Synthetic lethal gene pair prediction method and device based on graph convolutional neural network, terminal and medium

Abstract

The invention discloses a synthetic lethal gene pair prediction method and device based on a graph convolutional neural network, a terminal and a medium. The synthetic lethal gene pair prediction method comprises the following steps: obtaining structural characteristics of protein based on protein structure data; obtaining sequence characteristics of the protein based on the protein sequence data; obtaining functional characteristics of the protein based on the protein-protein interaction network; combining and standardizing the structural features, the sequence features and the functional features of the protein to obtain gene features of main generation genes of the protein; obtaining interaction among genes, and training a synthetic lethal gene pair prediction model based on a graph convolutional neural network according to the interaction among the genes and gene features; and obtaining the final feature representation of each gene based on the trained synthetic lethal gene pair prediction model, and predicting whether the two genes are the synthetic lethal gene pair or not according to the final feature representation. According to the method, the feature extraction efficiency and the gene interaction prediction capability are improved.

Description

Technical Field

[0001] The present invention relates to the field of bioinformatics, and in particular to a method, device, terminal and medium for predicting synthetic lethal gene pairs based on a graph convolutional neural network. Background Art

[0002] Among the many methods for cancer treatment, targeted therapy of the SL gene of oncogenic genes based on the principle of synthetic lethality (SL) can effectively kill cancer cells without damaging normal cell functions. Figure 1 As shown in the figure, synthetic lethality means that there are some gene pairs in human DNA, which together guide the completion of some life activities necessary for cells. When any one of the genes is inactivated, it will not have a fatal effect on the cell, and the life activity will be taken over by the remaining gene. However, when both are inactivated, the cell will no longer be able to complete its necessary life activities, leading to cell death. For cancer patients, when the main pathogenic gene produced by cancer cells in their bodies can be identified, the synthetic lethal gene of the gene can be attacked to carry out a highly targeted attack on cancer cells. The first PARP inhibitor made using the principle of synthetic lethality was approved for clinical treatment in 2014, which inhibits the repair of tumor cell DNA damage and promotes apoptosis of tumor cells.

[0003] Computational methods based on machine learning learn gene features from large amounts of data and construct prediction models to identify SL gene pairs. These methods can be divided into traditional machine learning methods and deep learning methods. Traditional machine learning methods mainly include support vector machines (SVM), random forests (RF), and logistic regression. These methods integrate multiple data sources, such as gene expression data, protein interaction networks (PPIs), and gene ontology (GO), to construct feature vectors, and use machine learning algorithms for training and prediction. For example, the SL2MF method extracts gene features from the gene differential expression similarity matrix through matrix decomposition technology to predict potential SL interactions. The advantage of these methods is that they can handle large-scale data sets, but their disadvantages are that they require manual feature engineering and are easily affected by noisy data.

[0004] Deep learning methods automatically learn features from data by building complex network architectures, effectively capturing the complex relationships between genes. Graph convolutional neural networks (GCNs) learn low-dimensional representations of genes through message passing on gene interaction networks, thereby predicting SL gene pairs. The GCN model can automatically learn features from graphs, avoiding the tedious feature engineering in traditional machine learning methods. The DDGCN model first applied graph neural networks (GNNs) to SL prediction, and adopted a GCN strategy with a double dropout mechanism to effectively solve the problem of data sparsity and prevent overfitting. The KG4SL model is based on the knowledge graph (KG) and learns gene representations by sampling subgraphs around a single gene in the KG. The SLGNN model focuses on gene preferences in various relationships in the knowledge graph and designs a GNN model to effectively learn gene representations.

[0005] Although the above methods have achieved good results and interpretability in SL prediction, they are still limited to the feature representation of genetic data and ignore the possible role of protein-related data.

[0006] Therefore, the existing technology still needs to be improved and developed. Summary of the invention

[0007] In view of the above-mentioned deficiencies in the prior art, the object of the present invention is to provide a method, device, terminal and medium for predicting synthetic lethal gene pairs based on a graph convolutional neural network, aiming to solve the problem that the existing synthetic lethal gene pair prediction method does not take into account protein-related data, resulting in low prediction accuracy.

