Construction method, device and equipment of stroke space knowledge graph database and medium

Abstract

The invention discloses a construction method and device of a cerebral apoplexy space knowledge graph database, equipment and a medium, and relates to the technical field of big data analysis. According to the application, the Tonyuqian question 2.1 model is utilized to deeply mine, analyze and publically obtain papers, identify and mark spatial distribution conditions of ischemic cerebral apoplexy genes, physiological regions and pathological regions, and the spatial distribution conditions are displayed in the form of a cerebral apoplexy spatial knowledge graph database. According to the method, genes and spatial distribution information thereof are automatically extracted from literatures, and a dynamically updated knowledge graph is constructed, so that the problems of low data integration efficiency, insufficient labeling accuracy, insufficient dynamic association and the like in cerebral apoplexy spatial heterogeneity research in the prior art are solved.

Description

Technical Field

[0001] The present application relates to the field of big data analysis technology, and in particular to a method, device, equipment and medium for constructing a stroke spatial knowledge graph database. Background Art

[0002] With the acceleration of global population aging, stroke has become one of the major public health challenges facing the world. It is not only the main cause of disability, but also the leading cause of death in the world. Although a large number of studies have made significant progress in the understanding of stroke in the past few decades, there is still a lack of in-depth understanding of its complex disease mechanisms, especially the spatial heterogeneity shown in the development of stroke. This spatial heterogeneity is reflected in the pathological changes in different stages of stroke, different brain regions and their microenvironments, which is of great significance for the diagnosis, treatment and prognosis of stroke.

[0003] In existing stroke research, spatial heterogeneity analysis faces the following technical bottlenecks.

[0004] (1) Low data integration efficiency: Gene-region association information is scattered in massive literature. Traditional manual extraction methods are time-consuming, costly, and difficult to ensure data consistency.

[0005] (2) Insufficient labeling accuracy: The labeling of physiological areas (such as hippocampus and cortex) and pathological areas (such as the core area of ​​infarction) relies on expert experience and is easily affected by subjective factors.

[0006] (3) Lack of dynamic analysis capabilities: Existing databases lack automated interaction with bioinformatics tools (such as STRING and KEGG), and are unable to construct protein interaction networks and pathway enrichment analysis in real time, which limits the in-depth exploration of spatial heterogeneity mechanisms.

[0007] The existing technologies for studying spatial heterogeneity of stroke have low data integration efficiency, insufficient annotation accuracy, and insufficient dynamic association, which has become a technical problem that needs to be solved urgently. Summary of the invention

[0008] The purpose of this application is to provide a method, device, equipment and medium for constructing a stroke spatial knowledge graph database, so as to achieve efficient integration of multi-source heterogeneous data through automated tools, accurately annotate the spatial distribution of genes, and construct a dynamically updated knowledge graph, so as to solve the problems of low data integration efficiency, insufficient annotation accuracy and insufficient dynamic association in the existing technology of stroke spatial heterogeneity research.

[0009] To achieve the above objectives, this application provides the following solutions.

[0010] In a first aspect, the present application provides a method for constructing a stroke spatial knowledge graph database, comprising the following steps.

[0011] Get papers related to ischemic stroke.

[0012] The Tongyi Qianwen 2.1 model was used to identify the genes and gene description information related to ischemic stroke recorded in each paper, and output them in the form of structured data; the structured data included gene names, description information and sources.

[0013] The Tongyi Qianwen 2.1 model is used to analyze the papers that record various structured data, and the physiological region annotations and pathological region annotations are performed on each structured data. The physiological region annotations and pathological region annotations are added to the structured data respectively to construct a structured dataset.

[0014] The structured data set is cleaned, invalid and repeated structured data in the structured data set is deleted, a cleaned structured data set is obtained, and the cleaned structured data set is output in the form of an Excel table to obtain Excel table data.

[0015] According to the Excel table data, the protein-protein interaction network was constructed using the STRING database.

