Drug relocation method based on multi-view stacked jump graph convolutional network

Abstract

The invention belongs to the technical field of biomedicine, and particularly relates to a drug relocation method based on a multi-view stacked skip graph convolutional network, which comprises the following specific steps of: acquiring drug and disease associated data, and respectively forming an adjacent matrix A and an associated graph G of drugs and diseases, and similarity matrixes Sr and Sd of the drugs and the diseases, a multi-layer graph convolution encoder block is used to encode the similarity matrix of the drugs and the diseases into high-order feature representation; and constructing a double-level sub-graph self-attention network, and performing structure-level fusion with a graph topological structure by combining the characteristics of a node level and a sub-graph level. When the high-order node information is used, the context information of the high-order node can be extracted, and the inconsistency of local information and global information is relieved; according to the method, deep information can be utilized more comprehensively, and the problem that the prediction capability of the model is reduced due to the fact that drug disease nodes cannot be distinguished after the network is in multiple layers is solved.

Description

Technical Field

[0001] The present invention relates to the field of biomedical technology, and in particular to a drug relocalization method based on a multi-view stacked skip-level graph convolutional network. Background Art

[0002] In the past few decades, despite the increasing investment in the biomedical field, the number of new treatments and drugs approved by the US Food and Drug Administration (FDA) has not been proportional to it. This situation has prompted researchers and pharmaceutical companies to look for more economical and faster drug development pathways. Drug repositioning strategies have attracted attention because of their ability to shorten drug development cycles and reduce costs.

[0003] Traditional drug repositioning methods mainly rely on experimental research and clinical observation, which are time-consuming and costly. With the development of bioinformatics and computing technology, computational drug repositioning methods have emerged, including those based on text mining, network analysis and machine learning.

[0004] Graph neural network (GNN), as a deep learning algorithm, can efficiently process graph structure data and capture the complex relationships and topological structures between nodes in the drug-disease network. It shows potential in predicting the potential association between drugs and diseases. However, existing methods have limitations, as follows:

[0005] 1. The long-distance node context information of drugs and diseases in the graph cannot be effectively utilized, resulting in a lack of consistency between global and local information in feature learning;

[0006] 2. Too much attention is paid to the information aggregation of connected nodes, ignoring the global information and high-order dependencies of the graph; in order to improve the prediction accuracy, the number of network layers will be increased, but too many layers will lead to the homogenization of the node features of drugs and diseases, affecting the prediction accuracy;

[0007] 3. The sparsity and imbalance of drug-disease connections further weaken the predictive ability.

[0008] Therefore, a drug relocalization method based on a multi-view stacked skip-level graph convolutional network is invented; by comprehensively understanding the complex relationship between drug-disease nodes on the graph from multiple perspectives, the accuracy of predicting the potential association between drugs and diseases is improved, which has important practical application value. Summary of the invention

[0009] To solve the above technical problems, according to one aspect of the present invention, the present invention provides the following technical solutions:

[0010] A drug relocalization method based on a multi-view stacked skip-level graph convolutional network includes the following specific steps:

[0011] S1: Obtain drug-disease association data, and construct the adjacency matrix A of drugs and diseases, the association graph G, and the similarity matrix S of drugs and diseases. r , S d , and use a multi-layer graph convolutional encoder block to encode the similarity matrix of drugs and diseases into a high-order feature representation;

[0012] S2: Construct a two-level subgraph self-attention network, combine the features of node level and subgraph level, and perform structural fusion with the graph topology structure;

[0013] S3: Construct a multi-view stacking skip-level sensing network, perform multi-view stacking learning on the association graph G, skip-level sense deep feature information in multi-layer graph convolution, and learn the horizontal features of nodes;

[0014] S4: The two-level sub-graph self-attention network and the multi-view stacked skip-level induction network are integrated, and a negative example distance balancing method is designed to enhance negative examples to optimize network performance;

[0015] S5: Use the optimized network to predict potential interactions between drugs and diseases.

[0016] As a preferred solution of the drug relocalization method based on multi-view stacked skip-level graph convolutional network described in the present invention, the specific steps of S1 are as follows:

[0017] S11: Obtain drug-disease association data, and construct the adjacency matrix A of drugs and diseases, the association graph G, and the similarity matrix S of drugs and diseases respectively. r , S d ;

[0018] S12: S r , S d Input to the multi-layer graph convolutional encoder block F encode , the specific form is as follows:

[0019]

[0020] Among them, σ is the activation function, v is the node, is the neighbor node of v, is the feature of node v, is the feature of neighbor node u, W l is the learnable weight matrix, are the degrees of node v and its neighbor node u respectively;

[0021] S13: Obtain drug high-order features and disease high-order features respectively, concatenate the two features together, and output drug-disease high-order features X.

