System for detecting snp interactions based on heterogeneous biomolecular networks

Abstract

The disclosure provides a SNP interaction detection system based on a heterogeneous biomolecule network, which comprises: obtaining SNP sequences of diseased individuals and non-diseased individuals, and preprocessing the data; constructing a statistical SNP correlation network according to the SNP sequences of the samples and the disease conditions of the samples; constructing a three-order SNP correlation network based on the known correlation between SNPs and biomolecules, and the correlation between biomolecules and diseases; constructing a two-order SNP correlation network based on the known correlation between SNPs and genes, and the functional correlation between genes; constructing a composite network for describing the correlation between SNPs based on the statistical SNP correlation network, the three-order SNP correlation network and the two-order SNP correlation network; clustering the SNP sequences based on the composite network; and detecting high-dimensional SNP combinations related to diseases from each cluster.

Description

Technical Field

[0001] The present disclosure belongs to the field of artificial intelligence data mining classification and bioinformatics technology, and in particular relates to a SNP interaction detection system based on heterogeneous biomolecular networks. Background Art

[0002] The statements in this section merely provide background information related to the present disclosure and do not necessarily constitute prior art.

[0003] In the context of artificial intelligence, machine learning technology can assist in the study of the relationship between SNP combinations and diseases by integrating heterogeneous biomolecular networks. How to quickly and accurately detect SNP interactions from massive data is a difficulty that still needs to be improved in current machine learning technology in association detection.

[0004] Current SNP interaction detection methods include: exhaustive search method: evaluating the correlation between each SNP combination and the disease one by one; using a clustering algorithm to first divide the complete SNP set into multiple clusters, and then searching for SNP combinations related to the disease within each cluster; considering that SNP-related genes, miRNAs, lncRNAs, etc. can assist in the identification of similar SNP pairs, integrating this heterogeneous biomolecular information into SNP interaction detection can help us obtain more accurate identification results.

[0005] The inventors found that due to the high-dimensional SNP combinations composed of massive SNP data in the whole genome, the number of possible SNP combinations increases exponentially with the increase of dimension, and some algorithms face a heavy computational burden. In addition, problems such as excessive false positives also mean that there is still a lot of room for improvement in SNP interaction detection in machine learning technology. Summary of the Invention

[0006] In order to solve the above problems, the present disclosure provides a SNP interaction detection system based on a heterogeneous biomolecule network. The scheme constructs a SNP statistical network and a SNP correlation network based on heterogeneous biomolecules. A composite network is constructed from multiple networks to describe the similarity between SNPs. The association information between known SNPs and biomolecules can be used as a guide to increase the accuracy of the definition of SNP similarity. At the same time, the whole genome SNPs are constructed into multiple clusters based on the established composite network, and disease-related SNP combinations are searched within each cluster. Compared with the original huge search space, the computational burden of the algorithm is reduced and the efficiency is improved.

[0007] According to a first aspect of an embodiment of the present disclosure, a SNP interaction detection system based on a heterogeneous biomolecular network is provided, comprising:

[0008] A data pre-acquisition module is configured to: acquire SNP sequences of diseased individuals and non-diseased individuals, and pre-process the data;

[0009] A statistical network construction module is configured to: construct a statistical SNP correlation network based on the SNP sequence of the sample and the disease condition of the sample;

[0010] A third-order SNP correlation network construction module is configured to: construct a third-order SNP correlation network based on known correlations between SNPs and biomolecules, and correlations between biomolecules and diseases;

[0011] A second-order SNP correlation network construction module is configured to: construct a second-order SNP correlation network based on the known correlation between SNPs and genes, and the functional correlation between genes;

[0012] A composite network generation module is configured to: construct a composite network for describing SNP correlation based on the statistical SNP correlation network, the third-order SNP correlation network, and the second-order SNP correlation network;

[0013] a clustering module, configured to: cluster the SNP sequences based on the composite network;

[0014] The interaction detection module is configured to detect high-dimensional SNP combinations associated with the disease from each cluster.