[0016] Based on the Excel table data, online bioinformatics tools were used to analyze the distribution differences of genes in different physiological and pathological regions, and the functional annotations and pathway enrichment of each gene in the Excel table data were determined.

[0017] A stroke spatial knowledge graph database was constructed based on the Excel table data, protein-protein interaction network, and functional annotations and pathway enrichment of each gene in the Excel table data.

[0018] In the second aspect, the present application provides a device for constructing a stroke spatial knowledge graph database, which applies the above-mentioned method for constructing a stroke spatial knowledge graph database, and includes the following modules.

[0019] The paper acquisition module is used to obtain papers related to ischemic stroke.

[0020] The gene identification module is used to use the Tongyi Qianwen 2.1 model to identify the genes and gene description information related to ischemic stroke recorded in each paper, and output them in the form of structured data; the structured data includes gene name, description information and source.

[0021] The region annotation module is used to analyze the papers that record various structured data using the Tongyi Qianwen 2.1 model, annotate the physiological regions and pathological regions of each structured data, and add the physiological region annotations and pathological region annotations to the structured data respectively to construct a structured data set.

[0022] The cleaning module is used to clean the structured data set, delete invalid and repeated structured data in the structured data set, obtain the cleaned structured data set, and output the cleaned structured data set in the form of an Excel table to obtain Excel table data.

[0023] The protein-protein interaction network construction module is used to construct a protein-protein interaction network based on Excel table data using the STRING database.

[0024] The distribution difference analysis module is used to analyze the distribution differences of genes in different physiological and pathological regions based on Excel table data using online bioinformatics tools, and to determine the functional annotations and pathway enrichment of each gene in the Excel table data.

[0025] The database construction module is used to construct a stroke spatial knowledge graph database based on Excel table data, protein-protein interaction network and functional annotations and pathway enrichment of each gene in Excel table data.

[0026] In a third aspect, the present application provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-mentioned method for constructing a stroke spatial knowledge graph database.

[0027] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned method for constructing a stroke spatial knowledge graph database.

[0028] According to the specific embodiments provided in this application, this application has the following technical effects.

[0029] The present application provides a method, device, equipment and medium for constructing a stroke spatial knowledge graph database. The present application uses the Tongyi Qianwen 2.1 model to deeply mine and analyze publicly available papers, identify and annotate the spatial distribution of genes, physiological regions, and pathological regions related to ischemic stroke (the spatial distribution refers to the specific location of the physiological regions (such as cortex, hippocampus) and pathological regions (such as infarct core area, ischemic penumbra) where genes act in stroke diseases, and through annotation and dynamic analysis, reveal the functional differences of genes in different spatial environments), and display them in the form of a stroke spatial knowledge graph database. The present application uses the natural language processing capabilities of the Tongyi Qianwen 2.1 model to automatically extract genes and their spatial distribution information from the literature, and construct a dynamically updated knowledge graph, which solves the problems of low data integration efficiency, insufficient annotation accuracy and insufficient dynamic association in the existing technology of spatial heterogeneity research on stroke.

[0030] This application uses the natural language processing capabilities of the Tongyi Qianwen 2.1 model to automatically extract genes and their spatial distribution information from the literature and generate a structured data set (to address data dispersion).

[0031] This application uses a pre-trained deep neural network model to identify physiological / pathological regions, combined with multiple rounds of data cleaning (such as keyword matching and standard gene set verification) to improve annotation accuracy (solve annotation complexity).

[0032] This application constructs a protein interaction network based on the STRING database, combines Metascape, KEGG and other tools to achieve pathway enrichment analysis, and forms a dynamically updateable knowledge graph (to solve the lack of dynamic analysis capabilities). BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0034] Figure 1 A flowchart of a method for constructing a stroke spatial knowledge graph database provided in one embodiment of the present application.

[0035] Figure 2 A flowchart of data collection and mining provided for an embodiment of the present application.

[0036] Figure 3 A schematic diagram of data verification results provided in one embodiment of the present application.

