and thus accurately reflects the size of the actual prediction error. The MAE calculation formula is as follows:

[0073] Where n is the number of samples, y i The actual observed value, is the true value.

[0074] Mean squared error (MSE) is a measure of the difference between the estimator and the estimated quantity. The MSE calculation formula is as follows:

[0075] Root mean squared error (RMSE) is the root mean square of the prediction error. It is used to measure the accuracy of regression predictions. The RMSE calculation formula is as follows:

[0076] R 2, that is, the coefficient of determination is a statistical indicator that quantitatively characterizes the correlation between qualitative variables or factors. It is often used to measure the strength of the linear relationship between two variables. It is a quantitative measurement indicator with a value range between 0 and 1, and can also be regarded as a proportion reflecting the total variation of the variable. When it is greater than 0.8, it indicates that the linear relationship between the two variables is very strong; when it is less than 0.4, it indicates that there is no linear relationship between them. It is the most important measurement indicator in correlation analysis statistics. The formula for calculating the coefficient of determination is as follows: R 2 =PCC 2

[0077] The coefficient of variation (CV) is a statistic that measures the degree of variation of each observation in the data. When comparing the degree of variation of two or more data, if the measurement unit is the same as the mean, the standard deviation can be used directly for comparison. If the unit and / or mean are different, the standard deviation cannot be used to compare the degree of variation. Instead, the ratio of the standard deviation to the mean (relative value) must be used to compare the degree of variation. The formula for calculating the mean CV of RMSE is as follows:

[0078] Among them, CV is the mean RMSE CV, and Mean Value is the mean RMSE.

[0079] SNPs primarily refer to DNA sequence polymorphisms caused by variations in a single nucleotide at the genomic level. They are the most common type of heritable variation in organisms, accounting for over 90% of all known polymorphisms. SNPs are widespread in biological genomes, occurring on average at one in every 300 base pairs, with an estimated total number of 3 million or more. SNPs are dimorphic markers, caused by single base transitions or transversions, or by base insertions or deletions. SNPs can occur within gene sequences or in non-coding sequences outside of genes. Genetic screening requires identifying highly heritable core SNPs from within an organism's SNPs.

[0080] However, among the tens of thousands of SNPs in organisms, only a small number are core SNPs with high genetic characteristics.

[0081] To this end, an embodiment of the present application provides a data processing method that can be applied to a data processing system. Figure 1 is a functional block diagram of a data processing system 10 provided in an embodiment of the present application. As shown in Figure 1, the data processing system 10 may include various submodules. For example, a communication module 11, a processing module 12, a storage module 13, and an interface module 14.

[0082] Optionally, the data processing system 10 may include more or fewer submodules, and each submodule may include multiple components. In addition, each submodule and component of the data processing system 10 may be interconnected in a wired or wireless manner.

[0083] The communication module 11 may include a wireless communication system that can communicate wirelessly with one or more devices directly or via a communication network. For example, the wireless communication system may use a third generation mobile communication technology (3G) cellular communication, such as code division multiple access (CDMA), global system for mobile communications (GSM) or general packet radio service (GPRS), or a fourth generation mobile communication technology (4G) cellular communication, such as long term evolution (LTE). Or a fifth generation mobile communication technology (5G) cellular communication. The wireless communication system may communicate with a wireless local area network (WLAN) using wireless fidelity (WiFi). In some embodiments, the wireless communication system may communicate directly with the device using an infrared link, Bluetooth or zigbee.

[0084] The processing module 12 may be any conventional processor, such as a commercially available CPU. Alternatively, the processing module 12 may include a graphics processing unit (GPU), a field programmable gate array (FPGA), a system on chip (SOC), an application specific integrated circuit (ASIC), or a combination thereof.

[0085] The storage module 13 is used to store instructions, wherein the instructions may be computer programs.