[0015] Furthermore, the construction of a statistical SNP correlation network based on the SNP sequence of the sample and the disease condition of the sample specifically includes:

[0016] For the data in the sample set, the number of diseased individuals and the number of non-diseased individuals under each genotype in each SNP combination were counted;

[0017] Based on statistical data, the association degree between each two SNPs was evaluated based on preset evaluation indicators;

[0018] For each SNP, only the t SNPs with the strongest association are retained as neighbors with the SNP, where t is an integer greater than 0.

[0019] Furthermore, based on the known correlations between SNPs and biomolecules, as well as the correlations between biomolecules and diseases, a third-order SNP correlation network is constructed. The specific steps include:

[0020] Based on the verified association between SNPs and genes, miRNAs, and lncRNAs, a matrix describing the association is constructed;

[0021] Based on the verified associations between diseases and genes, miRNAs, and lncRNAs, a matrix describing the associations is constructed;

[0022] Based on the SNP-to-SNP paths implied in the matrix, the SNP correlation network is constructed.

[0023] Furthermore, based on the known correlation between SNPs and genes, and the functional correlation between genes, a second-order SNP correlation network is constructed, specifically:

[0024] Determine the mapping relationship between SNPs, genes and gene ontology;

[0025] Calculate the similarity between gene ontologies;

[0026] It is converted into the similarity between SNPs to realize the construction of SNP correlation network.

[0027] Furthermore, the SNP sequences are clustered based on the composite network, specifically:

[0028] Initially, each SNP is considered as a community;

[0029] For each SNP node, calculate the module gain of the network if it is moved to the cluster where the neighboring node is located; take the module with the largest and positive gain as the module that the SNP will move into; repeat this process until there is no change in the community to which the SNP belongs;

[0030] Treat each community as a node and repeat the above process until the entire network meets the preset constraints;

[0031] Ultimately, each community represents a cluster, and SNPs in the same community belong to the same cluster.

[0032] Furthermore, the preset constraint condition is: the entire network no longer changes or the maximum number of iterations is reached.

[0033] Furthermore, high-dimensional SNP combinations associated with the disease are detected from each cluster. Specifically, the p-value corresponding to each SNP combination is calculated using the chi-square test, so as to select the SNP combinations that have a significant correlation with the disease.

[0034] According to a second aspect of an embodiment of the present disclosure, an electronic device is provided, including a memory, a processor, and a computer program stored and running on the memory, wherein when the processor executes the program, the following steps are performed:

[0035] Obtain SNP sequences of diseased and non-diseased individuals and preprocess the data;

[0036] Construct a statistical SNP correlation network based on the SNP sequence of the sample and the disease status of the sample;

[0037] Based on the known correlation between SNPs and biomolecules, as well as the correlation between biomolecules and diseases, a third-order SNP correlation network is constructed;

[0038] Based on the known correlation between SNPs and genes, as well as the functional correlation between genes, a second-order SNP correlation network is constructed;

[0039] Based on the statistical SNP correlation network, the third-order SNP correlation network and the second-order SNP correlation network, a composite network for describing SNP correlation relationships is constructed;

[0040] The SNP sequences are clustered based on the composite network, and high-dimensional SNP combinations associated with the disease are detected from each cluster.

[0041] According to a third aspect of an embodiment of the present disclosure, a non-transitory computer-readable storage medium is provided, on which a computer program is stored. When the program is executed by a processor, the following steps are performed:

[0042] Obtain SNP sequences of diseased and non-diseased individuals and preprocess the data;

[0043] Construct a statistical SNP correlation network based on the SNP sequence of the sample and the disease status of the sample;

[0044] Based on the known correlation between SNPs and biomolecules, as well as the correlation between biomolecules and diseases, a third-order SNP correlation network is constructed;

[0045] Based on the known correlation between SNPs and genes, as well as the functional correlation between genes, a second-order SNP correlation network is constructed;

[0046] Based on the statistical SNP correlation network, the third-order SNP correlation network and the second-order SNP correlation network, a composite network for describing SNP correlation relationships is constructed;

[0047] The SNP sequences are clustered based on the composite network, and high-dimensional SNP combinations associated with the disease are detected from each cluster.