[0037] Figure 4A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0038] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0039] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0040] In an exemplary embodiment, Figure 1 As shown, a method for constructing a stroke spatial knowledge graph database is provided, including the following steps 101 to 107.

[0041] Step 101, obtaining papers related to ischemic stroke.

[0042] Step 102, using the Tongyi Qianwen 2.1 model to identify the genes and gene description information related to ischemic stroke recorded in each paper, and output them in the form of structured data; the structured data includes gene name, description information and source.

[0043] Step 103, using the Tongyi Qianwen 2.1 model to analyze the papers recording each structured data, annotate each structured data with physiological regions and pathological regions, and add the physiological region annotations and pathological region annotations to the structured data, respectively, to construct a structured data set.

[0044] Step 104 , clean the structured data set, delete invalid and repeated structured data in the structured data set, obtain the cleaned structured data set, and output the cleaned structured data set in the form of an Excel table to obtain Excel table data.

[0045] Step 105, constructing a protein-protein interaction network based on the Excel table data using the STRING database.

[0046] Step 106, based on the Excel table data, using online bioinformatics tools to analyze the distribution differences of genes in different physiological regions and pathological regions, and determining the functional annotations and pathway enrichment of each gene in the Excel table data.

[0047] Step 107, constructing a stroke spatial knowledge graph database based on the Excel table data, the protein-protein interaction network, and the functional annotations and pathway enrichment of each gene in the Excel table data.

[0048] By implementing the above steps 101 to 107, it is possible to systematically sort out and summarize the knowledge related to spatial regions in stroke diseases.

[0049] This application aims to construct a comprehensive, experimentally validated, and traceable knowledge graph of ischemic stroke disease.

[0050] In another exemplary embodiment of the present application, it is determined that the keywords related to stroke are "ischemiastroke", "cerebral ischemia", "cerebral infarct" and "MCAO", and based on the keywords "ischemiastroke", "cerebral ischemia", "cerebral infarct" and "MCAO", papers related to ischemic stroke published in the past 20 years from January 2003 to December 2023 are retrieved from the PubMed database, and all PDFs of open access (OpenAccess, OA) papers are downloaded and uniformly named as "Stroke+serial number" to facilitate subsequent large language model calls and in-depth text mining analysis.

[0051] Among them, PubMed database is the most widely used free MEDLINE search tool on the Internet. The MEDLINE search tool is a WEB-based biomedical information retrieval system developed by the National Center for Biotechnology Information of the U.S. National Library of Medicine in April 2000. The PubMed database contains more than 32 million biomedical documents and abstracts.

[0052] Exemplarily, the present application embodiment collected academic papers on ischemic stroke published from January 2003 to December 2023 from the PubMed database, totaling 139,267 papers. During the detailed review of the downloaded papers, a total of 92,610 papers were attached with digital object identifiers (DOIs). Papers with DOIs included OA papers and papers obtained by other means, of which 56,903 were open access papers, and the full text could be downloaded directly for further analysis. Using the knowledge extraction technology of the Tongyi Qianwen 2.1 model, gene information and response space information related to ischemic stroke were identified from these OA papers, including 1,125 human genes and 1,094 mouse genes. Among them, there are 55 genes related to the infarct core area, 89 genes related to the ischemic penumbra, 36 genes related to the cortex, and 9 genes related to the hippocampus, but other brain regions such as the striatum and thalamus have been studied independently less. Finally, important information such as their gene names, species, references, detailed descriptions, gene locations, and synonyms were accurately recorded, e.g. Figure 2 As shown, Figure 2 A in the figure is a schematic diagram of the information mining process of Tongyi Qianwen 2.1 model; Figure 2 B in the column is the total literature information, which is divided into the number of all retrieved documents (ALL column), the number of documents with DOI numbers (DOI column) and the number of OA papers (OA column); Figure 2 C in the figure represents the species information and the quantitative statistics of the spatial regions in StrokeDB.

[0053] Considering the complexity of text mining tasks and the huge amount of data, the present embodiment uses the Tongyi Qianwen 2.1 model as the main text mining tool, which can effectively identify and extract key information from complex medical literature. Therefore, this model is selected for text mining.