[0086] The storage module 13 may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. The non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM may be used, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchlink DRAM (SLDRAM), and direct RAM bus RAM (DR RAM). The memory module 13 may also be a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compact disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), magnetic disk storage media or other magnetic storage devices. It should be noted that the memory of the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0087] The interface module 14 can be used to send or receive data, instructions or information. The processing module 12 can use the data, instructions or other information received by the interface module 14 to process and can send the processing completion information through the interface module 14.

[0088] Optionally, the above components are only an example. In actual applications, the components in the above modules may be added or deleted according to actual needs. FIG1 should not be understood as limiting the embodiments of the present invention.

[0089] FIG2 shows a data processing method provided by an embodiment of the present application, the method comprising:

[0090] S201. Obtain the first SNP set

[0091] The first SNP set includes N SNPs, where N is a positive integer.

[0092] Illustratively, N SNPs including soybeans may be obtained, and a set consisting of the obtained N SNPs of soybeans may be used as a first SNP set.

[0093] In a possible implementation, the first SNP set including N SNPs may be high-dimensional small sample data.

[0094] S202: Determine a second SNP set from the first SNP set according to the first SNP information.

[0095] The first information includes at least one of linkage disequilibrium information or variance, and the second SNP set includes M SNPs, where M is a positive integer and is smaller than N. The linkage disequilibrium information may include

[0096] In a possible implementation, when there are multiple SNPs with the same variance in the first SNP set, one SNP from the multiple SNPs with the same variance is retained and the other SNPs from the multiple SNPs are deleted to obtain the second SNP set.

[0097] For example, when there are 1000 SNPs with variance of 0.45 in the first SNP set, the first SNP that appears among the 1000 SNPs with variance of 0.45 can be retained in the first SNP set, and the remaining 999 SNPs with variance of 0.45 can be deleted from the first SNP set.

[0098] It is understandable that the genetic characteristics of SNPs with the same variance may also be the same. Therefore, only one SNP from multiple SNPs with the same variance can be retained, and the remaining SNPs from the multiple SNPs with the same variance can be deleted from the acquired SNP data to perform data dimensionality reduction on the acquired SNP data. The reduced dimensionality data is then subjected to feature screening using a target feature extraction method and a nonlinear feature extraction method, thereby screening out core SNPs with high genetic characteristics from the biological SNPs. In addition, the embodiment of the present application also selects SNPs with strong correlation with the dependent variable (phenotype) from the multiple SNPs acquired through the variance and linkage disequilibrium information of the SNPs.

[0099] In a possible implementation, the target SNP may be deleted from the first SNP set according to the LD coefficient of the SNP to obtain the second SNP set, where the target SNP is a SNP whose LD coefficient is outside a preset LD coefficient range.

[0100] For example, the LD coefficient of the SNP can be determined based on the corresponding frequency of the SNP characteristics in the population, and the target SNP can be deleted from the first SNP set based on the LD coefficient of the SNP to obtain the second SNP set.

[0101] For example, suppose two SNPs (x i ,x j ), its bases represent (A, B), the alleles are A, a, B, b, and the corresponding frequencies in the population are π A ,π a ,π B ,π b , the two SNPs have four haplotypes AB, Ab, aB, ab, corresponding to frequencies π AB ,π Ab ,π aB ,π ab , calculate D ab =π AB -π A π B Then, based on the calculated LD coefficients of the SNPs, the SNPs whose LD coefficients are outside the preset LD coefficient interval are deleted from the first SNP set.

[0102] LD coefficient satisfies: when when Where D'=0,r 2 =0 is in complete linkage equilibrium; D' = 1, r 2 =1 is in a state of complete linkage disequilibrium; the higher the metric from 0 to 1, the higher the LD and the stronger the degree of linkage.

[0103] S203 : Determine a third SNP set from the second SNP set according to a target feature extraction method and a nonlinear feature extraction method.

[0104] Among them, the above-mentioned target feature extraction method includes at least one of a filtering feature extraction method, a wrapping feature extraction method or an embedded feature extraction method, and the above-mentioned third SNP set includes H SNPs, the above-mentioned H is a positive integer, and the above-mentioned H is less than the above-mentioned M.