[0048] According to a fourth aspect of the embodiments of the present disclosure, there is provided a computer program product, comprising a computer program, wherein when the computer program is run on one or more processors, the computer program performs the following steps:

[0049] Obtain SNP sequences of diseased and non-diseased individuals and preprocess the data;

[0050] Construct a statistical SNP correlation network based on the SNP sequence of the sample and the disease status of the sample;

[0051] Based on the known correlation between SNPs and biomolecules, as well as the correlation between biomolecules and diseases, a third-order SNP correlation network is constructed;

[0052] Based on the known correlation between SNPs and genes, as well as the functional correlation between genes, a second-order SNP correlation network is constructed;

[0053] Based on the statistical SNP correlation network, the third-order SNP correlation network and the second-order SNP correlation network, a composite network for describing SNP correlation relationships is constructed;

[0054] The SNP sequences are clustered based on the composite network, and high-dimensional SNP combinations associated with the disease are detected from each cluster.

[0055] Compared with the prior art, the present invention has the following advantages:

[0056] (1) The present disclosure provides a SNP interaction detection system based on a heterogeneous biomolecule network. The scheme constructs a SNP statistical network and a SNP-related network based on heterogeneous biomolecules. A composite network is constructed by multiple networks to describe the similarity between SNPs. The association information between known SNPs and biomolecules can be used as a guide to increase the accuracy of the definition of SNP similarity. At the same time, the whole genome SNPs are constructed into multiple clusters based on the established composite network, and disease-related SNP combinations are searched within each cluster. Compared with the original huge search space, the computational burden of the algorithm is reduced and the efficiency is improved.

[0057] (2) The present disclosure integrates the SNP similarity defined by sample information and the SNP similarity defined by the relationship between known biological molecules, taking into account both the individual information of the current sample set and the common related information between verified biological molecules, thereby increasing the accuracy of the defined SNP similarity; using a clustering algorithm to divide the whole genome SNP set into multiple subsets, narrowing the search space and reducing the computational burden.

[0058] Advantages of additional aspects of the present disclosure will be given in part in the following description and in part will become apparent from the following description or learned through practice of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] The accompanying drawings, which constitute a part of the present disclosure, are used to provide a further understanding of the present disclosure. The exemplary embodiments of the present disclosure and their descriptions are used to explain the present disclosure and do not constitute an improper limitation to the present disclosure.

[0060] Figure 1 Schematic diagram of the structure of the SNP interaction detection system based on heterogeneous biomolecular network described in the embodiment of the present disclosure. DETAILED DESCRIPTION

[0061] The present disclosure will be further described below with reference to the accompanying drawings and embodiments.

[0062] It should be noted that the following detailed descriptions are illustrative and intended to provide further explanation of the present disclosure. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present disclosure belongs.

[0063] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present disclosure. As used herein, unless the context clearly indicates otherwise, the singular form is intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0064] In the absence of conflict, the embodiments of the present disclosure and the features thereof may be combined with each other.

[0065] Example 1:

[0066] The purpose of this embodiment is to provide a SNP interaction detection system based on a heterogeneous biomolecular network.

[0067] like Figure 1 As shown, the SNP interaction detection system based on heterogeneous biomolecular networks includes:

[0068] A data pre-acquisition module is configured to: acquire SNP sequences of diseased individuals and non-diseased individuals, and pre-process the data;

[0069] A statistical network construction module is configured to: construct a statistical SNP correlation network based on the SNP sequence of the sample and the disease condition of the sample;

[0070] A third-order SNP correlation network construction module is configured to: construct a third-order SNP correlation network based on known correlations between SNPs and biomolecules, and correlations between biomolecules and diseases;

[0071] A second-order SNP correlation network construction module is configured to: construct a second-order SNP correlation network based on the known correlation between SNPs and genes, and the functional correlation between genes;

[0072] A composite network generation module is configured to: construct a composite network for describing SNP correlation based on the statistical SNP correlation network, the third-order SNP correlation network, and the second-order SNP correlation network;

[0073] a clustering module, configured to: cluster the SNP sequences based on the composite network;

[0074] The interaction detection module is configured to detect high-dimensional SNP combinations associated with the disease from each cluster.