[0054] In another exemplary embodiment, the above step 102, based on the PDF document named "Stroke+serial number" obtained in step 101, uses the Tongyi Qianwen 2.1 model to perform an in-depth analysis of the downloaded literature to identify the genes related to ischemic stroke mentioned in the literature, specifically including the following steps 201-204.

[0055] Step 201: data preprocessing.

[0056] Convert the downloaded PDF documents named "Stroke+Serial Number" to text format so that they can be processed by the Tongyi Qianwen 2.1 model. During the conversion process, ensure the integrity and accuracy of the text, retain the chart titles, annotations and other contents that may contain gene information in the document, and minimize the problems of garbled text or information loss caused by format conversion.

[0057] Step 202: Model input and task setting.

[0058] Input the converted texts into the Tongyi Qianwen 2.1 model one by one. When inputting, explicitly inform the Tongyi Qianwen 2.1 model that the task is to identify gene information related to ischemic stroke, including gene names, gene variants (if mentioned in the text), and descriptions of the association between genes and stroke. For example, you can ask the Tongyi Qianwen 2.1 model to "extract all genes related to ischemic stroke and their description information from the paper."

[0059] Step 203: Model identification.

[0060] The Tongyi Qianwen 2.1 model performs gene recognition based on its deep neural network architecture and large-scale pre-trained knowledge system. During the pre-training process, the model learned a large amount of text patterns and knowledge in the fields of medicine and biology, and can recognize specific gene-related words and terms as well as their contextual relationships in the text.

[0061] When processing the input text, the Tongyi Qianwen 2.1 model uses natural language processing functions to perform natural language processing operations such as word segmentation and part-of-speech tagging on the text. Then, the deep neural network gene recognition function matches the words in the text with the gene-related patterns learned through pre-training through multi-layer calculations of the neural network. For example, the Tongyi Qianwen 2.1 model can recognize common gene name abbreviations (such as TNF-α for tumor necrosis factor-α), gene full names (such as brain-derived neurotrophic factor for brain-derived neurotrophic factor) and specific expressions when describing the relationship between gene function, mechanism of action and ischemic stroke (such as gene X is upregulated in the ischemic stroke model and is related to neuronal protection).

[0062] For the possible gene information identified, the Tongyi Qianwen 2.1 model will further analyze its context to confirm the relevance of the gene to ischemic stroke and extract relevant descriptive information. For example, if a text mentions "In brain tissue samples of patients with ischemic stroke, it was found that the mutation frequency of gene Y was high and was related to the abnormal activation of the apoptosis pathway", the Tongyi Qianwen 2.1 model can accurately extract the name of gene Y, its mutation status, and its association with the apoptosis pathway and ischemic stroke.

[0063] Step 204: Output and organize the results.

[0064] The Tongyi Qianwen 2.1 model outputs the identified gene information related to ischemic stroke in a structured form. The output may include a list of gene names, a description of each gene (such as its role in ischemic stroke, expression changes, etc.), and the source of the gene in the literature (such as the title and page number of the literature, etc., for subsequent verification).

[0065] The results output by the Tongyi Qianwen 2.1 model were sorted and preliminarily screened to remove possible misidentification or irrelevant information. For example, some gene information mentioned in other unrelated diseases or normal physiological processes but not directly related to ischemic stroke was excluded, and duplicated gene information was merged and deduplicated to ensure the accuracy and simplicity of the gene list, providing reliable basic data for subsequent analysis and research.

[0066] The specific method for organizing and preliminarily screening the results output by the Tongyi Qianwen 2.1 model includes the following steps A1-A3.

[0067] A1. Error identification filtering.

[0068] Preset stroke keywords (such as 'ischemia stroke', 'neuronal apoptosis', 'infarct core area') were matched with the descriptive information output by the model, and entries that did not contain keywords or were semantically irrelevant were deleted; Natural language processing techniques (such as semantic similarity calculation) are used to identify entries with ambiguous descriptions or that contradict the context (such as 'Gene X is highly expressed in normal brain tissue') and mark them as data for review.

