178 results about "Gene expression matrix" patented technology

The invention discloses a space transcriptomics cell type deconvolution method based on a generative adversarial network and a comparative learning framework. The method comprises the following steps: firstly, enhancing a gene expression matrix by using a histological image and spatial position information; then, in combination with GraphSAGE and self-supervised comparative learning, optimizing feature representation of space transcriptomics data, and reconstructing high-quality space gene expression data; single-cell RNA sequencing data gene expression features are extracted by using a variational automatic encoder, and meanwhile, the quality of a generated sample is optimized by using a discriminator, so that efficient feature extraction and accurate gene expression reconstruction are realized; and finally, mapping single-cell RNA sequencing data into spatial points through comparative learning and mapping matrix learning based on low-rank representation, thereby predicting the cell type of each point. According to the model, multi-modal information in single-cell RNA sequencing and space transcriptomics data is comprehensively considered, and a mapping matrix is constructed and optimized, so that the cell type of each space point is predicted.

The invention discloses a graph convolution spatial domain identification method of a fusion gene interaction network, and relates to the technical field of biological information, and the method comprises the steps: preprocessing spatial transcriptome data; constructing a spatial diagram according to the spatial position information of the spot and the gene expression matrix; constructing a conditional gene interaction network based on at least a part of key genes in the spatial transcriptome data, constructing a conditional gene interaction network matrix, combining the conditional gene interaction network matrix and the gene expression matrix to obtain an expression degree-interaction matrix, and constructing a feature map; performing graph neural network convolution on the space graph and the feature graph, and introducing an attention mechanism to obtain final embedding; and clustering and finally embedding, and identifying a spatial domain with a similar gene expression mode. According to the method, the spatial domain is identified through the fusion gene interaction network, the problem that the spatial domain is discontinuous is effectively solved, the robustness and accuracy of the spatial domain identification method are improved, and high performance is shown in complex and highly heterogeneous tissue samples.

The invention discloses a single-cell multi-omics cell type annotation method based on pre-training, and belongs to the technical field of biological information, and the method comprises the following steps: obtaining a gene expression matrix of paired scRNA-seq data and a chromatin accessibility peak matrix of scATAC-seq data, and preprocessing the gene expression matrix and the chromatin accessibility peak matrix, then inputting a pre-trained feature extractor, and obtaining a query feature matrix fused with multiple omics by adopting a multi-head cross attention network of the pre-trained feature extractor; the feature extractor comprises a multi-head cross attention network and two classifiers; and inputting the query feature matrix into a classification model, performing data splitting by adopting a linear layer, then inputting the query feature matrix into a Transform encoder to perform multi-omics genetic feature extraction, and then inputting the query feature matrix into a classifier to perform cell type annotation. According to the cell type annotation method, the problem of deviation of cell type annotation in an existing method is solved.

The invention discloses a transcriptomics spatial domain identification method, and belongs to the technical field of transcriptomics. The objective of the invention is to solve the problems of low data noise reduction precision and poor recognition effect of an existing spatial transcriptional spatial domain recognition method. The method comprises the following steps: firstly, obtaining an undirected neighborhood graph according to a gene expression matrix, obtaining embedded representation of the gene expression matrix by utilizing an encoder, obtaining a corresponding reconstruction matrix by utilizing a decoder, and further determining reconstruction loss; meanwhile, a ZINB model is used for fitting a reconstruction matrix, and a ZINB loss function is obtained; then, an augmented graph is constructed based on the undirected neighborhood graph, respective embedded matrixes are obtained through an encoder, the comparison loss of the undirected neighborhood graph and the comparison loss of the augmented graph are obtained through a comparison representation learning mechanism, and then the neighbor comparison loss is obtained; total target loss is obtained based on all losses, a joint optimization strategy is adopted for training, and after training of the whole model is completed, dimensionality reduction and spatial domain recognition are carried out on a generated reconstruction matrix.