[0075] Furthermore, individual SNP data is obtained and preprocessed;

[0076] Among them, individuals include sick individuals and non-sick individuals, and sick individuals are marked as 1; non-sick individuals are marked as 0;

[0077] Among them, the mutation status of each SNP site in the SNP data to the sample is marked as 0 if both alleles are not mutated; if one of the two alleles is mutated, it is marked as 1; if both alleles are mutated, it is marked as 2; if the data of the site is missing, it is marked as 3;

[0078] The specific steps of data preprocessing are as follows: delete samples with more than 5% missing SNPs; delete SNPs with more than 5% missing samples; calculate the p-value of each SNP using the chi-square test and delete SNPs with p-value > 0.0001; delete SNPs with minor allele frequency less than 0.1.

[0079] Furthermore, based on the known correlations between SNPs and genes, as well as the functional correlations between genes, a SNP correlation network is constructed, specifically including:

[0080] For each two-point SNP combination, the corresponding entropy score (ES) is calculated. The larger the entropy value, the stronger the association between the two SNPs. For each SNP, t SNPs with the largest association degree are retained to form a SNP correlation network. Each point in the network represents a SNP, and each edge represents an association relationship between the two connected SNPs. The weight value of the edge is its corresponding entropy.

[0081] Furthermore, a third-order SNP correlation network was constructed, specifically including:

[0082] If two SNPs are both associated with a gene, then the two SNPs are likely to be associated;

[0083] If two SNPs are associated with two genes, and both genes are associated with a disease, then the two SNPs are likely to be associated;

[0084] If two SNPs are both associated with a certain miRNA, then the two SNPs may be related;

[0085] If two SNPs are associated with two miRNAs, and both miRNAs are associated with a certain disease, then the two SNPs may be associated;

[0086] If two SNPs are both associated with a certain lncRNA, then the two SNPs may be related;

[0087] If two SNPs are associated with two lncRNAs, and both lncRNAs are associated with a certain disease, then the two SNPs may be associated;

[0088] The present disclosure uses the aforementioned six known biomolecule association information to construct a third-order SNP correlation network.

[0089] Furthermore, a statistical SNP correlation network N is established based on the SNP sequence of the sample and the disease status of the sample. ssn ; Specifically include:

[0090] S1021, for all two-site SNP combinations, calculate the corresponding entropy ES(S1, S2) in the sample set, which is defined as follows:

[0091]

[0092] in, h 1i =-p i logp i -(1-p i )log(1-p i ),

[0093] Where i = 0, 1, 2 represent three genotypes (AA, Aa, aa), i = 0, ..., 9 represent nine genotypes of two-site combination (AABB, AABb, ..., aabb) n represents the number of all samples, n total (i) represents the number of samples with genotype i, n case (i) is the number of diseased samples with genotype i. The larger the ES value, the

[0094] S1022: For each SNP, only the association relationships between the t SNPs with the strongest association are retained.

[0095] S1023, constructing a SNP correlation network based on the calculated association relationship: each point in the network represents a SNP, an edge represents that the two connected SNPs have an association relationship, and the weight value of the edge is the entropy corresponding to the two SNPs.

[0096] Furthermore, based on the known correlations between SNPs and biomolecules, and between biomolecules and diseases, a third-order SNP correlation network is constructed. Specifically, the following steps are involved:

[0097] S1031, construct matrix W snp-gene :Each row of the matrix represents a SNP, and each column represents a gene. If a SNP is associated with a gene, the value of the position uniquely determined by the row and column in the matrix is ​​1, otherwise it is 0; construct the matrix W gene-disease Represents a gene, and each column represents a disease. If a gene is associated with a disease, the value of the position uniquely determined by the row and column in the matrix is ​​1, otherwise it is 0; construct the matrix W SNP-miRNA , each row represents a SNP, each column represents a miRNA. If a SNP is associated with a miRNA, the value of the position uniquely determined by the row and column in the matrix is ​​1, otherwise it is 0. Construct the matrix W miRNA-disease , each row represents a miRNA, each column represents a disease. If a miRNA is associated with a disease, the value of the position uniquely determined by the row and column in the matrix is ​​1, otherwise it is 0; construct the matrix W SNP-lncRNA , each row represents a SNP, each column represents a lncRNA. If a SNP is associated with a lncRNA, the value of the position uniquely determined by the row and column in the matrix is ​​1, otherwise it is 0. Construct the matrix W lncRNA-disease , each row represents a lncRNA, and each column represents a disease. If a lncRNA is associated with a disease, the value of the position uniquely determined by the row and column in the matrix is ​​1, otherwise it is 0;

[0098] S1032, construct SNP-gene-disease network N sgd :Each point in the network represents a SNP, and the weight value of the edge in the network is defined as:

[0099]

[0100] Where: M ij =W snp-gene *W gene-disease *W gene-disease '*W snp-gene '.

[0101] M ijRepresents the number of paths between SNPi and SNPj in the meta-path snp-gene-disease. A larger weight indicates a stronger association between SNPs. For each SNP, only the k most correlated SNPs are retained as its neighbors.

[0102] Similarly, the SNP-miRNA-disease network N was constructed. smd and SNP-lncRNA-disease network N sld .

[0103] Furthermore, based on the known correlations between SNPs and genes, as well as the functional correlations between genes, a second-order SNP correlation network is constructed; specifically, the following steps are involved:

[0104] Map each SNP to its associated gene.

[0105] The similarity between different SNPs is defined based on the similarity of gene ontology.

[0106] Only the p most similar SNPs of each SNP are retained as its neighbors to construct the second-order SNP correlation network N snp-gene Each point in the network represents a SNP, an edge represents that the two SNPs it connects are related, and the weight value of the edge is the similarity between the SNPs calculated in S1042.

[0107] Furthermore, a composite network was constructed from the three SNP correlation networks to describe the final SNP correlation relationship; specifically, it includes:

[0108] Composite Network N scn Defined as:

[0109] N scn =N ssn +w 11 N sgd +w 12 N sld +w 13 N smd +w 21 N sg

[0110] where w 11 , w 12 , w 13 , w 21 They are all parameters, and their specific values ​​can be determined through parameter analysis.

[0111] Furthermore, clustering is performed on the composite network so that similar SNPs are grouped into the same cluster and dissimilar SNPs are grouped into different clusters. Specifically, the following steps are performed:

[0112] S1051, initially, each SNP in the composite network is considered as a community;

[0113] S1052, for each SNPi, calculate the change in the entire module gain if SNPi is removed from the current community and added to the community where its neighbor node SNPj is currently located. The community module gain ΔQ is defined as:

[0114]

[0115] Among them, ∑ in is the sum of the weight values ​​of the connections within the module, ∑ tot is the sum of the weights of the connections where the SNPs are located within the module, k i is the sum of the weight values ​​of the connections where SNPi is located, k i,in is the sum of the weight values ​​of the connections from the SNPi to the SNPs in the module, and m is the sum of the weight values ​​of all connections in the entire network.

[0116] Then, the SNPi is placed in the community where the neighboring node with the largest and positive gain is located; this process is repeated until the community to which the SNP belongs no longer changes;

[0117] S1053, using the communities in S1052 to define a new network: each community is regarded as a node, and the sum of the weight values ​​of the connections between two communities is regarded as the weight value of the edge.

[0118] S1052 and S1053 are repeatedly executed until the network no longer changes or the maximum number of iterations is reached.

[0119] Furthermore, high-dimensional SNP combinations associated with the disease are detected from each cluster, including:

[0120] The chi-square test was used to calculate the p-values ​​corresponding to all SNP combinations in each cluster, and the p-values ​​less than The SNP combination output is used as the detection result of the algorithm. Among them, α0 is pre-set, M is the number of SNPs, and K is the dimension of the combination.