[0105] In one possible implementation, a fourth SNP set can be determined from the second SNP set according to the target feature extraction method. A third SNP set can be determined from the fourth SNP set according to the nonlinear feature extraction method. The fourth SNP set includes K SNPs, where K is a positive integer, and K is less than M and greater than H.

[0106] In one possible implementation, SNPs are eliminated from the second SNP set according to the target feature extraction method; when the determination coefficient of the remaining SNPs in the second SNP set is greater than the first threshold or the first remaining ratio is less than the first ratio threshold, the remaining SNPs in the second SNP set are determined as the fourth SNP set, and the first remaining ratio is the ratio of the number of remaining SNPs in the second SNP set to the M.

[0107] Exemplarily, SNPs can be screened out from the above-mentioned second SNP set according to the filtering feature extraction method, the wrapping feature extraction method and the embedded feature extraction method. When the determination coefficient of the remaining SNPs in the above-mentioned second SNP set is greater than 0.7 or the first remaining ratio is less than 20% (that is, the second SNP set has screened out 80% of the SNPs), the remaining SNPs in the above-mentioned second SNP set are determined as the above-mentioned fourth SNP set.

[0108] Accordingly, when the determination coefficient of the remaining SNPs is less than the first threshold and the first remaining ratio is greater than the first ratio threshold, SNPs can be continuously eliminated from the second SNP set according to the above-mentioned target feature extraction method until the determination coefficient of the remaining SNPs in the second SNP set is greater than the first threshold or the first remaining ratio is less than the first ratio threshold.

[0109] It can be seen that the method provided in the embodiment of the present application determines the representative SNPs in each region from the obtained multiple SNPs through the variance and linkage disequilibrium information of the SNPs, eliminates the remaining redundant SNPs, thereby performing data dimensionality reduction on the obtained SNP data, and then further screening the obtained multiple SNPs according to the determination coefficient and number of the remaining SNPs through the target feature extraction method. The further screened SNPs are then feature-screened using nonlinear features, thereby screening core SNPs with high genetic characteristics from the biological SNPs. In addition, by further screening the obtained multiple SNPs according to the determination coefficient and number of the remaining SNPs through the target feature extraction method, SNPs with penalty terms for correlation between features can be obtained.

[0110] In a possible implementation, the PCC of the remaining SNPs may be determined, and the coefficient of determination of the SNPs may be determined based on the PCC of the remaining SNPs.

[0111] In one possible implementation, SNPs can be eliminated from the fourth SNP set according to the nonlinear feature extraction method; when the determination coefficient of the remaining SNPs in the fourth SNP set is greater than the second threshold or the second remaining ratio is less than the second ratio threshold, the remaining SNPs in the fourth SNP set are determined as the third SNP set, and the second remaining ratio is the ratio of the number of remaining SNPs in the fourth SNP set to the M.

[0112] Exemplarily, a nonlinear feature extraction method can be used to screen out SNPs from the above-mentioned fourth SNP set. When the determination coefficient of the remaining SNPs in the above-mentioned fourth SNP set is greater than 0.75 or the second remaining proportion is less than 1% (that is, 99% of the SNPs have been screened out by the second SNP set), the remaining SNPs in the above-mentioned fourth SNP set are determined as the above-mentioned third SNP set.

[0113] Accordingly, when the determination coefficient of the remaining SNPs is less than the second threshold and the second remaining ratio is greater than the second ratio threshold, SNPs can continue to be eliminated from the above-mentioned fourth SNP set according to the above-mentioned target feature extraction method until the determination coefficient of the remaining SNPs in the fourth SNP set is greater than the second threshold or the second remaining ratio is less than the second ratio threshold.

[0114] It can be seen that the method provided in the embodiment of the present application determines the representative SNPs in each region from the obtained multiple SNPs through the variance and linkage disequilibrium information of SNPs, eliminates the remaining redundant SNPs, thereby performing data dimensionality reduction on the obtained SNP data, and then further screens the multiple SNPs obtained through the target feature extraction method. The SNPs after further screening are further screened from the multiple SNPs obtained according to the determination coefficient and number of the remaining SNPs through nonlinear characteristics, thereby screening out core SNPs with high genetic characteristics from the biological SNPs.