[0008] The technical solution of the present invention is as follows:

[0009] In a first aspect, a method for predicting synthetic lethal gene pairs based on a graph convolutional neural network is provided, comprising:

[0010] Obtaining structural features of proteins based on protein structure data;

[0011] Obtaining protein sequence features based on protein sequence data;

[0012] Obtaining functional characteristics of proteins based on protein-protein interaction networks;

[0013] The structural features, sequence features and functional features of proteins are combined and standardized to obtain the gene features of the main gene producing the protein;

[0014] Obtain the interactions between genes, and use the interactions between genes and gene features to train a synthetic lethal gene pair prediction model based on a graph convolutional neural network;

[0015] The final feature representation of each gene is obtained based on the trained synthetic lethal gene pair prediction model, and whether two genes are a synthetic lethal gene pair is predicted based on the final feature representation.

[0016] The preferred technical solution, wherein the structural characteristics of the protein are obtained based on the protein structure data, specifically includes:

[0017] Alphafold2 is used to convert the protein sequence into a three-dimensional structure of the protein, and the three-dimensional structure is simplified into a contact map of the protein structure. The contact map is then processed using the node2vec method to obtain the structural characteristics of the protein.

[0018] A preferred technical solution, wherein the step of obtaining the sequence characteristics of a protein based on the protein sequence data specifically includes:

[0019] The seq2vec method is used to extract the semantic information of protein sequences and generate feature embedding to obtain the sequence features of proteins.

[0020] The preferred technical solution, wherein the functional characteristics of proteins are obtained based on the protein-protein interaction network, specifically includes:

[0021] The node2vec method is used to extract the semantic information of nodes in the protein-protein interaction network and generate feature embeddings to obtain the functional characteristics of proteins.

[0022] The preferred technical solution is that obtaining the interaction between genes specifically includes:

[0023] The interactions between genes were obtained from the SynlethDB database. Gene pairs with synthetic lethal relationships were used as positive samples, and gene pairs without synthetic lethal relationships were used as negative samples. Since the number of negative samples was less than the number of positive samples, a generative adversarial network was used to generate negative samples so that the ratio of positive to negative samples was 1:1.

[0024] The preferred technical solution, wherein the synthetic lethal gene pair prediction model based on graph convolutional neural network is trained based on the interaction between genes and gene characteristics, specifically includes:

[0025] Genes are nodes, interactions between genes are edges between nodes, and gene features are node features;

[0026] set up is the initial feature vector of node i, A is the adjacency matrix of the graph, D is the degree matrix, I is the identity matrix, and the graph convolution operation of the graph convolutional neural network model is expressed as:

[0027] in, It is obtained by adding self-connection to the original adjacency matrix A. yes The degree matrix, W (k) is the weight matrix of the kth layer, and σ is the nonlinear activation function.

[0028] The preferred technical solution, wherein the prediction of whether two genes are a synthetic lethal gene pair based on the final feature representation specifically includes:

[0029] The feature vectors of the two gene nodes u and v and Combined, a linear layer is used to predict whether they are a synthetic lethal gene pair. The predicted probability that gene nodes u and v are a synthetic lethal gene pair is expressed as:

[0030] In a second aspect, a synthetic lethal gene pair prediction device based on a graph convolutional neural network is provided, comprising:

[0031] A first acquisition module is configured to acquire structural features of the protein based on the protein structure data;

[0032] A second acquisition module is configured to acquire sequence features of proteins based on protein sequence data;

[0033] a third acquisition module, configured to acquire functional characteristics of proteins based on a protein-protein interaction network;

[0034] A fusion module is configured to merge and standardize the structural features, sequence features, and functional features of the protein to obtain the gene features of the main gene producing the protein;

[0035] A model building module is configured to obtain the interactions between genes and train a synthetic lethal gene pair prediction model based on a graph convolutional neural network using the interactions between genes and gene features;

[0036] The prediction module is configured to obtain a final feature representation of each gene based on the trained synthetic lethal gene pair prediction model, and predict whether two genes are a synthetic lethal gene pair based on the final feature representation.

[0037] In a third aspect, a terminal is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for predicting synthetic lethal gene pairs based on a graph convolutional neural network as described in the first aspect is implemented.

[0038] In a fourth aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the synthetic lethal gene pair prediction method based on graph convolutional neural network described in the first aspect.

[0039] Beneficial effects: Compared with the prior art, the present invention provides a method for predicting synthetic lethal gene pairs based on graph convolutional neural networks. This method comprehensively considers the interactions between proteins and the mutual influence or dependency of the genes they produce, which not only improves the efficiency of feature extraction, but also enhances the prediction model's ability to predict gene interactions. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 is a schematic diagram of the synthetic lethality process.