[0121] The present disclosure first constructs a statistical SNP correlation network based on the SNP information of diseased samples and non-diseased samples; then, based on the known correlation between SNPs and biomolecules and the correlation between biomolecules and diseases, multiple third-order SNP correlation networks are constructed; then, based on the correlation between SNPs and genes, the similarity between SNPs is measured using the similarity between gene ontologies to construct a second-order SNP correlation network; then, multiple SNP correlation networks are weighted to form a final composite network to describe the association between SNPs; then, clustering is performed based on the composite network to divide the complete SNP set into multiple SNP clusters, and finally, within each cluster, a chi-square test is used to select SNP combinations that have a significant correlation with the disease. The present disclosure incorporates information related to a variety of heterogeneous biomolecules and SNPs and diseases, making the measurement of the association between SNPs more accurate and better interpretable.

[0122] Example 2:

[0123] The purpose of this embodiment is to provide an electronic device.

[0124] An electronic device includes a memory, a processor, and a computer program stored and running on the memory, wherein the processor performs the following steps when executing the program:

[0125] Obtain SNP sequences of diseased and non-diseased individuals and preprocess the data;

[0126] Construct a statistical SNP correlation network based on the SNP sequence of the sample and the disease status of the sample;

[0127] Based on the known correlation between SNPs and biomolecules, as well as the correlation between biomolecules and diseases, a third-order SNP correlation network is constructed;

[0128] Based on the known correlation between SNPs and genes, as well as the functional correlation between genes, a second-order SNP correlation network is constructed;

[0129] Based on the statistical SNP correlation network, the third-order SNP correlation network and the second-order SNP correlation network, a composite network for describing SNP correlation relationships is constructed;

[0130] The SNP sequences are clustered based on the composite network, and high-dimensional SNP combinations associated with the disease are detected from each cluster.

[0131] Example 3:

[0132] The purpose of this embodiment is to provide a non-transitory computer-readable storage medium.

[0133] A non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0134] Obtain SNP sequences of diseased and non-diseased individuals and preprocess the data;

[0135] Construct a statistical SNP correlation network based on the SNP sequence of the sample and the disease status of the sample;

[0136] Based on the known correlation between SNPs and biomolecules, as well as the correlation between biomolecules and diseases, a third-order SNP correlation network is constructed;

[0137] Based on the known correlation between SNPs and genes, as well as the functional correlation between genes, a second-order SNP correlation network is constructed;

[0138] Based on the statistical SNP correlation network, the third-order SNP correlation network and the second-order SNP correlation network, a composite network for describing SNP correlation relationships is constructed;

[0139] The SNP sequences are clustered based on the composite network, and high-dimensional SNP combinations associated with the disease are detected from each cluster.

[0140] Example 4:

[0141] The purpose of this embodiment is to provide a computer program product.

[0142] A computer program product comprising a computer program which, when executed on one or more processors, performs the following steps:

[0143] Obtain SNP sequences of diseased and non-diseased individuals and preprocess the data;

[0144] Construct a statistical SNP correlation network based on the SNP sequence of the sample and the disease status of the sample;

[0145] Based on the known correlation between SNPs and biomolecules, as well as the correlation between biomolecules and diseases, a third-order SNP correlation network is constructed;

[0146] Based on the known correlation between SNPs and genes, as well as the functional correlation between genes, a second-order SNP correlation network is constructed;

[0147] Based on the statistical SNP correlation network, the third-order SNP correlation network and the second-order SNP correlation network, a composite network for describing SNP correlation relationships is constructed;

[0148] The SNP sequences are clustered based on the composite network, and high-dimensional SNP combinations associated with the disease are detected from each cluster.

[0149] The above embodiment provides a SNP interaction detection system based on a heterogeneous biomolecular network, which is feasible and has broad application prospects.

[0150] The foregoing description is merely a preferred embodiment of the present disclosure and is not intended to limit the present disclosure. Those skilled in the art will readily appreciate that various modifications and variations are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present disclosure shall be included within the scope of protection of the present disclosure.