[0069] A2. Merge duplicate data.

[0070] Gene names were compared with the NCBI standard gene set using a string matching algorithm based on edit distance (e.g., Levenshtein distance algorithm) to merge different expressions of the same gene (e.g., 'BDNF' and 'brain-derived neurotrophic factor').

[0071] For duplicate gene entries (same gene name, same description, same source), only the record of the first appearance was retained, and the DOI number of the other literature was appended to the 'Source' field.

[0072] A3. Manual review and verification.

[0073] Two researchers independently reviewed the screened data, focusing on checking the accuracy of gene names and descriptions, the completeness of the sources, and resolving inconsistent entries through cross-comparison.

[0074] The data that passes the review is included in the cleaned structured data set, and the data that fails is returned to the model for reprocessing or directly eliminated.

[0075] The Tongyi Qianwen 2.1 model normalizes the diverse descriptions of the same gene in different literature into standard terms through semantic analysis and term mapping, while retaining the differential information of the original descriptions to form a structured "main description-additional description-multiple sources" data framework.

[0076] In another exemplary embodiment, in step 103, based on the ischemic stroke disease-related genes obtained in the above step, the Tongyi Qianwen 2.1 model is used, and the model further analyzes the contextual information in the literature to identify and annotate whether each gene is experimentally studied in the relevant brain region, such as the hippocampus, cortex, striatum, etc. In addition to the annotation of physiological regions, it also focuses on identifying the pathological partitions of genes mentioned in the literature, such as the infarct core area and the ischemic penumbra. This step is crucial to understanding how genes work in different physiological / pathological brain region environments, and reveals the specific areas that may be affected by stroke diseases.

[0077] Among them, the basis for marking physiological areas and pathological areas includes: (1) Anatomical terms from medical knowledge bases (e.g., Allen Brain Atlas, AHA Stroke Classification).

[0078] (2) Description of the experiment in the context of the literature (e.g., “hippocampal slice analysis,” “gene expression in the infarct core”).

[0079] (3) Pathological feature library for model pre-training.

[0080] In another exemplary embodiment, the structured data includes gene name, description, species, physiological region, pathological region, literature source and functional annotation, wherein the physiological / pathological region field is directly associated with the spatial distribution information of brain regions (such as cortex, hippocampus, infarct core area).

[0081] This structured data not only records the association information between genes and brain regions, but also enables in-depth analysis through the following steps B1-B3.

[0082] B1. Construct a protein interaction network based on the STRING database to analyze the functional synergy of genes in specific brain regions.

[0083] B2. Use Metascape and KEGG tools to perform pathway enrichment and clarify the regulatory mechanism of genes in pathological areas.

[0084] B3. Dynamically associate multi-dimensional data (such as gene expression levels and experimental verification results) through knowledge graphs to support spatial heterogeneity research.

[0085] In another exemplary embodiment, the above step 104 cleans the structured data set obtained in the above step, matches it with the standard gene set of a biotechnology information center, removes irrelevant items, and deletes duplicate items to ensure the accuracy and consistency of the data. Subsequently, in order to ensure the convenience of subsequent analysis and display, the cleaned data is formatted and organized into an Excel table, which records in detail the key information such as gene name, species information, related brain region location, pathological region division, etc., to ensure that the data has a certain logical structure and readability, specifically including the following steps 301-305.

[0086] Step 301 , checking the integrity of each structured data in the structured data set, deleting incomplete structured data, and obtaining a structured data set after the first cleaning.

[0087] Step 302, matching the description information of each structured data in the structured data set after the first cleaning with the preset stroke keywords, deleting the structured data whose description information does not match the preset stroke keywords in the structured data set after the first cleaning, and obtaining the structured data set after the second cleaning.

[0088] Step 303 , uniformly standardize the structured data in the structured data set after the second cleaning to obtain a standardized structured data set.