A single cell sequencing data interpolation method based on a biclustering and graph convolutional neural network comprises the following steps: 1) respectively downloading matrix-form gene expression data of human and mouse from a gene database, and screening out cells of which the total number of gene expressions is less than 3 and genes of which the total number of cell expressions is less than 200; 2) performing feature selection on the gene expression matrix, finding out cells with consistent growth trend, forming a new two-dimensional matrix, and constructing a cell connection diagram; 3) constructing a BiGCN single-cell interpolation model by using a graph convolutional neural network, learning a nonlinear relationship between cells, dividing a data set into a training set and a test set, training the model by using the training set, and reserving the trained BiGCN single-cell interpolation model; 4) inputting the test set into the trained BiGCN single cell interpolation model to obtain an interpolation result, and evaluating the interpolation result by using an evaluation index; the method has the characteristics that the feature extraction capability is improved, interference of redundant zero expression is avoided, data sparsity is effectively relieved, and the interpolation accuracy is improved.

The invention discloses a spatial domain identification method based on a multi-view weighted fusion GCN network, and the method comprises the steps: taking the gene expression and spatial position of spatial transcriptome data as input, and combining the spatial information based on different similarity measurement modeling with the gene expression after modeling to construct a multi-view weighted fusion graph convolutional network model; and reconstructing a gene expression matrix by using a decoder, capturing global information of ST data, calculating spatial regularization constraint loss by using similarity information and spatial neighbor information, and completing identification of a spatial transcriptome spatial domain. According to the method provided by the scheme, the model can extract the spatial information from the whole situation and the local situation, so that the spatial information is more fully and comprehensively mined, and accurate spatial domain identification is realized.

The invention provides a clustering method and system for transcriptome sequencing data. The method is applied to the technical field of data processing, and comprises the following steps: collecting cervical adenocarcinoma and para-carcinoma tissue specimens of a plurality of patients, and carrying out data preprocessing to obtain a standardized gene expression matrix; performing clustering operation on the standardized gene expression matrix based on a clustering analysis method combining a potential category model and sub-alliance division to obtain a clustered gene set; performing gene function annotation on the clustered gene set, and performing correlation analysis on a clustering result after function annotation and clinical characteristics of cervical adenocarcinoma to obtain correlation between gene expression and the clinical characteristics; and on the basis of clustering analysis and correlation analysis results, constructing a deep clustering prediction model based on a variational auto-encoder and a Gamma hybrid model so as to predict the prognosis risk or chemotherapy sensitivity of the patient. The problems that high-dimensional transcriptome data is poor in clustering stability and low in prognosis prediction precision are solved.

The invention relates to the technical field of bioinformatics, in particular to a data classification method and system for single cell sequencing. The method comprises the following steps: acquiring an original gene expression matrix; performing gene data preprocessing on the original gene expression matrix to obtain a preprocessed gene expression matrix; constructing a cell topological graph according to the preprocessed gene expression matrix to obtain the cell topological graph; performing cell division rate calculation according to the cell topological graph and the preprocessed gene expression matrix, and performing RNA rate calculation to obtain RNA rate data; adding node rates to the cell topological graph according to the cell division rate data and the RNA rate data to obtain an enhanced cell topological graph. According to the method, the dynamic graph is constructed and the change rule of the dynamic graph is analyzed, so that the dynamic change of the cell state is better captured, the sensitivity to parameters is reduced, a complex cell population structure is processed, and a quantitative cell heterogeneity evaluation index is provided.

The invention discloses a single cell sequencing data clustering method and system based on similarity constraint, and the method comprises the steps: obtaining a Fastq sequencing file, carrying out the quantitative analysis and data preprocessing, and obtaining a preprocessed gene expression matrix; performing similarity priori knowledge and non-similarity priori knowledge extraction processing on the preprocessed gene expression matrix, and constructing a similarity matrix and a non-similarity matrix; constructing a joint non-negative matrix factorization model according to the similarity matrix and the non-similarity matrix, and performing dimension reduction processing to obtain a common coefficient matrix; and carrying out hierarchical clustering and visualization processing to obtain a single cell sequencing data clustering result. According to the method, similarity and non-similarity constraints of a plurality of gene expression matrixes are mined, and the feature extraction capability of the model is improved, so that a better single cell sequencing data clustering effect is realized. The single cell sequencing data clustering method and system based on similarity constraint can be widely applied to the technical field of single cell sequencing data clustering.