[0022] As a preferred solution of the drug relocalization method based on multi-view stacked skip-level graph convolutional network described in the present invention, the specific steps of S2 are as follows:

[0023] S21: Use the METIS software package to split the drug-disease association graph G into multiple equally divided subgraph sets P G , and use the high-order features X obtained by S1 to extract two-level features, namely the sub-map level and the node level X sub , X node , and use graph convolutional networks to extract structural features X strut ;

[0024] S22: Structurally fuse node-level features, sub-layer-level features, and structure-level features, and output the fused feature X bi .

[0025] As a preferred solution of the drug relocalization method based on multi-view stacked skip-level graph convolutional network described in the present invention, the specific steps of S21 are as follows:

[0026] S211: Get sub-layer level feature X sub ;

[0027] S212: Obtain node level feature X node , the specific steps are as follows:

[0028]

[0029] Among them, X is the high-order feature of drug-disease, d X is the drug-disease node vector dimension;

[0030] S213: Get structure-level feature X strut .

[0031] As a preferred solution of the drug relocalization method based on multi-view stacked skip-level graph convolutional network described in the present invention, the specific steps of S211 are as follows:

[0032] S2111: Use the METIS tool to split the drug-disease association graph G into multiple equally divided subgraph sets P G ;

[0033] S2112: Pool the node features in each subgraph so that the nodes in each subgraph share one feature. The specific form is as follows:

[0034]

[0035] in, is the feature of the nodes contained in each subgraph, X sis the sub-graph feature;

[0036] S2113: Extract sub-graph hierarchical features using sub-graph self-attention:

[0037]

[0038] Among them, X s is the subgraph feature, is the subgraph vector dimension.

[0039] As a preferred solution of the drug relocalization method based on multi-view stacked skip-level graph convolutional network described in the present invention, the specific steps of S213 are as follows:

[0040] S2131: Copy the subgraph hierarchical features and assign them to all nodes under the same subgraph:

[0041] X' sub =repeat(X sub )

[0042] S2132: Construct the graph convolutional network combination block GCN_F to extract structural level features:

[0043]

[0044] X strut =GCN_F(X' sub +X node )

[0045] in, is the feature of the disease node associated with the drug, is the feature of the drug node, W is the learnable weight matrix, are the degrees of disease node d and drug node r, respectively.

[0046] As a preferred solution of the drug relocalization method based on multi-view stacked skip-level graph convolutional network described in the present invention, the specific steps of S22 are as follows:

[0047] The node-level features X obtained in the above steps are node , sub-layer level feature X sub and structural feature X strut Perform structural level fusion and output the fused feature X bi :

[0048] X bi =fusion(X' sub ,X node ,X strut )

[0049] Among them, fusion is structural-level fusion, and the specific operation is to add the corresponding elements of each feature.

[0050] As a preferred solution of the drug repositioning method based on the multi-view stacked skip graph convolutional network described in the present invention, wherein: the specific steps of S3 are as follows:

[0051] S31: Generate multiple channels by combining a random mask and a graph convolutional network block;

[0052] S32: Define a domain skip induction mechanism to extract deep node information;

[0053] S33: Stack and splice all views, and output the feature X mv ;

[0054] The specific steps of S31 are as follows:

[0055] S311: Perform random masking on the drug-disease high-order feature X in S1 for the following different views:

[0056] X' = qX

[0057] Among them, 0 < q < 1 is the random masking ratio, and X' is the masked high-order feature;

[0058] S312: Construct GCN_H as a graph convolutional network block to learn deep graph network information:

[0059]

[0060] Among them, is the feature of the disease node associated with the drug, is the feature of the drug node, W l is a learnable weight matrix, are the degrees of the disease node d and the drug node r respectively;

[0061] S313: Design a hierarchical representation for skip induction in the channel, and repeat L times to generate L layers:

[0062] Layer i = σ(BatchNorm(GraphSizeNorm((GCN_H(GCN_H(X'))))))

[0063] Among them, GraphSizeNorm is graph normalization, specifically by dividing the feature of each node by the square root of the number of nodes in the graph, and BatchNorm is batch normalization, specifically by normalizing the feature using the mean and variance in the current batch;

[0064] The specific steps of S32 are as follows:

[0065] S321: Using the characteristics of self-attention mechanism and graph convolutional network, design a domain skipping induction mechanism, including Layer i Perform pooling, calculate the hierarchical self-attention scores, and select the K highest scores as the aggregated levels:

[0066] P i =Pooling(Layer i )

[0067]

[0068] TopLayer=select(score topK (Layer i ))

[0069] Among them, P i To pool the node features of each layer, score is used to calculate the attention score, TopLayer is the feature set of the K layers before the score, and d P is the level vector dimension;