[0041] Figure 2 Schematic diagram of the synthetic lethal gene pair prediction model based on graph convolutional neural network. DETAILED DESCRIPTION

[0042] The present invention provides a method, device, terminal and medium for predicting synthetic lethal gene pairs based on a graph convolutional neural network. In order to make the objectives, technical solutions and effects of the present invention clearer and more specific, the present invention is further described in detail below.

[0043] Protein is the product of gene, and the effect of gene on biological traits directly depends on the protein produced by it. The inventor believes that the interaction between proteins can reflect the mutual influence or dependence of the genes that produce them, that is, if there is a synthetic lethal effect between two genes, it is actually caused by the simultaneous inactivation of the proteins produced by the two genes. This means that protein is one of the direct actors of synthetic lethality, and the characteristic description of protein should be able to well reflect the relevant mechanism of synthetic lethality. Alphafold2 is a completely open source prediction model for protein spatial structure based on deep learning, which provides the most accurate prediction effect to date (Journal of chemical information and modeling, 2021, 61 (10): 4827-4831.). The spatial structure of protein is closely related to its biological effect, and the characterization of protein spatial structure should be able to describe the effect of protein. Based on this, the present invention proposes an improved synthetic lethal gene pair prediction method based on graph convolutional neural network, named Struct2SL, which comprehensively considers the characteristic characterization performance of protein sequence, PPI network and three-dimensional structure of protein on gene interaction, so as to better predict potential SL pairs.

[0044] The embodiment of the present invention provides a method for predicting synthetic lethal gene pairs based on a graph convolutional neural network, comprising:

[0045] Obtaining structural features of proteins based on protein structure data;

[0046] Obtaining protein sequence features based on protein sequence data;

[0047] Obtaining functional characteristics of proteins based on protein-protein interaction networks;

[0048] The structural features, sequence features and functional features of proteins are combined and standardized to obtain the gene features of the main gene producing the protein;

[0049] Obtain the interactions between genes, and use the interactions between genes and gene features to train a synthetic lethal gene pair prediction model based on a graph convolutional neural network;

[0050] The final feature representation of each gene is obtained based on the trained synthetic lethal gene pair prediction model, and whether two genes are a synthetic lethal gene pair is predicted based on the final feature representation.

[0051] In one embodiment, the step of obtaining the structural features of a protein based on protein structural data specifically includes:

[0052] Alphafold2 is used to convert the protein sequence into a three-dimensional structure of the protein, and the three-dimensional structure is simplified into a contact map of the protein structure. The contact map is then processed using the node2vec method to obtain the structural characteristics of the protein.

[0053] Specifically, for the structural characteristics of proteins, this embodiment uses Alphafold2 to predict the three-dimensional structure of proteins and simplifies the three-dimensional structure into a contact graph; the nodes in the contact graph are amino acids in the protein, and when the Euclidean distance between two amino acid molecules is less than 10 angstroms, it is recorded as contact between the two molecules, otherwise it is recorded as no contact; for the convenience of calculation, this embodiment flattens the contact graph using the node2vec method, obtains the node vector representation in the contact graph for each amino acid node, and then flattens the feature matrix to obtain the feature vector; since the number of amino acids required for the composition of different proteins is not the same, the length of the feature vector is padded with a trainable value, and finally a fully connected layer is used to compress the length of the feature vector to an acceptable range, thereby obtaining the structural characteristics of the protein.

[0054] First, alphafold2 was used to describe the three-dimensional structure of the protein. The pdb file data of the protein was output by alphafold2, and the Cα atomic coordinates in the three-dimensional structure of the protein were extracted. For each residue in the structure, there are: i =(x i ,y i ,z i ), where r i Represents the coordinate vector of the Cα atom of the ith residue. Using the coordinates of the Cα atom, calculate the distances between all residue pairs and construct the distance matrix D:

[0055]

[0056] Among them, D ij represents the Euclidean distance between the ith and jth residues.

[0057] Next, the protein sequence is converted to a one-hot encoding representation to capture the composition information of the sequence. For a sequence containing V different amino acids, the one-hot encoding vector of the i-th amino acid is defined as:

[0058]

[0059] Then, according to the given threshold cmap_thresh, the distance matrix is ​​converted into an adjacency matrix A to represent the contact relationship in the protein structure:

[0060]

[0061] Finally, the adjacency matrix A and the one-hot encoding vector of the protein sequence are integrated together to form a graph representation of the protein structure. This graph representation can be used for subsequent graph convolutional network analysis to extract high-level features of the protein structure.