[0089] Step 304 , compare each structured data in the structured data set after the standard processing, delete the repeated structured data in the structured data set after the standard processing, and obtain the structured data set after the third cleaning.

[0090] Step 305, matching the gene name of each structured data in the structured data set after the third cleaning with the standard gene set, deleting the structured data whose gene name in the structured data set after the third cleaning does not match the standard gene set, and obtaining a cleaned structured data set.

[0091] The above cleaning process specifically includes the following steps 401 to 403.

[0092] Step 401, data screening.

[0093] First, the raw data extracted from a large number of literatures were preliminarily screened. The data integrity was checked and obviously incomplete or damaged data records were removed. For example, if some data items were missing key information, such as unclear gene names or unlabeled brain region locations, they were excluded.

[0094] At the same time, stroke-related data are filtered out according to specific rules. For example, through keyword matching, only data records containing specific stroke-related keywords (such as "stroke", "cerebrovascular accident", "cerebral ischemia", etc.) are retained.

[0095] Step 402, format standardization.

[0096] The data format is standardized. The data formats in different documents may be different, which will bring difficulties to subsequent analysis. Therefore, the format of key information such as gene name, species information, brain region location and pathological region division is standardized.

[0097] For example, for gene names, the internationally accepted nomenclature is uniformly adopted; for species information, the standard taxonomic naming method is adopted; for brain region locations and pathological area divisions, unified anatomical terms are used for labeling.

[0098] Step 403, removing noise data.

[0099] Use natural language processing technology and machine learning algorithms to identify and remove noisy data. Noisy data may include incorrect annotations, unclear descriptions, or information irrelevant to the topic.

[0100] The specific method for identifying and removing noise data includes the following steps C1-C3.

[0101] C1. Natural language processing technology.

[0102] Semantic similarity analysis: Use pre-trained medical word vector models (such as BioBERT) to calculate the semantic similarity between text descriptions and preset stroke keywords (such as "ischemic penumbra" and "apoptosis regulation"), and filter out entries with similarity below the threshold.

[0103] Named Entity Recognition (NER): Key entities such as gene names and brain region names are identified through NER models (such as those based on the BiLSTM-CRF architecture). Entries missing key entities are marked as incomplete data and removed.

[0104] C2. Machine learning algorithm.

[0105] Supervised learning classification: A random forest model was used to train a classifier based on a labeled dataset (containing 'relevant' and 'noise' labels) to automatically identify descriptions not related to stroke (such as 'normal physiological process', 'other disease associations').

[0106] Anomaly Detection: Use the Isolation Forest algorithm to detect unusual patterns in description information (such as contradictory statements, non-standard terminology).

[0107] C3. Manual review.

[0108] The screened data were reviewed by researchers, with a focus on checking semantically ambiguous items and algorithm classification results to ensure the accuracy and consistency of the final data set.

[0109] For example, text analysis algorithms can be used to detect data records with unclear descriptions or ambiguous semantics, and manual review can be performed to determine whether they need to be removed. At the same time, machine learning algorithms can be used to classify data and filter out and remove information that is not related to stroke.

[0110] During the data cleaning process, if there are multiple literature sources for the same gene, the following steps D1-D3 are performed.

[0111] D1. Extract common descriptions as the main description and mark the difference descriptions as "additional information".

[0112] D2. Record the title, DOI and page number of all source documents.

[0113] D3. Prioritize sources based on the authority of the document to ensure that high-credible sources are displayed first.

[0114] The above text analysis algorithms include: Named Entity Recognition (NER): A model based on the BiLSTM-CRF architecture was used to identify gene names (such as 'TNF-α'), brain region names (such as 'hippocampus', 'cortex'), and pathological regions (such as 'infarct core area') in the text. Entries with missing key entities were considered incomplete data and removed; Semantic similarity calculation: Use pre-trained medical domain word vector models (such as BioBERT) to calculate the semantic similarity between text descriptions and preset stroke keywords (such as "apoptosis regulation" and "ischemic penumbra"), and filter out entries with similarity below a threshold (such as 0.7); Rule matching: Non-standard terms (such as 'cerebral infarction', 'stroke') are matched by regular expressions and standardized into unified terms (such as 'ischemic stroke').