[0115] Optionally, the nonlinear feature extraction method includes but is not limited to the HSIC Lasso feature extraction method.

[0116] The present application does not limit the specific implementation method for determining the SNP determination coefficient, and any method that can be conceived by those skilled in the art can be used. For example, the SNP determination coefficient can be determined by a regression ensemble algorithm (Ridge regression, Lasso regression, Elastic net regularization, SVR, or LightGBM).

[0117] It should be noted that among the multiple SNPs in a region, there may be one SNP that can represent the type of most SNPs in the region. The embodiment of the present application determines the representative SNPs in each region from the multiple SNPs obtained through the variance and linkage disequilibrium information of the SNPs, eliminates the remaining redundant SNPs, and thus performs data dimensionality reduction on the obtained SNP data. The target feature extraction method and the nonlinear feature extraction method are then used to perform feature screening on the reduced data, thereby screening out core SNPs with high genetic characteristics from the biological SNPs. In addition, the embodiment of the present application also selects SNPs with strong correlation with the dependent variable (phenotype) from the multiple SNPs obtained through the variance and linkage disequilibrium information of the SNPs.

[0118] Table 1 shows the prediction results obtained by the data processing method provided in the embodiment of the present application for soybean phenotype prediction. As can be seen from Table 1, the prediction correlation coefficient (PCC) index obtained by the data processing method provided in the embodiment of the present application reaches over 94%, close to the heritability of 95%. After preprocessing and HSIC LASSO screening, the number of characteristic SNPs is 440, a reduction of 99.7%, which is an overall improvement. Compared with the test results of related technologies, the data processing method provided in the embodiment of the present application has advanced phenotype prediction based on genotype data.

[0119] Table 1

[0120] The data processing device for executing the above data processing method will be described below with reference to FIG. 3 .

[0121] It is understandable that, in order to implement the above functions, the data processing device includes hardware and / or software modules that perform the corresponding functions. In combination with the algorithm steps of each example described in the embodiments disclosed herein, the embodiments of the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application in combination with the embodiments, but such implementation should not be considered to exceed the scope of the embodiments of the present application.

[0122] In the embodiment of the present application, the data processing device can be divided into functional modules according to the above method example. For example, each functional module can be divided according to each function, or two or more functions can be integrated into one processing module. The above integrated modules can be implemented in the form of hardware. It should be noted that the division of modules in this embodiment is schematic and is only a logical function division. In actual implementation, other division methods may be used.

[0123] In the case of dividing each functional module according to each function, Figure 3 shows a possible composition diagram of the data processing device involved in the above embodiment. As shown in Figure 3, the data processing device 300 may include: a transceiver unit 301 and a processing unit 302.

[0124] The transceiver unit 301 is configured to obtain a first SNP set, where the first SNP set includes N SNPs, where N is a positive integer.

[0125] The processing unit 302 is used to determine a second SNP set from the first SNP set based on the first information of the SNP, wherein the first information includes at least one of linkage disequilibrium information or variance, and the second SNP set includes M SNPs, where M is a positive integer and is less than N.

[0126] The above-mentioned processing unit 302 is also used to determine a third SNP set from the above-mentioned second SNP set according to a target feature extraction method and a nonlinear feature extraction method, the above-mentioned target feature extraction method includes at least one of a filtering feature extraction method, a wrapper feature extraction method or an embedded feature extraction method, and the above-mentioned third SNP set includes H SNPs, the above-mentioned H is a positive integer, and the above-mentioned H is less than the above-mentioned M.

[0127] In one possible implementation, the first information includes linkage disequilibrium information, the linkage disequilibrium information includes the linkage disequilibrium LD coefficient, and the processing unit is specifically used to: delete the target SNP from the first SNP set according to the LD coefficient of the SNP to obtain the second SNP set, and the target SNP is a SNP whose LD coefficient is outside the preset LD coefficient range.