[0062] In one embodiment, the obtaining of sequence features of proteins based on protein sequence data specifically includes:

[0063] The seq2vec method is used to extract the semantic information of protein sequences and generate feature embedding to obtain the sequence features of proteins.

[0064] Specifically, a protein sequence is composed of different amino acids, and the protein sequence can be regarded as a "sentence", so that features can be extracted from the protein sequence using natural language processing methods. This embodiment uses the ELMo model with pre-trained parameters to extract features from the protein sequence. The ELMo model uses a deep bidirectional recurrent neural network to learn sub-vector representations in order to obtain richer semantic information.

[0065] For the sequence features of proteins, this embodiment uses the seq2vec method based on deep learning to automatically extract the semantic information of protein sequences and generate feature embedding. This method avoids the complexity of manually designing features and captures complex patterns in sequences by using pre-trained deep learning models.

[0066] During the seq2vec conversion process, the ELMo model is first used to obtain the embedding representation of the sequence. The ELMo model is a pre-trained deep bidirectional language model that can capture the semantics of each element (e.g., amino acid) in a sequence in different contexts.

[0067] First, the protein sequence file is built into a sequence dictionary. Then, a function for processing embedding is defined, which processes the embedding representation output by the Elmo model and calculates the average of all time steps to generate a single embedding vector for each protein:

[0068]

[0069] Among them, h t is the hidden state at time step t, and T is the length of the sequence.

[0070] After that, the embedding representation is updated layer by layer in an iterative manner to obtain higher-level features:

[0071]

[0072] in, represents the embedded representation of entity i at layer l, N(i) represents the set of neighbor entities of entity i, and aggregate is an aggregation function used to integrate the information of the current entity and its neighbors.

[0073] Finally, the final protein embedding is obtained by accumulating the embedding representations of all layers:

[0074]

[0075] Among them, L is the number of layers in the network.

[0076] In one embodiment, the method of obtaining the functional characteristics of a protein based on a protein-protein interaction network specifically includes:

[0077] The node2vec method is used to extract the semantic information of nodes in the protein-protein interaction network and generate feature embeddings to obtain the functional characteristics of proteins.

[0078] Specifically, when proteins play their roles in life activities, the interaction with other proteins occupies a major part. Therefore, the interaction between proteins can also be used to describe the functional characteristics of proteins.

[0079] In the stage of processing protein-protein interaction (PPI) network data, this embodiment uses a deep learning-based method, node2vec, to automatically extract the semantic information of nodes (proteins) in the network and generate feature embeddings. This method effectively captures the topological structure and neighborhood information of nodes in the network.

[0080] The node2vec method explores the network by simulating random walks to learn embedded representations of nodes. First, a starting node is selected from the network and a random walk is performed according to predefined transition probabilities. The transition probability is calculated by considering the neighborhood of the current node and parameters p and q. Specifically, for each neighbor of the current node, the probability of it being selected as the next node is calculated, which is proportional to the weight of the neighbor and is adjusted by the parameters p and q. Mathematically, the transition probability can be expressed as:

[0081]

[0082] Among them, w uv is the edge weight between nodes u and v, and N(v) is the set of neighbors of node v.

[0083] Next, multiple random walks are repeated from each node to generate a series of walk paths. Each walk path can be regarded as a series of node sequences that contain the neighborhood information of the node.

[0084] These walk paths are then used as input to learn the embedding representation of the nodes using the Word2Vec model. The Word2Vec model is a pre-trained word embedding model that learns the vector representation of words (or nodes) by optimizing the Skip-gram objective function. In this case, the node sequence in each walk path is regarded as a sequence of words in a sentence, and the embedding vector of each node is learned by the model.

[0085] Finally, the final node embedding is obtained by averaging the embedding vectors of each node in all walks. These embedding vectors capture the structural and semantic information of the nodes in the network and extract data features, namely the functional features of proteins, from the original PPI network data.

[0086] For each protein, this embodiment extracts its structural features, sequence features, and functional features. These three features are combined and standardized for multiple features to obtain the protein features of each protein. Considering that there is a main gene for each protein, this embodiment uses a one-to-one approach to map gene features with protein features to describe the feature vector of the gene. For the case where multiple proteins exist corresponding to a main gene, the features of the main protein produced by the gene indicated in the uniprot database are used to describe the gene features to ensure the rationality of the model.