[0115] The machine learning algorithms mentioned above include: Supervised learning classification: A random forest model was used to train a classifier based on a labeled dataset (containing 'relevant' and 'noise' labels), with features including word frequency (TF-IDF), entity type, and semantic similarity score, to automatically identify descriptions not related to stroke (such as 'normal metabolism', 'cancer association'); Anomaly detection: An Isolation Forest algorithm was used to detect unusual patterns in the description information, such as contradictory statements (e.g., 'Gene X is highly expressed and not expressed in the infarct core') or non-standard terms (e.g., 'Brain region A' does not match the anatomical term).

[0116] The above-mentioned manual review specifically includes: the screened data is independently reviewed by two researchers, with a focus on verifying semantically ambiguous items and algorithm classification results, and resolving inconsistencies through cross-comparison to ensure the final data quality.

[0117] The above matching process specifically includes the following steps 501 to 503.

[0118] Step 501: data preparation.

[0119] First, obtain the standard gene set from the National Center for Biotechnology Information (NCBI), which has been rigorously screened and validated and contains detailed information on various known genes.

[0120] At the same time, the cleaned raw data is organized and prepared to ensure that the format and content of the data meet the matching requirements. For example, the gene names are standardized so that they can be accurately matched with the gene names in the standard gene set.

[0121] Step 502, matching calculation.

[0122] Advanced string matching algorithms and bioinformatics tools were used to match genes in the cleaned raw data with the NCBI standard gene set.

[0123] For example, an algorithm based on edit distance can be used to calculate the similarity between the gene names in the original data and the gene names in the standard gene set. If the similarity exceeds a certain threshold, the two are considered to be matched successfully.

[0124] At the same time, matching is performed in combination with the functional information and biological characteristics of genes. For example, for some genes with unknown functions, they can be matched with genes in the standard gene set by comparing their conservation, expression patterns and other characteristics in different species.

[0125] Step 503: manual review.

[0126] In order to ensure the accuracy of the data extracted by the Tongyi Qianwen 2.1 model, a manual proofreading method was adopted, that is, two researchers independently checked the same batch of data and then compared their proofreading results to identify and resolve any inconsistencies. Figure 3As shown in the figure, the verification results of a randomly displayed document show that the gene information, species information, physiological brain region and pathological partition data extracted by the Tongyi Qianwen 2.1 model are highly consistent with the original document records. By checking hundreds of records extracted by the large language model in detail, it was found that the model showed a high degree of accuracy in gene name recognition and reference matching. These results verify the effectiveness and reliability of the Tongyi Qianwen 2.1 large language model in complex document information extraction tasks, and significantly improve the quality and credibility of the final database. Figure 3 A in the figure is a screenshot of the txt format document output of the results of the Tongyi Qianwen 2.1 model; Figure 3 B in the figure is the manual verification result of the relevant content of the specific document.

[0127] That is, after the matching is completed, manual review is performed to ensure the accuracy of the match. Manual review can find some problems that the automatic matching algorithm cannot identify, such as ambiguous gene names and confusion between genes with similar but different functions.

[0128] For problems found during manual review, further analysis and processing will be carried out to ensure the accuracy and consistency of the data.

[0129] In order to reveal potential protein-protein interactions in the study, the STRING database was used to construct the corresponding protein-protein interaction network. The STRING database provides a platform that can integrate information from multiple sources to identify and analyze protein-protein interactions. This interaction information includes but is not limited to text mining, experimental verification, existing database records, protein co-expression patterns, protein neighborhood relationships, gene fusion events, and gene co-occurrence. Among them, the parameter sets the minimum interaction score to 0.4 to ensure that the analyzed protein-protein interactions have a certain degree of credibility.