The invention discloses a breast cancer neoadjuvant chemotherapy pyroptosis related prognosis model construction method which comprises the following steps: obtaining an original data set, and preprocessing to obtain an annotated whole-gene expression matrix and a survival clinical information matrix; screening out a gene database having significant differential expression on patient survival; obtaining a pyroptosis related gene database; taking an intersection of the pyroptosis related gene database and the gene database having significant differential expression on the survival of the patient to obtain a list of differential expression genes related to the survival of the pyroptosis; and carrying out disease survival prediction model batch construction on the differential expression gene list, screening out two optimal disease survival prediction models as sub-models, and weighting to obtain a prognosis model. According to the method, the problem that gene pathways of a prognosis model of the breast cancer neoadjuvant chemotherapy patient are deficient is solved, and the prognosis prediction capability of the breast cancer patient receiving the neoadjuvant chemotherapy is expected to be improved.

A gene regulation inference method based on causal diagram embedding and conditional cellular network relates to the technical field of gene regulation network prediction, and comprises the following steps: 1, obtaining a gene expression matrix from scRNA-seq, and generating a causal diagram; 2, generating a local feature embedding matrix of a gene by using a graph neural network model based on a causal graph and a known gene regulation and control network graph; 3, constructing a CCSN based on scRNA-seq, converting the CCSN into gene connectivity vectors, and integrating the gene connectivity vectors of all cells to form a CNDM as a global feature matrix; 4, integrating the local feature embedding matrix and the global feature matrix to form a final gene feature matrix; 5, screening a core gene from the gene feature matrix, and constructing a regulation edge matrix; and 6, inputting the regulatory edge matrix into a gene link prediction module to realize inference of the gene regulatory network. By applying the method, the causal relationship and the cell specificity can be integrated, the core gene is effectively screened, the feature fusion is optimized, and the accuracy and the biological interpretation of network inference are improved.

The invention provides a multi-scale causal discovery method and system based on collective behavior modeling, and relates to the technical field of causal discovery, and the method comprises the steps: obtaining single-cell RNA sequencing data, the single-cell RNA sequencing data being a gene expression matrix of a plurality of cells; single-cell RNA sequencing data is input into a multi-scale causal model for reasoning, a multi-scale causal diagram is obtained, and the multi-scale causal diagram is used for describing the multi-scale causal relationship between genes and between genes and gene programs; wherein the multi-scale causal model takes a single gene as an individual variable, through collective behavior modeling, the single gene is aggregated into a collective variable representing a gene program, and a multi-scale collaborative learning strategy is adopted to collaboratively learn the multi-scale causal relationship between individuals and between individuals and collective. According to the method, a collective behavior modeling framework is innovatively constructed, key collective behaviors driving state changes can be effectively identified, and the causal discovery accuracy, rationality and robustness are improved.

The invention discloses a genetic state cancer clustering analysis method based on deep learning, which comprises the following steps: S1, constructing a single cell sequencing data set which comprises a gene expression matrix; s2, constructing a deep neural network composed of an input layer, a hidden layer and an output layer to obtain converted low-dimensional high-variation gene expression features; s3, constructing a gridding preprocessing model, and performing networked preprocessing including grid segmentation and isolated point removal on the low-dimensional high-variation gene expression features; constructing a clustering algorithm model, inputting the preprocessed low-dimensional high-variation gene expression features into the clustering algorithm model, and designing an optimization strategy based on a target function by the clustering algorithm model to enhance label entropy on a KL divergence index and amplify the ratio weight of auxiliary probability distribution and clustering distribution of the gene features; and iteratively realizing clustering result optimization as the output of the clustering algorithm model. Compared with the prior art, the clustering precision can be improved.

Provided are a gene expression matrix optimization method, an electronic device, and a storage medium. The method comprises: acquiring cell data of a plurality of cells, wherein the cell data comprises an initial gene expression matrix of each of the plurality of cells; determining a preset number of nearest neighbor cells of a target cell among the plurality of cells on the basis of the cell data; determining a gene expression distance between the target cell and each nearest neighbor cell, determining a distance parameter on the basis of a target non-zero value among the gene expression distances, and determining, on the basis of the distance parameter, a weight parameter corresponding to each nearest neighbor cell; and on the basis of the weight parameter corresponding to each nearest neighbor cell and the initial gene expression matrix of each nearest neighbor cell, obtaining a target gene expression matrix obtained by smoothing of the target cell. By using the described method, the noise in a gene expression matrix can be reduced to improve the accuracy of performing downstream analysis on the basis of the gene expression matrix.