[0128] In one possible implementation, the first information includes variance, and the processing unit 302 is specifically used to: when there are multiple SNPs with the same variance in the first SNP set, retain one SNP from the multiple SNPs with the same variance and delete the other SNPs from the multiple SNPs to obtain the second SNP set.

[0129] In one possible implementation, the processing unit 302 is specifically used to: determine a fourth SNP set from the second SNP set according to the target feature extraction method, the fourth SNP set including K SNPs, K being a positive integer, K being less than M and greater than H; and determine a third SNP set from the fourth SNP set according to the nonlinear feature extraction method.

[0130] In one possible implementation, the processing unit 302 is specifically used to: eliminate SNPs from the second SNP set according to the target feature extraction method; when the determination coefficient of the remaining SNPs in the second SNP set is greater than the first threshold or the first remaining ratio is less than the first ratio threshold, the remaining SNPs in the second SNP set are determined as the fourth SNP set, and the first remaining ratio is the ratio of the number of remaining SNPs in the second SNP set to the M.

[0131] In one possible implementation, the processing unit 302 is specifically used to: eliminate SNPs from the fourth SNP set according to the nonlinear feature extraction method; when the determination coefficient of the remaining SNPs in the fourth SNP set is greater than the second threshold or the second remaining ratio is less than the second ratio threshold, determine the remaining SNPs in the fourth SNP set as the third SNP set, and the second remaining ratio is the ratio of the number of remaining SNPs in the fourth SNP set to the M.

[0132] The present application also provides a chip. FIG4 shows a schematic diagram of the structure of a chip 1000. The chip 400 includes one or more processors 401 and an interface circuit 402. Optionally, the chip 400 may also include a bus 403.

[0133] The processor 401 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above data processing method can be completed by the hardware integrated logic circuit in the processor 401 or the software instruction.

[0134] Optionally, the processor 401 may be a general-purpose processor, a digital signal processing (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The methods and steps disclosed in the embodiments of the present application may be implemented or executed. The general-purpose processor may be a microprocessor or any conventional processor.

[0135] The interface circuit 402 can be used to send or receive data, instructions or information. The processor 401 can use the data, instructions or other information received by the interface circuit 402 to process it, and can send the processing completion information through the interface circuit 402.

[0136] Optionally, the chip also includes a memory, which may include a read-only memory and a random access memory, and provides operating instructions and data to the processor. Part of the memory may also include a non-volatile random access memory (NVRAM).

[0137] Optionally, the memory stores an executable software module or a data structure, and the processor can perform corresponding operations by calling an operation instruction stored in the memory (the operation instruction may be stored in an operating system).

[0138] Optionally, the chip can be used in a data processing device according to an embodiment of the present application. Optionally, the interface circuit 402 can be used to output the execution result of the processor 401. For the data processing method provided in one or more embodiments of the present application, reference can be made to the aforementioned embodiments and will not be repeated here.

[0139] It should be noted that the corresponding functions of the processor 401 and the interface circuit 402 can be implemented through hardware design, software design, or a combination of hardware and software, and there is no limitation here.

[0140] FIG5 is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device 500 may be a processor or a chip or functional module in the processor. As shown in FIG5 , the electronic device 500 includes a processor 501 , a transceiver 502 , and a communication line 503 .

[0141] Among them, the processor 501 is used to execute any step in the data processing method provided in the embodiment of the present application, and in the process of executing any step in the data processing method provided in the embodiment of the present application, the transceiver 502 and the communication line 503 can be optionally called to complete the corresponding operation.

[0142] Furthermore, the electronic device 500 may further include a memory 504 , wherein the processor 501 , the memory 504 and the transceiver 502 may be connected via a communication line 503 .

[0143] The processor 501 is a processor, a general-purpose processor, a network processor (NP), a digital signal processor (DSP), a microprocessor, a microcontroller, a programmable logic device (PLD), or any combination thereof. The processor 501 may also be other devices with processing functions, such as circuits, devices, or software modules, without limitation.