[0087] In one embodiment, obtaining the interaction between genes specifically includes:

[0088] The interactions between genes were obtained from the SynlethDB database. Gene pairs with synthetic lethal relationships were used as positive samples, and gene pairs without synthetic lethal relationships were used as negative samples. Since the number of negative samples was less than the number of positive samples, a generative adversarial network was used to generate negative samples so that the ratio of positive to negative samples was 1:1.

[0089] Specifically, SynLethDB is a comprehensive database of synthetic lethal gene pairs. Since the amount of real data of negative samples in the SynLethDB database is small, this embodiment uses data enhancement based on a generative adversarial network (GAN), using a model architecture of a pair of generators (Generator) and discriminators (Discriminator) to generate realistic negative sample data through an adversarial process, thereby enhancing the original data set.

[0090] The goal of the generator is to generate fake samples that are sufficient to "fool" the discriminator. Given a random noise vector z, the generator G converts it into a fake data sample

[0091] Among them, z is random noise sampled from the standard normal distribution, G consists of 3 hidden layers with 256, 128 and 64 neurons respectively, each layer is followed by a ReLU activation function, and the last layer uses a Sigmoid activation function to ensure that the output is in the range [0,1].

[0092] The goal of the discriminator is to distinguish between real samples and fake samples generated by the generator. The discriminator D accepts an input sample x and outputs a probability value indicating the probability that the sample is a real sample:

[0093] D(x)=σ(Wx+b);

[0094] Among them, x is the input sample (can be a real sample or a fake sample), W and b are the parameters of the discriminator, σ is the Sigmoid activation function, and the output value is in the range [0,1].

[0095] The process of training a GAN involves minimizing a combined loss function consisting of two parts: the discriminator loss and the generator loss.

[0096] The discriminator loss is the discriminator's attempt to maximize its ability to correctly classify real samples and fake samples. The discriminator's loss function L D Defined as:

[0097]

[0098] Among them, the first term is about the loss of real samples, and the second term is about the loss of fake samples generated by the generator.

[0099] The generator loss is the generator trying to minimize the probability that the discriminator classifies the fake samples it generates as fake. The loss function L of the generator G Defined as:

[0100]

[0101] During the training process, the parameters of the discriminator and the generator are updated alternately. First, the parameters of the generator are fixed and the parameters of the discriminator are updated; then, the parameters of the discriminator are fixed and the parameters of the generator are updated. This process is repeated until a certain number of iterations is reached or the generated samples are realistic enough, that is, the error is small.

[0102] Through the above process, this embodiment can automatically learn the distribution characteristics of data from the original data and generate new data samples, thereby increasing the diversity and quantity of non-SL pair data sets, providing a richer and more accurate data basis for subsequent link prediction.

[0103] In one embodiment, the synthetic lethal gene pair prediction model based on graph convolutional neural network is trained based on gene interactions and gene features, specifically comprising:

[0104] Genes are nodes, interactions between genes are edges between nodes, and gene features are node features;

[0105] set up is the initial feature vector of node i, A is the adjacency matrix of the graph, D is the degree matrix, I is the identity matrix, and the graph convolution operation of the synthetic lethal gene pair prediction model based on the graph convolutional neural network is expressed as:

[0106]

[0107] in, It is obtained by adding self-connection to the original adjacency matrix A. yes The degree matrix, W (k) is the weight matrix of the kth layer, and σ is the nonlinear activation function.

[0108] Specifically, the graph convolutional neural network (GCN) is a deep learning model used to process graph structured data. It can retain the structural information between nodes when processing graph data, and can perform link prediction on the graph by embedding node attributes in feature engineering.

[0109] This example proposes to train a GCN-based synthetic lethal gene pair prediction model for predicting the links of synthetic lethal gene pairs. The core of the model is to automatically extract complex relationship features between genes, thereby effectively capturing the interactions and functional associations between genes.

[0110] First, genes are represented as nodes in the graph, and the interactions between genes are represented as edges between nodes. On this basis, the goal of GCN is to capture the structural information of the entire graph by iteratively updating the feature representation of the nodes, thereby predicting whether there is a link between gene pairs.

[0111] The core idea of ​​GCN is to update the feature representation of each node by aggregating the information of neighboring nodes. This process can be described by the following mathematical formula:

[0112] set up is the initial feature vector of node i, A is the adjacency matrix of the graph, D is the degree matrix, and I is the identity matrix. The graph convolution operation of GCN can be expressed as:

[0113]

[0114] in, It is obtained by adding self-connection to the original adjacency matrix A. yes The degree matrix, W (k) is the weight matrix of the kth layer, and σ is the nonlinear activation function.