[0070] S322: Fusing the selected skip-level graph convolutional network layers with the drug-disease high-order feature X described in S1:

[0071]

[0072] Among them, X is the high-order feature of drug-disease, Σ k∈K TopLayer is the selected deep features and;

[0073] The specific steps of S33 are as follows:

[0074] Using Channel i As a basis, feature fusion is performed through view stacking and S2 cross-domain features. At the same time, in order to ensure the differences between each view, different features are aggregated separately. The operation is as follows:

[0075] VIEW i =GCN_H(Channel i +VIEW i-1 +X bi )

[0076] VIEW i =GCN_H(Channel i +VIEW i-1 +X node +X sub )

[0077] Among them, X bi , X node , X sub Features extracted from the two-level subgraph self-attention network in S2;

[0078] Finally, multiple views are concatenated as the output feature X mv :

[0079]

[0080] Among them, VIEW i For different views, A splicing symbol.

[0081] As a preferred solution of the drug relocalization method based on multi-view stacked skip-level graph convolutional network described in the present invention, the specific steps of S4 are as follows:

[0082] S41: Fusion of two-level sub-graph self-attention network and multi-view stacked skip-sensing network;

[0083] S42: Design negative example distance balancing method for unbalanced negative examples;

[0084] The specific steps of S41 are as follows:

[0085] Fusion of the two-layer subgraph self-attention network and the multi-view stacked skip-sensing network:

[0086] X final =αX mv +(1-α)X bi

[0087] Among them, α is a hyperparameter, X bi , X mv Features extracted for the two networks respectively;

[0088] The specific steps of S42 are as follows:

[0089] S421: Calculate the mean of positive examples, the mean of negative examples, the cosine similarity between each negative example and the mean of positive examples, the cosine similarity between each negative example and the mean of negative examples, the Pearson correlation coefficient between each negative example and the mean of positive examples, and the Pearson correlation coefficient between each negative example and the mean of negative examples under the same batch of training;

[0090] S422: In the cosine similarity calculation result in S421, if the cosine similarity between the negative example and the mean of the negative example is less than the cosine similarity between the negative example and the mean of the positive example, and the cosine similarity between the negative example and the mean of the positive example is greater than 0, record it as true;

[0091] S423: In the result of calculating the Pearson correlation coefficient in S421, if the Pearson correlation coefficient between the negative example and the mean of the negative example is less than the Pearson correlation coefficient between the negative example and the mean of the positive example, and the Pearson correlation coefficient between the negative example and the mean of the positive example is greater than 0, record it as true;

[0092] S424: When both S422 and S423 are true, distance balancing is used for negative examples. The specific operations are as follows:

[0093] Link' neg =Link neg +WLink pos_mean

[0094] Where W is the learnable weight matrix, Link pos_mean is the positive example mean, Link neg is a negative example.

[0095] As a preferred solution of the drug relocalization method based on multi-view stacked skip-level graph convolutional network described in the present invention, the specific steps of S5 are as follows:

[0096] Use trained positive and negative examples to predict potential interactions between drugs and diseases

[0097]

[0098] Pred=MLP(Link)

[0099] in, is a concatenation operation, MLP is a multi-layer perceptron, and Link is the positive and negative drug-disease samples in the same batch of training.

[0100] Compared with existing technologies:

[0101] When using high-order node information, the present invention can extract contextual information of high-order nodes, thereby alleviating the inconsistency between local information and global information; the present invention can more comprehensively utilize deep information, thereby solving the problem that the network cannot distinguish between drug-disease nodes after multiple layers, thereby causing a decrease in the prediction ability of the model; the present invention can more accurately distinguish between positive and negative examples, thereby solving the problem of data imbalance, which causes the problem that the model cannot accurately identify positive and negative examples. BRIEF DESCRIPTION OF THE DRAWINGS

[0102] Figure 1 Schematic diagram of the main structure of the multi-view stacked skip-level graph convolutional network of the present invention;

[0103] Figure 2 This is a schematic diagram of the overall process of the multi-view stacked skip-level graph convolutional network of the present invention;

[0104] Figure 3It is a schematic diagram of the dual-level subgraph self-attention network process of the present invention;

[0105] Figure 4 This is a schematic diagram of the multi-view stacking hopping sensing network process of the present invention;

[0106] Figure 5 Schematic diagram of the negative distance balance method of the present invention. DETAILED DESCRIPTION

[0107] In order to make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0108] The present invention provides a drug relocalization method based on multi-view stacked skip-level graph convolutional network, which is used to predict the potential interaction between drugs and diseases; it includes constructing a two-level subgraph self-attention network, a view stacked skip-level induction network, and designing a negative example distance balance method. Figure 1-Figure 5 , including the following specific steps:

[0109] S1: Obtain drug-disease association data, and construct the adjacency matrix A of drugs and diseases, the association graph G, and the similarity matrix S of drugs and diseases. r , S d , and use a multi-layer graph convolutional encoder block to encode the similarity matrix of drugs and diseases into a high-order feature representation;

[0110] The specific steps of S1 are as follows:

[0111] S11: Obtain drug-disease association data, and construct the adjacency matrix A of drugs and diseases, the association graph G, and the similarity matrix S of drugs and diseases respectively. r , S d ;

[0112] S12: S r , S d Input to the multi-layer graph convolutional encoder block F encode , the specific form is as follows:

[0113]

[0114] Among them, σ is the activation function, v is the node, is the neighbor node of v, is the feature of node v, is the feature of neighbor node u, W l is the learnable weight matrix, are the degrees of node v and its neighbor node u respectively;

[0115] S13: Obtain drug high-order features and disease high-order features respectively, concatenate the two features together, and output drug-disease high-order features X;

[0116] S1 includes but is not limited to the following embodiments:

[0117] In the experiment of the present invention, a B-dataset data set was obtained, which consists of 269 drug nodes, 598 disease nodes, and 18416 drug-disease association pairs; the molecular fingerprint and Gaussian interaction distribution (GIP) kernel similarity were used to calculate the similarity between drugs, and the average value was used as the similarity matrix of the drugs; the phenotype and GIP similarity were used to calculate the similarity between diseases, and the average value was used as the similarity matrix of the diseases; the drug similarity matrix scale was 269×269, and the disease similarity matrix scale was 598×598; the two similarity matrices were respectively passed through the constructed multi-layer graph convolution encoder block to extract their respective high-order features, and spliced ​​together as the input of subsequent network learning, with a scale of 867×128, that is, a total of 867 drug and disease nodes, and each node feature was 128 dimensions;

[0118] S2: Construct a two-level subgraph self-attention network, combine the features of node level and subgraph level, and perform structural fusion with the graph topology; The present invention designs a two-level subgraph self-attention network to integrate the local and global information of drug disease nodes, and the process is: extract the node level features and subgraph level features of drug diseases and perform structural fusion; integrate local and global multi-level features;

[0119] The specific steps of S2 are as follows:

[0120] S21: Use the METIS software package to split the drug-disease association graph G into multiple equally divided subgraph sets P G , and use the high-order features X obtained by S1 to extract two-level features, namely the sub-map level and the node level X sub , X node , and use graph convolutional networks to extract structural features X strut ; Use the high-level drug-disease features obtained from S1 to construct two-level features and use graph convolutional networks to extract structural features;

[0121] The specific steps of S21 are as follows:

[0122] S211: Get sub-layer level feature X sub ;

[0123] The specific steps of S211 are as follows:

[0124] S2111: Use the METIS tool to split the drug-disease association graph G into multiple equally divided subgraph sets P G;

[0125] S2112: Pool the node features in each subgraph so that the nodes in each subgraph share one feature. The specific form is as follows:

[0126]

[0127] in, is the feature of the nodes contained in each subgraph, X s is the sub-graph feature;

[0128] S2113: Extract sub-graph hierarchical features using sub-graph self-attention:

[0129]

[0130] Among them, X s is the subgraph feature, is the subgraph vector dimension;

[0131] Among them, METIS (Multilevel Partitioning of Graphs and Matrices) is a widely used toolkit that focuses on solving the partitioning and graph partitioning problems of large-scale data sets;

[0132] S212: Obtain node level feature X node , the specific steps are as follows:

[0133]

[0134] Among them, X is the high-order feature of drug-disease, d X is the drug-disease node vector dimension;

[0135] S213: Get structure-level feature X strut ;

[0136] The specific steps of S213 are as follows:

[0137] S2131: Copy the subgraph hierarchical features and assign them to all nodes under the same subgraph:

[0138] X' sub =repeat(X sub )

[0139] S2132: Construct the graph convolutional network combination block GCN_F to extract structural level features:

[0140]

[0141] X strut =GCN_F(X' sub +Xnode )

[0142] in, is the feature of the disease node associated with the drug, is the feature of the drug node, W l is the learnable weight matrix, are the degrees of disease node d and drug node r respectively;

[0143] S22: Structurally fuse node-level features, sub-layer-level features, and structure-level features, and output the fused feature X bi ;

[0144] The specific steps of S22 are as follows:

[0145] The node-level features X obtained in the above steps are node , sub-layer level feature X sub and structural feature X strut Perform structural level fusion and output the fused feature X bi :

[0146] X bi =fusion(X' sub ,X node ,X strut )

[0147] Among them, fusion is a structural level fusion, and the specific operation is to add the corresponding elements of each feature;