[0144] The transceiver 502 is used to communicate with other devices or other communication networks, such as Ethernet, radio access networks (RAN), wireless local area networks (WLAN), etc. The transceiver 502 can be a module, a circuit, a transceiver, or any device capable of implementing communication.

[0145] The transceiver 502 is mainly used for sending and receiving commands and information, and may include a transmitter and a receiver for sending and receiving commands and information, respectively. Operations other than sending and receiving commands and information are implemented by the processor.

[0146] The communication line 503 is used to transmit information between the components included in the electronic device 500.

[0147] In one design, the processor can be considered as the logic circuit and the transceiver as the interface circuit.

[0148] The memory 504 is used to store instructions, where the instructions may be computer programs.

[0149] The memory 504 may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. The non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM may be used, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchlink DRAM (SLDRAM), and direct RAM bus RAM (DR RAM). Memory 504 may also be a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compact disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), magnetic disk storage media or other magnetic storage devices, etc. It should be noted that the memory of the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0150] It should be noted that the memory 504 can exist independently of the processor 501 or can be integrated with the processor 501. The memory 504 can be used to store instructions, program code, or some data. The memory 504 can be located inside the electronic device 500 or outside the electronic device 500, without limitation. The processor 501 is configured to execute the instructions stored in the memory 504 to implement the methods provided in the above embodiments of the present application.

[0151] In one example, processor 501 may include one or more processors, such as processor 0 and processor 1 in FIG. 5 .

[0152] As an optional implementation, the electronic device 500 includes multiple processors. For example, in addition to the processor 501 in FIG. 5 , it may also include a processor 507 .

[0153] As an optional implementation, the electronic device 500 further includes an output device 505 and an input device 506. For example, the input device 506 is a keyboard, a mouse, a microphone, a joystick, or the like, and the output device 505 is a display screen, a speaker, or the like.

[0154] It should be pointed out that the electronic device 500 can be a chip system or a device with a similar structure as shown in Figure 5. Among them, the chip system can be composed of chips, or it can include chips and other discrete devices. The actions, terms, etc. involved in the various embodiments of this application can refer to each other without limitation. The message names or parameter names in the messages exchanged between the various devices in the embodiments of this application are only an example. Other names can also be used in the specific implementation without limitation. In addition, the component structure shown in Figure 5 does not constitute a limitation on the electronic device 500. In addition to the components shown in Figure 5, the electronic device 500 may include more or fewer components than those shown in Figure 5, or combine certain components, or arrange the components differently.

[0155] The processor and transceiver described in this application can be implemented on an integrated circuit (IC), an analog IC, a radio frequency integrated circuit, a mixed-signal IC, an application specific integrated circuit (ASIC), a printed circuit board (PCB), an electronic device, etc. The processor and transceiver can also be manufactured using various IC process technologies, such as complementary metal oxide semiconductor (CMOS), N-type metal oxide semiconductor (NMOS), P-type metal oxide semiconductor (positive channel metal oxide semiconductor, PMOS), bipolar junction transistor (BJT), bipolar CMOS (BiCMOS), silicon germanium (SiGe), gallium arsenide (GaAs), etc.

[0156] Figure 6 is a schematic diagram of the structure of a data processing device provided in an embodiment of the present application. The data processing device can be applied to the scenarios shown in the above method embodiments. For ease of explanation, Figure 6 only shows the main components of the data processing device, including a processor 601, a memory 602, a control circuit 603, and an input / output device 604. The processor 601 is mainly used to process communication protocols and communication data, execute software programs, and process data of software programs. The memory 602 is mainly used to store software programs and data. The control circuit 603 is mainly used for power supply and transmission of various electrical signals. The input / output device 604 is mainly used to receive data input by the user and output data to the user.