[0115] This example defines a synthetic lethal gene pair prediction model based on a graph convolutional neural network. The construction process is as follows: Figure 2 As shown in Figure 2, the model implements the above graph convolution operation through a series of linear layers and nonlinear activation functions. For each hidden layer, firstly, the linear transformation W (k) Update the feature representation of the node. Next, apply a nonlinear activation function (ReLU) to introduce nonlinear factors and enhance the expressiveness of the model. Finally, regularize through the Dropout layer to prevent overfitting. This allows iterative updates of the feature representation of the node.

[0116] After multiple layers of graph convolution, the final feature representation of each node can be obtained Where L is the number of layers of GCN. These feature representations will be used for the link prediction task, that is, predicting whether there is a link between two nodes (genes).

[0117] In one embodiment, L=256, that is, the synthetic lethal gene pair prediction model based on graph convolutional neural network has 256 hidden layers.

[0118] In one embodiment, link prediction, i.e., predicting whether two genes are a synthetic lethal gene pair based on the final feature representation, specifically includes:

[0119] The feature vectors of the two gene nodes u and v and Combined, a linear layer or a nonlinear layer is used to predict the existence of the link between them, that is, whether it is a synthetic lethal gene pair. The calculation formula is:

[0120]

[0121] in, is the predicted probability that there is a link between gene nodes u and v, that is, the predicted probability that gene nodes u and v are a synthetic lethal gene pair.

[0122] In the model training process, this embodiment uses a binary cross-entropy loss function to optimize the model parameters and updates the weights through a back-propagation algorithm. In addition, this embodiment also introduces a regularization term, such as L2 regularization, to further prevent overfitting.

[0123] Through the above method, this embodiment can automatically extract complex relationship features between genes from graph structure data, complete link prediction tasks, and identify synthetic lethal gene pairs. This method not only improves the efficiency of feature extraction, but also enhances the model's ability to predict gene interactions.

[0124] In a specific embodiment, as described above, the synthetic lethal gene pair prediction method (Struct2SL) based on graph convolutional neural network uses the protein sequence in the Uniprot database, the PPI network in the STRING database and the SynlethDB database as original data, and converts the protein sequence into the three-dimensional structure of the protein through Alphafold2. For the obtained three-dimensional structure, it is converted into a contact map of the protein structure, and processed to obtain structural features. At the same time, the PPI network is characterized by nodes to obtain the structural features, sequence features and functional features of the protein.

[0125] Proteins are mapped one by one to their main generating genes, thereby describing the genetic characteristics of the genes. The SL network is reconstructed, node features (i.e., the gene features mentioned above) and edge features are added, and GAN is used to enhance the non-SL data at a ratio of 1:1 to obtain negative samples. The ratio of training set, validation set, and test set is 7:1:2. At the same time, the model learning rate is set to 0.004, and the early stopping strategy is adopted. For all baseline methods, the hyperparameters and model parameters described in the original paper are used. The experimental results are as follows:

[0126]

[0127] The original papers of the baseline methods are as follows:

[0128] DDGCN:Cai R,Chen X,Fang Y,Wu M,Hao Y.Dual-dropout graph convolutionalnetwork for predicting synthetic lethality in humancancers.Bioinformatics.2020Aug 15;36(16):4458-4465.doi:10.1093 / bioinformatics / btaa211.PMID:32221609.

[0129] KG4SL:Li J,Lu X,Jiang K,Tang D,Ning B,Sun F.TARSL:Triple-attentioncross-network representation learning to predict synthetic lethality foranti-cancer drug discovery.IEEE J Biomed Health Inform.2023Aug 21;PP.doi:10.1109 / JBHI.2023.3306768.Epub ahead of print.PMID:37603479.

[0130] TAR4SL:Wang S,Xu F,Li Y,Wang J,Zhang K,Liu Y,Wu M,Zheng J.KG4SL:knowledge graph neural network for synthetic lethality prediction in human cancers.Bioinformatics.2021Jul12;37(Suppl_1):i418-i425.doi:10.1093 / bioinformatics / btab271.PMID:34252965;PMCID:PMC8336442.

[0131] SLGNN: Zhu Y, Zhou Y, Liu Y, Wang X, Li J. SLGNN: synthetic lethality prediction in human cancers based on factor-aware knowledge graph neuralnetwork.Bioinformatics.2023Feb3;39(2):btad015.doi:10.1093 / bioinformatics / btad015.PMID:36645245;PMCID:PMC9907046.