The invention discloses a spatial omics single cell data acquisition method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring a dyeing image of a biological sample, and acquiring a gene expression visual image of the biological sample; carrying out registration on the dyed image and the gene expression visualization image to obtain a registered image; performing cell segmentation on the registered image to obtain a cell mask pattern; determining cells of background gene molecules of unknown cells in the gene expression visualized image according to the cell mask graph, and marking the cells to obtain a space group molecular expression matrix with cell markers. Compared with the prior art, the method has the advantages that high-precision spatial single cell data is obtained through the processing flows of gene expression matrix graph, splicing, registration, segmentation and correction.

The invention is applicable to the technical field of cell type annotation, and provides an interpretable single cell annotation method, which comprises the following steps: based on a source cancer data set, performing cell type label prediction on a target cancer data set through a first module of an annotation model to obtain a prediction label of each cell in the target cancer data set, the annotation model is constructed by adopting an information bottleneck-based graph interpretation technology, and based on the health data set and the prediction label, interpretability analysis is performed on the to-be-interpreted cancer data set through a second module of the annotation model to obtain an interpretation result, the to-be-explained cancer data set is composed of the cancer cell gene expression matrix and the prediction label of the target cancer data set, so that the explanatory mechanism is integrated into the model architecture, the prediction accuracy is ensured, the explanatory stability and reliability are improved, the inconsistency caused by dependence on a post explanatory method is avoided, and the prediction accuracy is improved. And the prediction logic of the model can be understood.

The invention discloses a spatial transcriptome data spatial domain identification method based on multi-space self-supervised contrast learning, and the method comprises the steps: carrying out the modeling to generate a spatial neighborhood graph, keeping the graph structure unchanged, disorganizing node features, and carrying out the data enhancement, thereby obtaining an enhanced graph; constructing an encoder based on a graph neural network, extracting spatial transcriptome data fused with spatial information and gene information to obtain potential embedding, and sending the potential embedding into a multi-space generator to generate multiple groups of rich graph feature representations; fusing graph feature representation and potential embedding to obtain refined representation, reconstructing a gene expression matrix through a decoder, and adding contrast learning loss and reconstruction loss as a total objective function; and updating network parameters by adopting an Adam optimizer according to the obtained total objective function to complete spatial transcriptome spatial domain identification. Spatial transcriptome data are fully mined from global and local angles, and accurate spatial domain identification is realized.

The invention provides an expression data preprocessing method, device and system and a storage medium, and relates to the technical field of space-time omics. The expression data preprocessing method comprises the following steps: obtaining sequencing data based on a space-time group technology, and obtaining a gene expression matrix according to the sequencing data; carrying out dimension reduction processing to obtain a cell population; determining a differential gene list among the cell populations, and constructing a gene set expression matrix; and converting the gene set expression matrix into contribution degrees, and constructing a gene contribution degree matrix. According to the method, cell grouping and gene screening technologies are applied, a gene set expression matrix is accurately converted, matrix dimensionality reduction is remarkably achieved, similar feature gene sets are more aggregated, some potential valuable features are reserved as far as possible, the analysis accuracy is improved, the use efficiency of computing resources is optimized, and the method is suitable for large-scale popularization and application. And a more reliable and efficient research way is provided for scientific researchers.

The invention discloses a spatial transcriptome data analysis method based on artificial intelligence, and belongs to the technical field of spatial transcriptomics data analysis. Firstly, self-adaptive normalization and hypervariant gene screening preprocessing are carried out on original gene expression data; then constructing a hierarchical map integrating spatial proximity and transcription similarity, and ensuring the connectivity and robustness of the map through a dynamic radius pruning and neighborhood inheritance strategy; dividing positive and negative sample sets based on the atlas, and inputting a type modulation contrast graph auto-encoder for training; and finally, spatial domain identification and downstream function analysis are completed based on the low-dimensional potential representation or reconstructed gene expression matrix output by the model. The method effectively improves the accuracy and stability of spatial domain recognition, adapts to multi-technology-source data, enhances the biological interpretability of model output, and can be widely applied to biomedical scenes such as tumor microenvironment analysis and organ development research.

The invention discloses a method and a device for processing single cell gene expression profile data of different platforms, electronic equipment and a storage medium, which are used for converting the single cell gene expression profile data of different platforms into a gene expression matrix file of a single sample, a first column in the gene expression matrix file is a gene name, reading the gene expression matrix file of a single sample, and converting the gene expression matrix file of the single sample into a single sample data object; performing principal component analysis on all the single-sample data objects to obtain a preset number of principal components of a gene expression profile in each single-sample data object; correcting a preset number of principal components of the gene expression profile by taking samples of different platforms and single cell experimental methods as covariables to obtain a corrected first principal component and a corrected second principal component; and drawing a double plot of the corrected first principal component and the second principal component.