[0157] When the data processing device is a processor 601, the control circuit 603 may be a motherboard, the memory 602 may include a hard disk, RAM, ROM, or other media with storage functions, the processor 601 may include a baseband processor 601 and a central processing unit, the baseband processor is mainly used to process communication protocols and communication data, the central processing unit is mainly used to control the entire data processing device, execute software programs, and process software program data, the input and output devices 604 include a display screen, a keyboard, and a mouse, etc.; the control circuit 603 may further include or be connected to a transceiver circuit or transceiver, such as a network cable interface, for sending or receiving data or signals, such as for data transmission and communication with other devices. Furthermore, it may also include an antenna for transmitting and receiving wireless signals for data / signal transmission with other devices.

[0158] An embodiment of the present application further provides a data processing device, which includes: at least one processor, and when the at least one processor executes program code or instructions, it implements the above-mentioned related method steps to implement the data processing method in the above-mentioned embodiment.

[0159] Optionally, the apparatus may further include at least one memory configured to store the program code or instruction.

[0160] An embodiment of the present application also provides a computer storage medium, which stores computer instructions. When the computer instructions are executed on a data processing device, the data processing device executes the above-mentioned related method steps to implement the data processing method in the above-mentioned embodiment.

[0161] An embodiment of the present application further provides a computer program product. When the computer program product is run on a computer, the computer is caused to execute the above-mentioned related steps to implement the data processing method in the above-mentioned embodiment.

[0162] The present application also provides a data processing device, which may be a chip, integrated circuit, component, or module. Specifically, the device may include a processor and a memory for storing instructions, or the device may include at least one processor configured to retrieve instructions from an external memory. When the device is running, the processor may execute the instructions, causing the chip to perform the data processing methods described in the above method embodiments.

[0163] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0164] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0165] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0166] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0167] The units described above as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, and may be located in one place or distributed across multiple network units. Some or all of these units may be selected based on actual needs to achieve the objectives of this embodiment.

[0168] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0169] If the above functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the above methods of each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0170] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A data processing method, It is characterized in that The method comprises: Obtaining a first single nucleotide polymorphism (SNP) set, wherein the first SNP set includes N SNPs, where N is a positive integer; Determine a second SNP set from the first SNP set according to first information of SNPs, wherein the first information includes at least one of linkage disequilibrium information or variance, and the second SNP set includes M SNPs, where M is a positive integer and is less than N; A third SNP set is determined from the second SNP set according to a target feature extraction method and a nonlinear feature extraction method, wherein the target feature extraction method includes at least one of a filtering feature extraction method, a wrapping feature extraction method or an embedded feature extraction method, and the third SNP set includes H SNPs, where H is a positive integer and is less than M.

2. The method according to claim 1, It is characterized in that The first information includes linkage disequilibrium information, and the linkage disequilibrium information includes a linkage disequilibrium LD coefficient. The determining of the second SNP set from the first SNP set according to the first information of the SNP includes: According to the LD coefficient of the SNP, a target SNP is deleted from the first SNP set to obtain the second SNP set, wherein the target SNP is a SNP whose LD coefficient is outside a preset LD coefficient range.

3. The method according to claim 1 or 2, It is characterized in that The first information includes variance, and determining the second SNP set from the first SNP set according to the first SNP information includes: In the case that there are multiple SNPs with the same variance in the first SNP set, one SNP among the multiple SNPs with the same variance is retained and other SNPs among the multiple SNPs are deleted to obtain the second SNP set.

4. The method according to any one of claims 1 to 3, It is characterized in that The step of determining the third SNP set from the second SNP set according to the target feature extraction method and the nonlinear feature extraction method comprises: Determine a fourth SNP set from the second SNP set according to the target feature extraction method, wherein the fourth SNP set includes K SNPs, where K is a positive integer, and K is less than M and greater than H; A third SNP set is determined from the fourth SNP set according to the non-linear feature extraction method.

5. The method according to claim 4, It is characterized in that Determining a fourth SNP set from the second SNP set according to the target feature extraction method comprises: Eliminate SNPs from the second SNP set according to the target feature extraction method; When the determination coefficient of the remaining SNPs in the second SNP set is greater than the first threshold or the first remaining ratio is less than the first ratio threshold, the remaining SNPs in the second SNP set are determined as the fourth SNP set, and the first remaining ratio is the ratio of the number of remaining SNPs in the second SNP set to M.