[0132] The performance of Struct2SL is compared with the four baseline models described previously according to three evaluation metrics: AUC, AUPR, and F1-score. As shown in the table, Struct2SL outperforms all baseline methods. The AUC, AUPR, and F1-score of Struct2SL are 0.9764, 0.9808, and 0.9425, respectively, which are 1.3%, 1.0%, and 3.7% higher than the second-best model SLGNN, respectively.

[0133] From the above comparison results, it can be found that the GCN-based method uses the topological structure of the graph to propagate information and uses the similarity of known SL interactions to improve the model performance. However, the baseline methods basically only use the characteristics of the gene itself to distinguish the different relationships between genes when predicting SL. Among them, TAR4SL comprehensively considers the interaction relationship of the protein PPI network, but does not consider the characteristic information of the protein three-dimensional structure and protein sequence. It can be clearly seen from the comparison that Struct2SL comprehensively considers this and performs better in generating gene embeddings for SL interaction prediction.

[0134] Based on the same inventive concept, an embodiment of the present invention provides a synthetic lethal gene pair prediction device based on a graph convolutional neural network, comprising:

[0135] A first acquisition module is configured to acquire structural features of the protein based on the protein structure data;

[0136] A second acquisition module is configured to acquire sequence features of proteins based on protein sequence data;

[0137] a third acquisition module, configured to acquire functional characteristics of proteins based on a protein-protein interaction network;

[0138] A fusion module is configured to merge and standardize the structural features, sequence features, and functional features of the protein to obtain the gene features of the main gene producing the protein;

[0139] A model building module is configured to obtain the interactions between genes and train a synthetic lethal gene pair prediction model based on a graph convolutional neural network using the interactions between genes and gene features;

[0140] The prediction module is configured to obtain a final feature representation of each gene based on the trained synthetic lethal gene pair prediction model, and predict whether two genes are a synthetic lethal gene pair based on the final feature representation.

[0141] Based on the same inventive concept, an embodiment of the present invention provides a terminal, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for predicting synthetic lethal gene pairs based on a graph convolutional neural network as described above is implemented.

[0142] Based on the same inventive concept, an embodiment of the present invention provides a computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable a computer to execute the synthetic lethal gene pair prediction method based on graph convolutional neural network as described above.

[0143] In summary, the key to the present invention is:

[0144] 1. The characteristics of proteins are used to describe the characteristics of genes. Proteins are the products of gene expression, and the effects of genes on biological traits are directly dependent on the proteins they produce. The present invention believes that the interaction between proteins can reflect the mutual influence or dependency of the genes that produce them, that is, if there is a synthetic lethal effect between two genes, it is actually caused by the simultaneous inactivation of the proteins produced by the two genes.

[0145] 2. For protein feature extraction, the present invention obtains three aspects of features and integrates them, namely, sequence features extracted from protein sequence data, structural features extracted from protein structure data, and functional features extracted from PPI network. The three features are merged and standardized to obtain the genetic features of each protein.

[0146] 3. Considering the interaction between genes based on protein interactions. In addition to the similarity with the synthetic lethal gene itself, potential synthetic lethal gene pairs may also be related to the dependence of biological processes between the proteins produced by the synthetic lethal gene.

[0147] 4. The three-dimensional structure of the protein was described using Alphafold2. Using the pdb file data output by Alphafold2, the present invention extracted the Cα atomic coordinates in the protein structure. For each residue in the structure, the Euclidean distance between each residue was calculated and converted into an adjacency matrix through one-hot encoding, thereby obtaining a graphical representation of the protein structural features.

[0148] 5. For the functional characteristics of proteins, the verified PPI network data provided by the public database STRING was used, and its vector feature representation was obtained through the node2vec method.

[0149] 6. For link prediction, the present invention defines a GCN-based synthetic lethal gene pair prediction model. For each hidden layer, the feature representation of the node is updated by linear transformation. Next, ReLU is used as the activation function to introduce nonlinear factors and enhance the expressive power of the model. Finally, regularization is performed through the Dropout layer to prevent overfitting and iteratively update the feature representation of the node. After multiple layers of graph convolution, the final feature representation of each node is obtained for the final prediction task, that is, predicting whether there is a link between two nodes (genes). The final feature representations of the two gene nodes can be combined to predict the existence of a link between them through a linear layer (the output is the probability of synthetic lethality between the two gene nodes).

[0150] It should be understood that the application of the present invention is not limited to the above examples. For ordinary technicians in this field, improvements or changes can be made based on the above description. All these improvements and changes should fall within the scope of protection of the claims attached to the present invention.