Claims

1. A SNP interaction detection system based on a heterogeneous biomolecular network, characterized in that: include: A data pre-acquisition module is configured to: acquire SNP sequences of diseased individuals and non-diseased individuals, and pre-process the data; A statistical network construction module is configured to: construct a statistical SNP correlation network based on the SNP sequence of the sample and the disease condition of the sample; The third-order SNP correlation network construction module is configured to: construct a third-order SNP correlation network based on the known correlation between SNPs and biomolecules, and the correlation between biomolecules and diseases. The specific steps include: Based on the verified association between SNPs and genes, miRNAs, and lncRNAs, a matrix describing the association is constructed; Based on the verified associations between diseases and genes, miRNAs, and lncRNAs, a matrix describing the associations is constructed; Based on the SNP-to-SNP paths implied in the matrix, the SNP correlation network is constructed; The second-order SNP correlation network construction module is configured as follows: based on the known correlation between SNPs and genes, as well as the functional correlation between genes, only the p most similar SNPs of each SNP are retained as its neighbors to construct the second-order SNP correlation network. Specifically: Determine the mapping relationship between SNPs, genes and gene ontology; Calculate the similarity between gene ontologies; Convert it into the similarity between SNPs to build a SNP-related network; A composite network generation module is configured to: construct a composite network for describing SNP correlation based on the statistical SNP correlation network, the third-order SNP correlation network, and the second-order SNP correlation network; A composite network was constructed from the three SNP correlation networks to describe the final SNP correlation relationship; specifically, it includes: Composite Network N scn Defined as: where w 11 , w 12 , w 13 , w 21 They are all parameters, and their specific values ​​are determined through parameter analysis; a clustering module, configured to: cluster the SNP sequences based on the composite network; The interaction detection module is configured to detect high-dimensional SNP combinations associated with the disease from each cluster.

2. A SNP interaction detection system based on a heterogeneous biomolecular network as claimed in claim 1, characterized in that: The method of constructing a statistical SNP correlation network based on the SNP sequence of the sample and the disease condition of the sample specifically includes: For the data in the sample set, the number of diseased individuals and the number of non-diseased individuals under each genotype in each SNP combination were counted; Based on statistical data, the association degree between each two SNPs was evaluated based on preset evaluation indicators; For each SNP, only the t SNPs with the strongest association are retained as neighbors with the SNP, where t is an integer greater than 0.

3. A SNP interaction detection system based on a heterogeneous biomolecular network as claimed in claim 1, characterized in that: Clustering of SNP sequences based on the composite network is specifically as follows: Initially, each SNP is considered as a community; For each SNP node, calculate the module gain of the network if it is moved to the cluster where the neighboring node is located; take the module with the largest and positive gain as the module that the SNP will move into; Repeat this process until no SNP-to-community changes again; Treat each community as a node and repeat the above process until the entire network meets the preset constraints; Ultimately, each community represents a cluster, and SNPs in the same community belong to the same cluster.

4. A SNP interaction detection system based on a heterogeneous biomolecular network as claimed in claim 3, characterized in that: The preset constraint condition is: the entire network no longer changes or the maximum number of iterations is reached.

5. The SNP interaction detection system based on heterogeneous biomolecular networks as claimed in claim 1, characterized in that: Detect high-dimensional SNP combinations associated with the disease from each cluster. Specifically, use the chi-square test to calculate the p-value corresponding to each SNP combination, so as to select the SNP combination with a significant correlation with the disease.