[0148] S2 includes but is not limited to the following embodiments:

[0149] The drug-disease high-order feature X obtained by S1 is used as input, with a scale of 867×128, indicating a total of 867 nodes (including drug and disease nodes), and the feature of each node is 128 dimensions; first, node self-attention is used to extract node-level features, with a scale of 867×128. Secondly, METIS is used to divide the drug-disease association graph into 128 subgraphs, with the same number of nodes in each subgraph; the node features in the 128 subgraphs are pooled as the features of each subgraph, with a scale of 128×128; the subgraph self-attention mechanism is used to extract subgraph-level features, with a scale of 128×128; the subgraph-level features are copied to each node so that the nodes in the subgraph have the same features, with a scale of 867x128; the subgraph-level features and the node-level features are fused, and the structure-level features are extracted through the custom GCN_F. Finally, the structure-level features are structurally fused with the subgraph-level features and the node-level features as the output features of the current network;

[0150] S3: Construct a multi-view stacked skip induction network to perform multi-view stacked learning on the association graph G, skip and induce deep feature information in multi-layer graph convolution, and learn the horizontal features of nodes. The present invention designs a multi-view stacked skip induction network for extracting deep feature information from multiple perspectives of nodes. Its steps include generating multiple channels using a random mask and a graph convolutional network combination block; defining a domain skip induction mechanism to extract deep node information; stacking and splicing all views, and outputting the feature X mv ;

[0151] The specific steps of S3 are as follows:

[0152] S31: Generate multiple channels using a random mask and a graph convolutional network combination block; the generated channels extract the horizontal features of drug-disease nodes from multiple perspectives;

[0153] The specific steps of S31 are as follows:

[0154] S311: Perform random masking on the drug-disease high-order feature X in S1 for the following different views:

[0155] X' = qX

[0156] where 0 < q < 1 is the random masking ratio, and X' is the masked high-order feature;

[0157] S312: Construct GCN_H as a graph convolutional network combination block for learning deep information of the graph network:

[0158]

[0159] where, is the feature of the disease node associated with the drug, is the feature of the drug node, W l is a learnable weight matrix, are the degrees of the disease node d and the drug node r respectively;

[0160] S313: Design the hierarchical representation for skip induction in the channel, and repeat L times to generate L layers:

[0161] Layer i = σ(BatchNorm(GraphSizeNorm((GCN_H(GCN_H(X'))))))

[0162] where GraphSizeNorm is graph normalization, specifically by dividing the feature of each node by the square root of the number of nodes in the graph, and BatchNorm is batch normalization, specifically by normalizing the feature using the mean and variance in the current batch;

[0163] S32: Define domain skipping sensing mechanism to extract deep node information;

[0164] The specific steps of S32 are as follows:

[0165] S321: Using the characteristics of self-attention mechanism and graph convolutional network, design a domain skipping induction mechanism, including Layer i Perform pooling, calculate the hierarchical self-attention scores, and select the K highest scores as the aggregated levels:

[0166] P i =Pooling(Layer i )

[0167]

[0168] TopLayer=select(score topK (Layer i ))

[0169] Among them, P i To pool the node features of each layer, score is used to calculate the attention score, TopLayer is the feature set of the K layers before the score, and d P is the level vector dimension;

[0170] S322: Fusing the selected skip-level graph convolutional network layers with the drug-disease high-order feature X described in S1:

[0171]

[0172] Among them, X is the high-order feature of drug-disease, ∑ k∈K TopLayer is the selected deep features and;

[0173] S33: Stack the views and concatenate all views to output feature X mv ;

[0174] The specific steps of S33 are as follows:

[0175] Using Channel i As a basis, feature fusion is performed through view stacking and S2 cross-domain features. At the same time, in order to ensure the differences between each view, different features are aggregated separately. The operation is as follows:

[0176] VIEW i =GCN_H(Channel i +VIEW i-1 +X bi )

[0177] VIEW i =GCN_H(Channel i +VIEW i-1 +X node +X sub )

[0178] Among them, X bi , X node , X sub Features extracted from the two-level subgraph self-attention network in S2;

[0179] Finally, multiple views are concatenated as the output feature X mv :

[0180]

[0181] Among them, VIEW i For different views, is the splicing symbol;

[0182] S3 includes but is not limited to the following embodiments:

[0183] The high-order feature X obtained by S1 is randomly masked with a ratio of 0.2 and distributed into 4 channels; each channel extracts deep features of 6 layers of GCN_H, and each layer undergoes GCN_H, graph normalization and batch normalization; retain the features of each layer, use self-attention to calculate the importance of the features of each layer, and use the top 3 scores as the number of layers for skip-level sensing; therefore, the final scale is 4×867×32, representing 4 channels, 867 nodes, and 32 as the feature dimension; each channel aggregates different sensing levels; each channel fuses the features extracted by S2 and the features of the previous channel to generate a view stack; finally, the 4 channels are spliced ​​to a scale of 867×128 as the output features of this network;