6. The method according to claim 4 or 5, It is characterized in that The determining the third SNP set from the fourth SNP set according to the nonlinear feature extraction method comprises: Eliminate SNPs from the fourth SNP set according to the nonlinear feature extraction method; When the determination coefficient of the remaining SNPs in the fourth SNP set is greater than the second threshold or the second remaining ratio is less than the second ratio threshold, the remaining SNPs in the fourth SNP set are determined as the third SNP set, and the second remaining ratio is the ratio of the number of remaining SNPs in the fourth SNP set to M.

7. A data processing device, It is characterized in that The device comprises a transceiver unit and a processing unit; The transceiver unit is used to obtain a first SNP set, where the first SNP set includes N SNPs, where N is a positive integer; The processing unit is used to determine a second SNP set from the first SNP set according to first information of SNPs, wherein the first information includes at least one of linkage disequilibrium information or variance, and the second SNP set includes M SNPs, wherein M is a positive integer and is less than N; The processing unit is also used to determine a third SNP set from the second SNP set according to a target feature extraction method and a nonlinear feature extraction method, the target feature extraction method includes at least one of a filtering feature extraction method, a wrapping feature extraction method or an embedded feature extraction method, and the third SNP set includes H SNPs, where H is a positive integer and is less than M.

8. The device according to claim 7, It is characterized in that The first information includes linkage disequilibrium information, and the linkage disequilibrium information includes a linkage disequilibrium LD coefficient. The processing unit is specifically used for: According to the LD coefficient of the SNP, a target SNP is deleted from the first SNP set to obtain the second SNP set, wherein the target SNP is a SNP whose LD coefficient is outside a preset LD coefficient range.

9. The device according to claim 7 or 8, It is characterized in that The first information includes variance, and the processing unit is specifically configured to: In the case that there are multiple SNPs with the same variance in the first SNP set, one SNP among the multiple SNPs with the same variance is retained and other SNPs among the multiple SNPs are deleted to obtain the second SNP set.

10. The device according to any one of claims 7 to 9, It is characterized in that The processing unit is specifically used for: Determine a fourth SNP set from the second SNP set according to the target feature extraction method, wherein the fourth SNP set includes K SNPs, where K is a positive integer, and K is less than M and greater than H; A third SNP set is determined from the fourth SNP set according to the non-linear feature extraction method.

11. The device according to claim 10, It is characterized in that The processing unit is specifically used for: Eliminate SNPs from the second SNP set according to the target feature extraction method; When the determination coefficient of the remaining SNPs in the second SNP set is greater than the first threshold or the first remaining ratio is less than the first ratio threshold, the remaining SNPs in the second SNP set are determined as the fourth SNP set, and the first remaining ratio is the ratio of the number of remaining SNPs in the second SNP set to M.

12. The device according to claim 10 or 11, It is characterized in that The processing unit is specifically used for: Eliminate SNPs from the fourth SNP set according to the nonlinear feature extraction method; When the determination coefficient of the remaining SNPs in the fourth SNP set is greater than the second threshold or the second remaining ratio is less than the second ratio threshold, the remaining SNPs in the fourth SNP set are determined as the third SNP set, and the second remaining ratio is the ratio of the number of remaining SNPs in the fourth SNP set to M.

13. A data processing device comprising at least one processor and a memory, It is characterized in that The at least one processor executes a program or instruction stored in the memory, so that the data processing device implements the method according to any one of claims 1 to 6.

14. A computer-readable storage medium for storing a computer program, It is characterized in that When the computer program is executed on a computer or a processor, the computer or the processor is enabled to implement the method according to any one of claims 1 to 6.

15. A computer program product comprising instructions, It is characterized in that When the instructions are executed on a computer or a processor, the computer or the processor is enabled to implement the method according to any one of claims 1 to 6.

16. A chip comprising at least one processor and a memory, It is characterized in that The at least one processor executes a program or instruction stored in the memory, so that the data processing device implements the method according to any one of claims 1 to 6.

Images