The invention belongs to the technical field of single cell data analysis, and particularly relates to a single cell pseudo-trajectory identification method based on a gene co-expression network. Comprising the following steps: S1, constructing a single cell transcription data matrix of behavior gene names and cell names; s2, preprocessing the single-cell transcription data matrix, constructing a cell expression matrix, and performing characteristic gene screening on the cell expression matrix to obtain a single-cell gene expression matrix; s3, constructing a gene co-expression network; s4, constructing a gene rank differential expression network in combination with the single-cell gene expression matrix and the gene co-expression network; and S5, carrying out integrated clustering on the gene rank differential expression network, fusing by using a similarity network to obtain a final clustering result, determining an initial cluster, distributing pseudo time for cells in each cluster, and deducing a cell pseudo trajectory. According to the method, the differentiation track of the cell can be identified, and compared with an original gene expression matrix, a more stable identification result can be obtained by utilizing a relative expression sequence.

The invention discloses a prediction model generation method and device of a gene regulatory network, a server and a medium, and relates to the technical field of computers. Gene activity changes at different time points in a gene expression time sequence are captured through a gene expression matrix encoder to obtain global gene expression characteristics; a text encoder is used for analyzing the corresponding relation between a gene expression mode and text semantics, multi-time-point gene expression feature data and biological text feature data are aligned through a comparative learning method, the learning rate is gradually and smoothly reduced through a cosine recession learning rate scheduling mechanism, it is ensured that the model is closer to a globally optimal solution, and the optimal solution is obtained. Relevance modeling between multi-time-point gene expression data and biological text description is achieved through the prediction model of the gene regulation and control network, and the precision and generalization ability of a statistical model are improved.

The invention provides a single-cell RNA sequencing data clustering method and system based on a multi-scale hypergraph. The method comprises the following steps: constructing a multi-scale similarity matrix through a gene expression matrix; generating a cell hypergraph through the multi-scale similarity matrix; carrying out cell hypergraph fusion; carrying out SVD (Singular Value Decomposition) on the hypergraph Laplacian matrix fused with the cell hypergraph; and extracting feature vectors corresponding to the first # imgabs0 # minimum feature values, and performing K-means to obtain a clustering result, thereby realizing the function of improving the clustering effect of the single-cell RNA sequencing data. The invention provides a novel hypergraph generation method, the high-order correlation among a plurality of cells is better captured, and the correlation among the cells is more accurately captured by using a plurality of scales. Compared with an existing single-cell RNA sequencing data clustering method, the single-cell RNA sequencing data clustering method has the advantages that a unified clustering framework is provided, the framework comprehensively learns topological information between cells on multiple scales, cell hypergraphs are constructed for clustering, and a good clustering result can be obtained.

The invention belongs to the field of spatial transcriptome spatial domain recognition, and provides an idling data feature extraction method, a spatial domain recognition method and a spatial domain recognition system in order to solve the problem of poor recognition accuracy of a spatial transcriptome spatial domain. The idling data feature extraction method comprises the following steps: acquiring idling data of a pathological section, wherein the idling data comprises an initial gene expression matrix and spatial position information; performing denoising processing on the initial gene expression matrix to obtain a denoised gene expression matrix; calculating a position code for each site according to the spatial position information, and splicing the position code with the denoised gene expression matrix to obtain an enhanced gene expression matrix; a mask auto-encoder is utilized to encode an enhanced gene expression matrix, low-order potential representation is obtained and serves as extracted idle data features, accurate recognition of a spatial domain can be achieved, and a good upstream analysis basis is provided for downstream tasks.

The invention provides a subcellular level space transcriptome function analysis method and system, and relates to the field of bioinformatics, the method comprises the following steps: automatically processing core information of a sample, and optimizing space transcriptome expression matrix data to realize expression matrix data reception and automatic processing; timed polling and the like are added to realize automation of standard analysis from a space gene expression matrix to a space transcriptome, and gene ID conversion, cell unit type selection and cell unit selection are performed according to expression matrix data to obtain data after selection; and according to the converted and selected data, carrying out statistics on spatial data features, and visualizing the data features. According to the method, through a multi-module automatic process, the analysis efficiency, accuracy and biological interpretation of space transcriptomics and single cell data are remarkably improved, and the method is suitable for multi-sample, high-resolution and big data analysis scenes.