6. An electronic device comprising a memory, a processor, and a computer program stored and running on the memory, characterized in that: When the processor executes the program, the following steps are performed: Obtain SNP sequences of diseased and non-diseased individuals and preprocess the data; Construct a statistical SNP correlation network based on the SNP sequence of the sample and the disease status of the sample; Based on the known correlations between SNPs and biomolecules, as well as the correlations between biomolecules and diseases, a third-order SNP correlation network is constructed. The specific steps include: Based on the verified association between SNPs and genes, miRNAs, and lncRNAs, a matrix describing the association is constructed; Based on the verified associations between diseases and genes, miRNAs, and lncRNAs, a matrix describing the associations is constructed; Based on the SNP-to-SNP paths implied in the matrix, the SNP correlation network is constructed; Based on the known correlation between SNPs and genes, as well as the functional correlation between genes, only the p most similar SNPs of each SNP are retained as its neighbors to construct a second-order SNP correlation network, specifically: Determine the mapping relationship between SNPs, genes and gene ontology; Calculate the similarity between gene ontologies; Convert it into the similarity between SNPs to build a SNP-related network; Based on the statistical SNP correlation network, the third-order SNP correlation network and the second-order SNP correlation network, a composite network for describing the SNP correlation relationship is constructed, and a composite network is constructed from the three SNP correlation networks to describe the final SNP correlation relationship; specifically including: Composite Network N scn Defined as: where w 11 , w 12 , w 13 , w 21 They are all parameters, and their specific values ​​are determined through parameter analysis; The SNP sequences are clustered based on the composite network, and high-dimensional SNP combinations associated with the disease are detected from each cluster.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by the processor, the following steps are performed: Obtain SNP sequences of diseased and non-diseased individuals and preprocess the data; Construct a statistical SNP correlation network based on the SNP sequence of the sample and the disease status of the sample; Based on the known correlations between SNPs and biomolecules, as well as the correlations between biomolecules and diseases, a third-order SNP correlation network is constructed. The specific steps include: Based on the verified association between SNPs and genes, miRNAs, and lncRNAs, a matrix describing the association is constructed; Based on the verified associations between diseases and genes, miRNAs, and lncRNAs, a matrix describing the associations is constructed; Based on the SNP-to-SNP paths implied in the matrix, the SNP correlation network is constructed; Based on the known correlation between SNPs and genes, as well as the functional correlation between genes, only the p most similar SNPs of each SNP are retained as its neighbors to construct a second-order SNP correlation network, specifically: Determine the mapping relationship between SNPs, genes and gene ontology; Calculate the similarity between gene ontologies; Convert it into the similarity between SNPs to build a SNP-related network; Based on the statistical SNP correlation network, the third-order SNP correlation network and the second-order SNP correlation network, a composite network for describing the SNP correlation relationship is constructed, and a composite network is constructed from the three SNP correlation networks to describe the final SNP correlation relationship; specifically including: Composite Network N scn Defined as: where w 11 , w 12 , w 13 , w 21 They are all parameters, and their specific values ​​are determined through parameter analysis; The SNP sequences are clustered based on the composite network, and high-dimensional SNP combinations associated with the disease are detected from each cluster.

8. A computer program product comprising a computer program, characterized in that The computer program, when executed on one or more processors, performs the following steps: Obtain SNP sequences of diseased and non-diseased individuals and preprocess the data; Construct a statistical SNP correlation network based on the SNP sequence of the sample and the disease status of the sample; Based on the known correlations between SNPs and biomolecules, as well as the correlations between biomolecules and diseases, a third-order SNP correlation network is constructed. The specific steps include: Based on the verified association between SNPs and genes, miRNAs, and lncRNAs, a matrix describing the association is constructed; Based on the verified associations between diseases and genes, miRNAs, and lncRNAs, a matrix describing the associations is constructed; Based on the SNP-to-SNP paths implied in the matrix, the SNP correlation network is constructed; Based on the known correlation between SNPs and genes, as well as the functional correlation between genes, only the p most similar SNPs of each SNP are retained as its neighbors to construct a second-order SNP correlation network, specifically: Determine the mapping relationship between SNPs, genes and gene ontology; Calculate the similarity between gene ontologies; Convert it into the similarity between SNPs to build a SNP-related network; Based on the statistical SNP correlation network, the third-order SNP correlation network and the second-order SNP correlation network, a composite network for describing the SNP correlation relationship is constructed, and a composite network is constructed from the three SNP correlation networks to describe the final SNP correlation relationship; specifically including: Composite Network N scn Defined as: where w 11 , w 12 , w 13 , w 21 They are all parameters, and their specific values ​​are determined through parameter analysis; The SNP sequences are clustered based on the composite network, and high-dimensional SNP combinations associated with the disease are detected from each cluster.

Images