[0184] S4: The two-level subgraph self-attention network and the multi-view stacked skip-level induction network are integrated, and a negative example distance balancing method is designed to enhance negative examples to optimize network performance; the features extracted by the two networks of S3 and S4 are integrated, and the deviation of negative examples is measured by cosine similarity and Pearson correlation coefficient to balance negative examples;

[0185] The specific steps of S4 are as follows:

[0186] S41: Fusion of two-level sub-graph self-attention network and multi-view stacked skip-sensing network;

[0187] S42: Design a negative example distance balancing method for unbalanced negative examples; the operation of the negative example distance balancing method is as follows Figure 5 As shown;

[0188] The specific steps of S41 are as follows:

[0189] Fusion of the two-layer subgraph self-attention network and the multi-view stacked skip-sensing network:

[0190] X final =αX mv +(1-α)X bi

[0191] Among them, α is a hyperparameter, X bi , X mv Features extracted for the two networks respectively;

[0192] The specific steps of S42 are as follows:

[0193] S421: Calculate the mean of positive examples, the mean of negative examples, the cosine similarity between each negative example and the mean of positive examples, the cosine similarity between each negative example and the mean of negative examples, the Pearson correlation coefficient between each negative example and the mean of positive examples, and the Pearson correlation coefficient between each negative example and the mean of negative examples under the same batch of training;

[0194] S422: In the cosine similarity calculation result in S421, if the cosine similarity between the negative example and the mean of the negative example is less than the cosine similarity between the negative example and the mean of the positive example, and the cosine similarity between the negative example and the mean of the positive example is greater than 0, record it as true;

[0195] S423: In the result of calculating the Pearson correlation coefficient in S421, if the Pearson correlation coefficient between the negative example and the mean of the negative example is less than the Pearson correlation coefficient between the negative example and the mean of the positive example, and the Pearson correlation coefficient between the negative example and the mean of the positive example is greater than 0, record it as true;

[0196] S424: When both S422 and S423 are true, distance balancing is used for negative examples. The specific operations are as follows:

[0197] Link' neg =Link neg +WLink pos_mean

[0198] Where W is the learnable weight matrix, Link pos_mean is the positive example mean, Link neg is a negative example;

[0199] S4 includes but is not limited to the following embodiments:

[0200] First, the features of the two networks are fused, and α is taken as 0.7 to obtain drug-disease features with a scale of 867×128. There are 18416 associations in the drug-disease association graph. According to the drug-disease association graph, features are obtained for each positive example and negative example. Associated drug-disease pairs are taken as positive examples, and unassociated drug-disease pairs are taken as negative examples, and finally 18416×128 positive example features and 18416×128 negative example features are obtained. The mean of all negative example features and the mean of positive example features are calculated, and the cosine similarity and Pearson correlation coefficient of each negative example with the negative example mean and the positive example mean are compared. If the cosine similarity and Pearson correlation coefficient of the negative example and the positive example mean are both greater than 0, and the cosine similarity and Pearson correlation coefficient of the negative example and the positive example mean are both greater than the cosine similarity and Pearson correlation coefficient of the negative example and the positive example mean, then the positive example mean of the learnable weight matrix is ​​fused for this negative example to expose the imbalance of negative examples.

[0201] Among them, in this network, positive examples are samples with a connection relationship between drugs and diseases, and negative examples are samples with no connection relationship between drugs and diseases;

[0202] S5: Use the optimized network to predict potential interactions between drugs and diseases;

[0203] The specific steps of S5 are as follows:

[0204] Use trained positive and negative examples to predict potential interactions between drugs and diseases

[0205]

[0206] Pred=MLP(Link)

[0207] in, is a concatenation operation, MLP is a multi-layer perceptron, and Link is the positive and negative drug-disease samples in the same batch of training.

[0208] Based on the above, a multi-view stacked skip-level graph convolutional network is designed in the present invention, and its structure includes two networks and one method: a two-level sub-graph self-attention network, a multi-view stacked skip-level induction network and a negative example distance balancing method;

[0209] The multi-view stacked skip-level graph convolutional network is used for the complex relationships and topological structures between nodes in the drug-disease network; the two-level subgraph self-attention network extracts node-level features, subgraph-level features and structure-level fusion, captures the local and global semantic information of drugs and diseases on the graph and the contextual information of long-distance nodes, and effectively alleviates the inconsistency of global and local information; the multi-view stacked skip-level induction network stacks multiple views and simultaneously skips and aggregates different high-order neighbor information, effectively utilizing the deep information of nodes from multiple aspects, solving the problem of homogenization of node features of drugs and diseases due to too many network layers, and improving the prediction accuracy; the negative example distance balance method adjusts the weights through the relationship between negative examples and positive examples, alleviates the sparsity and imbalance of drug and disease connections, and improves the network's ability to distinguish samples.