Claims

1. A method for predicting synthetic lethal gene pairs based on graph convolutional neural networks, characterized in that: include: Obtaining structural features of proteins based on protein structure data; Obtaining protein sequence features based on protein sequence data; Obtaining functional characteristics of proteins based on protein-protein interaction networks; The structural features, sequence features and functional features of proteins are combined and standardized to obtain the gene features of the main gene producing the protein; Obtain the interactions between genes, and use the interactions between genes and gene features to train a synthetic lethal gene pair prediction model based on a graph convolutional neural network; The final feature representation of each gene is obtained based on the trained synthetic lethal gene pair prediction model, and whether two genes are a synthetic lethal gene pair is predicted based on the final feature representation.

2. The method for predicting synthetic lethal gene pairs based on graph convolutional neural network according to claim 1, characterized in that: The step of obtaining the structural features of the protein based on the protein structural data specifically includes: Alphafold2 is used to convert the protein sequence into a three-dimensional structure of the protein, and the three-dimensional structure is simplified into a contact map of the protein structure. The contact map is then processed using the node2vec method to obtain the structural characteristics of the protein.

3. The method for predicting synthetic lethal gene pairs based on graph convolutional neural network according to claim 1, characterized in that: The step of obtaining the sequence features of the protein based on the protein sequence data specifically includes: The seq2vec method is used to extract the semantic information of protein sequences and generate feature embedding to obtain the sequence features of proteins.

4. The method for predicting synthetic lethal gene pairs based on graph convolutional neural network according to claim 1, characterized in that: The method of obtaining the functional characteristics of proteins based on the protein-protein interaction network specifically includes: The node2vec method is used to extract the semantic information of nodes in the protein-protein interaction network and generate feature embeddings to obtain the functional characteristics of proteins.

5. The method for predicting synthetic lethal gene pairs based on graph convolutional neural network according to claim 1, characterized in that: The obtaining of the interaction between genes specifically includes: The interactions between genes were obtained from the SynlethDB database. Gene pairs with synthetic lethal relationships were used as positive samples, and gene pairs without synthetic lethal relationships were used as negative samples. Since the number of negative samples was less than the number of positive samples, a generative adversarial network was used to generate negative samples so that the ratio of positive to negative samples was 1:

1.

6. The method for predicting synthetic lethal gene pairs based on graph convolutional neural network according to claim 1, characterized in that: The synthetic lethal gene pair prediction model based on graph convolutional neural network is trained based on the interaction between genes and gene features, specifically including: Genes are nodes, interactions between genes are edges between nodes, and gene features are node features; set up is the initial feature vector of node i, A is the adjacency matrix of the graph, D is the degree matrix, I is the identity matrix, and the graph convolution operation of the synthetic lethal gene pair prediction model based on the graph convolutional neural network is expressed as: in, It is obtained by adding self-connection to the original adjacency matrix A. yes The degree matrix, W (k) is the weight matrix of the kth layer, and σ is the nonlinear activation function.

7. The method for predicting synthetic lethal gene pairs based on graph convolutional neural network according to claim 1, characterized in that: The predicting whether the two genes are a synthetic lethal gene pair according to the final feature representation specifically includes: The feature vectors of the two gene nodes u and v and Combined, a linear layer is used to predict whether they are a synthetic lethal gene pair. The predicted probability that gene nodes u and v are a synthetic lethal gene pair is expressed as:

8. A synthetic lethal gene pair prediction device based on graph convolutional neural network, characterized in that: include: A first acquisition module is configured to acquire structural features of the protein based on the protein structure data; A second acquisition module is configured to acquire sequence features of proteins based on protein sequence data; a third acquisition module, configured to acquire functional characteristics of proteins based on a protein-protein interaction network; A fusion module is configured to merge and standardize the structural features, sequence features, and functional features of the protein to obtain the gene features of the main gene producing the protein; A model building module is configured to obtain the interactions between genes and train a synthetic lethal gene pair prediction model based on a graph convolutional neural network using the interactions between genes and gene features; The prediction module is configured to obtain a final feature representation of each gene based on the trained synthetic lethal gene pair prediction model, and predict whether two genes are a synthetic lethal gene pair based on the final feature representation.

9. A terminal comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method for predicting synthetic lethal gene pairs based on graph convolutional neural network according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the synthetic lethal gene pair prediction method based on graph convolutional neural network according to any one of claims 1 to 7.

Images