The method disclosed by the invention comprises the following steps: S1, collecting a gene expression matrix of scRNA-seq, a gene activity matrix of scATAC-seq and cell type annotation of the scRNA-seq; s2, the scRNA-seq data and the scATAC-seq data are subjected to preprocessing, and a gene shared by the scRNA-seq data and the scATAC-seq data is obtained; s3, constructing an encoder network, and realizing joint representation of different omics data in the shared embedding space; s4, performing supervised learning guidance on the embedded space by adopting a cell type label in the scRNA-seq data; s5, in the embedding space and the label space, performing alignment on the scRNA-seq data and the scATAC-seq data by adopting an unbalanced optimal transmission algorithm; and S6, according to the matching probability matrix, endowing each cell in the scATAC-seq data with a cell type annotation, and realizing integration of the scRNA-seq data and the scATAC-seq data. The problem that the accuracy and reliability of data integration are affected due to distortion of biological related signals caused by application of an existing OT frame to single cell data integration is solved.

The invention relates to a single-cell gene expression quantitative trait site identification method and equipment based on high-throughput multi-omics sequencing. The method comprises the following steps: S1, constructing a data storage structure object; S2, standardizing a gene expression matrix; s3, filtering the data; s4, identifying single cell gene expression quantitative character sites; and S5, result visualization is carried out. The method provided by the invention realizes conjoint analysis of single cell transcriptomics data and genomics data, provides selection of a plurality of models for identifying gene expression quantitative trait sites for users, and can analyze data on a single cell level, thereby revealing association between genotypes and gene expression. And a new perspective is provided for understanding the molecular mechanism of complex diseases.

The invention discloses a cell interaction identification method based on cell space transcriptome data, which comprises the following steps: constructing cell pairs according to cell space proximity, and calculating an interaction signal matrix and space information of each pair of cells by combining a ligand-receptor database; then generating cell pair embedding representation by using a graph neural network, calculating the similarity between cell pairs, and calculating the similarity between cells based on a gene expression matrix; updating the similarity matrix through iteration until convergence to obtain a final similarity matrix between cells and between cell pairs; and clustering the similarity matrix, and constructing a cell type and cell pair type model, thereby realizing accurate cell interaction identification. According to the method, space and expression information can be fully utilized, inter-cell heterogeneity and context specificity interaction can be captured, the analysis precision and biological authenticity of a cell communication network are improved, and an effective tool is provided for research on tissue development, disease mechanisms, immune response and the like.

The invention relates to a new crown growing typing method and device, a molecular subtype and clinical variable association method and device and electronic equipment, and relates to the technical field of new crown growing typing. The typing method comprises the following steps: acquiring a gene expression matrix of a to-be-detected sample; and processing the gene expression matrix of a to-be-detected sample by adopting the new crown growing typing model to obtain a typing result of the to-be-detected sample. The association method comprises the following steps: acquiring a gene expression matrix and a typing result of an experimental sample and a gene expression matrix of a control sample; respectively carrying out differential gene analysis on the experimental samples of each molecular subtype and a control sample to obtain significant differential genes; performing functional pathway enrichment analysis, and calculating a gene set expression average value of the functional pathways with significant difference; and carrying out correlation analysis after normalizing the gene set expression average value and clinical variable data to obtain clinical variables associated with each molecular subtype.

The invention discloses a method and a device for identifying a transcription factor of a tissue gene, relates to the technical field of biological information, and mainly aims to solve the problem of poor identification accuracy of the transcription factor of the existing tissue gene. Comprising the following steps: acquiring an initial gene expression matrix of tissue genes, and constructing a full-connection directed graph based on the initial gene expression matrix; after the full-connection directed graph is converted into an undirected role graph, the undirected role graph and a feature matrix are reconstructed based on a graph neural network model after model training is completed, a reconstructed target gene expression matrix is obtained, and a loss value of the graph neural network model is obtained through calculation after edges are dynamically removed; and generating a directed gene regulatory network based on the target gene expression matrix, and identifying a transcription factor of the tissue gene based on the directed gene regulatory network.