[0210] Although the present invention has been described above with reference to the embodiments, various modifications may be made thereto and parts thereof may be replaced by equivalents without departing from the scope of the present invention. In particular, as long as there is no structural conflict, the various features in the embodiments disclosed in the present invention may be used in combination with each other in any manner, and the fact that these combinations are not exhaustively described in this specification is only for the sake of omitting space and saving resources. Therefore, the present invention is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A drug relocalization method based on multi-view stacked skip-level graph convolutional network, characterized in that: The specific steps are as follows: S1: Obtain drug-disease association data, and construct the adjacency matrix A of drugs and diseases, the association graph G, and the similarity matrix S of drugs and diseases. r , S d , and use a multi-layer graph convolutional encoder block to encode the similarity matrix of drugs and diseases into a high-level feature representation; S2: Construct a two-level subgraph self-attention network, combine the features of the node level and the sublayer level, and perform structural-level fusion with the graph topology structure; S3: Construct a multi-view stacked skip-connection induction network, perform multi-view stacked learning on the associated graph G, skip-connect and sense deep feature information in multi-layer graph convolutions, and learn the horizontal features of nodes; S4: Integrate the two-level subgraph self-attention network and the multi-view stacked skip-connection induction network, and design a negative example distance balancing method to enhance negative examples to optimize the network performance; S5: Use the optimized network to predict the potential interactions between drugs and diseases.

2. The drug relocalization method based on multi-view stacked skip-level graph convolutional network according to claim 1, characterized in that: The specific steps of S1 are as follows: S11: Obtain drug-disease association data, and construct the adjacency matrix A of drugs and diseases, the association graph G, and the similarity matrix S of drugs and diseases respectively. r , S d ; S12: S r , S d Input to the multi-layer graph convolutional encoder block F encode , the specific form is as follows: Among them, σ is the activation function, v is the node, is the neighbor node of v, is the feature of node v, is the feature of neighbor node u, W l is the learnable weight matrix, are the degrees of node v and its neighbor node u respectively; S13: Obtain the drug high-order features and the disease high-order features respectively, concatenate the two types of features together, and output the drug-disease high-order feature X.

3. The drug relocalization method based on multi-view stacked skip-level graph convolutional network according to claim 1, characterized in that: The specific steps of S2 are as follows: S21: Use the METIS software package to split the drug-disease association graph G into multiple equally divided subgraph sets P G , and use the high-order features X obtained by S1 to extract two-level features, namely the sub-map level and the node level X sub , X node , and use graph convolutional networks to extract structural features X strut ; S22: Structurally fuse node-level features, sub-map-level features, and structure-level features, and output the fused feature X bi .

4. The drug relocalization method based on multi-view stacked skip-level graph convolutional network according to claim 3, characterized in that: The specific steps of S21 are as follows: S211: Get sub-layer level feature X sub ; S212: Obtain node level feature X node , the specific steps are as follows: Among them, X is the high-order feature of drug-disease, d X is the drug-disease node vector dimension; S213: Get structure-level feature X strut .

5. The drug relocalization method based on multi-view stacked skip-level graph convolutional network according to claim 4, characterized in that: The specific steps of S211 are as follows: S2111: Use the METIS tool to split the drug-disease association graph G into multiple equally divided subgraph sets P G ; S2112: Pool the node features in each subgraph so that the nodes within each subgraph share one feature. The specific form is as follows: in, is the feature of the nodes contained in each subgraph, X s is the sub-graph feature; S2113: Use subgraph self-attention to extract sublayer-level features: Among them, X s is the subgraph feature, is the subgraph vector dimension.

6. The drug relocalization method based on multi-view stacked skip-level graph convolutional network according to claim 5, characterized in that: The specific steps of S213 are as follows: S2131: Copy the sublayer-level features and assign them to all nodes under the same subgraph: X' sub =repeat(X sub ) S2132: Construct a graph convolutional network combination block GCN_F to extract structural-level features: X strut =GCN_F(X' sub +X node ) in, is the feature of the disease node associated with the drug, is the feature of the drug node, W l is the learnable weight matrix, are the degrees of disease node d and drug node r, respectively.

7. The drug relocalization method based on multi-view stacked skip-level graph convolutional network according to claim 6, characterized in that: The specific steps of S22 are as follows: The node-level features X obtained in the above steps are node , sub-layer level feature X sub and structural feature X strut Perform structure-level fusion and output the fused feature Xbi: X bi =fusion(X' sub ,X node ,X strut ) Among them, fusion is structural-level fusion, and the specific operation is to add the corresponding elements of each feature.

8. The drug relocalization method based on multi-view stacked skip-level graph convolutional network according to claim 1, characterized in that: The specific steps of S3 are as follows: S31: Use random masking and a graph convolutional network combination block to generate multiple channels; S32: Define a domain skip-connection induction mechanism to extract node deep information; S33: Stack the views and concatenate all views to output feature X mv ; The specific steps of S31 are as follows: S311: Perform random masking on the drug-disease high-order feature X in S1 for the following different views: X' = qX where 0 < q < 1 is the random masking ratio, and X' is the masked high-order feature; S312: Construct GCN_H as a graph convolutional network combination block to learn the deep information of the graph network: in, is the feature of the disease node associated with the drug, is the feature of the drug node, W l is the learnable weight matrix, are the degrees of disease node d and drug node r respectively; S313: Design a hierarchical representation for skip-connection induction in the channel, and repeat it L times to generate L layers: Layer i =σ(BatchNorm(GraphSizeNorm((GCN_H(GCN_H(X')))))) Among them, GraphSizeNorm is graph normalization. Specifically, it divides the feature of each node by the square root of the number of nodes in the graph. BatchNorm is batch normalization. Specifically, it normalizes the feature using the mean and variance in the current batch; The specific steps of S32 are as follows: S321: Using the self-attention mechanism and the characteristics of graph convolutional networks, design a domain skipping induction mechanism, including Layer i Perform pooling, calculate the hierarchical self-attention scores, and select the K highest scores as the aggregated levels: P i =Pooling(Layer i ) TopLayer=select(score topK (Layer i )) Among them, P i To pool the node features of each layer, score is used to calculate the attention score, TopLayer is the feature set of the K layers before the score, and d P is the level vector dimension; S322: Integrate the selected hierarchical levels of the skip-connection graph convolutional network with the drug-disease high-order feature X described in S1: Among them, X is the high-order feature of drug-disease, ∑ k∈K TopLayer is the selected deep features and; The specific steps of S33 are as follows: Using Channel i As a basis, feature fusion is performed through view stacking and S2 cross-domain features. At the same time, in order to ensure the differences between each view, different features are aggregated separately. The operation is as follows: VIEW i =GCN_H(Channel i +VIEW i-1 +X bi ) VIEW i =GCH_H(Channel i +VIEW i-1 +X node +X sub ) Among them, X bi , X node , X sub Features extracted from the two-level subgraph self-attention network in S2; Finally, multiple views are concatenated as the output feature X mv : Among them, VIEW i For different views, A splicing symbol.

9. The drug relocalization method based on multi-view stacked skip-level graph convolutional network according to claim 1, characterized in that: The specific steps of S4 are as follows: S41: Integrate the two-level subgraph self-attention network and the multi-view stacked skip-connection induction network; S42: Design a negative example distance balancing method for unbalanced negative examples; The specific steps of S41 are as follows: Integrate the two-level subgraph self-attention network and the multi-view stacked skip-connection induction network: X final =αX mv +(1-a)X bi Among them, α is a hyperparameter, X bi , X mv Features extracted for the two networks respectively; The specific steps of S42 are as follows: S421: Calculate the mean of positive examples, the mean of negative examples, the cosine similarity between each negative example and the mean of positive examples, the cosine similarity between each negative example and the mean of negative examples, the Pearson correlation coefficient between each negative example and the mean of positive examples, and the Pearson correlation coefficient between each negative example and the mean of negative examples under the same batch of training; S422: In the cosine similarity calculation result in S421, if the cosine similarity between the negative example and the mean of the negative example is less than the cosine similarity between the negative example and the mean of the positive example, and the cosine similarity between the negative example and the mean of the positive example is greater than 0, record it as true; S423: In the result of calculating the Pearson correlation coefficient in S421, if the Pearson correlation coefficient between the negative example and the mean of the negative example is less than the Pearson correlation coefficient between the negative example and the mean of the positive example, and the Pearson correlation coefficient between the negative example and the mean of the positive example is greater than 0, record it as true; S424: When both S422 and S423 are true, distance balancing is used for negative examples. The specific operations are as follows: Link' neg =Link neg +WLink pos_mean Where W is the learnable weight matrix, Link pos_mean is the positive example mean, Link neg is a negative example.

10. The drug relocalization method based on multi-view stacked skip-level graph convolutional network according to claim 1, characterized in that: The specific steps of S5 are as follows: Use trained positive and negative examples to predict potential interactions between drugs and diseases Pred=MLP(Link) in, is a concatenation operation, MLP is a multi-layer perceptron, and Link is the positive and negative drug-disease samples in the same batch of training.

Images