[0130] In another exemplary embodiment, the above step 105 uses the STRING database to construct a corresponding protein-protein interaction network based on the cleaned structured data set, including the following steps 601 to 605.

[0131] Step 601, data preparation.

[0132] Organize the gene list to ensure that the gene names are accurate and standardized and consistent with the recognition format of the STRING database.

[0133] Step 602, access the database and input the gene list.

[0134] Open the STRING database website and upload or enter the gene list in the designated area.

[0135] Step 603, setting parameters.

[0136] The minimum interaction score was set to 0.4 to ensure the inclusion of credible interactions.

[0137] Step 604, building a network.

[0138] Click the button to start the construction. The STRING database integrates multiple information sources and algorithm analysis and calculation, including text mining, experimental verification, database record integration, protein co-expression pattern analysis, neighborhood relationship inference, gene fusion event analysis and gene co-occurrence statistics, etc., and selects high-confidence interaction relationship generation networks based on set parameters.

[0139] Step 605: Obtain and analyze results.

[0140] View the network graphical display and obtain interaction details, statistical characteristics, functional annotations and other data, which can be downloaded for further analysis, such as revealing disease-related biological pathways and regulatory mechanisms through key node and network module analysis.

[0141] On the basis of constructing the knowledge graph, the gene lists extracted from the literature were classified according to their specific regions in the brain, and pathway enrichment analysis was further performed on the retrieved genes in different regions.

[0142] In another exemplary embodiment, in step 106, based on the cleaned structured data set, online bioinformatics tools such as Metascape, DAVID, and KEGG are used to perform functional annotation and pathway enrichment on genes, focusing on analyzing the distribution differences of genes in different physiological and pathological regions, and interpreting the importance of spatial heterogeneity to stroke diseases based on the pathway enrichment results. In the embodiment of the present application, because it is necessary to pay attention to spatial heterogeneity, different genes are expressed differently in space and play different functions. Therefore, it is necessary to see whether the genes selected from the pdf are reflected in which position in the text, and then collect the gene sets at different positions to see what their functions are, including the following steps 701-step 704.

[0143] Step 701, data preparation.

[0144] Prepare a cleaned structured dataset (in Excel spreadsheet format).

[0145] Step 702: Use Metascape to analyze the distribution differences of genes in different physiological and pathological regions.

[0146] Upload data: Open Metascape, drag and drop the data file or upload it as required, and select the appropriate format and gene identifier type.

[0147] Set up and start: Set the parameters as required and click the Analyze button.

[0148] Obtaining results: Metascape integrates multiple databases and provides functional annotations (biological processes, cellular components, molecular functions, etc.) and pathway enrichment results (presented in charts, tables, and visual networks) through statistical analysis, based on which the distribution differences of genes in different physiological and pathological regions are analyzed.

[0149] Step 703: Use DAVID to analyze the distribution differences of genes in different physiological regions and pathological regions.

[0150] Upload and convert: Access DAVID, use the "Gene ID Conversion" tool to convert the format if necessary, upload the gene list, and select the species.

[0151] Start analysis: Click tools such as "Functional Annotation", set parameters and start analysis.

[0152] Interpretation of results: DAVID integrates multiple databases and provides functional annotations, pathway enrichment tables and visual pathway maps based on statistical tests, thereby focusing on the differences in gene distribution in stroke-related areas.

[0153] Step 704: Use KEGG to analyze the distribution differences of genes in different physiological regions and pathological regions.

[0154] Prepare and enter: Prepare the data and enter the KEGG official website analysis portal.

[0155] Submit analysis: Enter or upload the gene list as required and select the analysis option.

[0156] View results: KEGG compares genes and pathways, and presents pathway enrichment results in a table after statistical calculation, thereby exploring the impact of gene distribution differences in different regions on stroke.

[0157] The process of pathway enrichment analysis in different regions is as follows.

[0158] 1. Pathway enrichment analysis in different injury areas.

[0159] When ischemic stroke occurs, the blood flow to the brain suddenly decreases, which leads to hypoxia and insufficient energy supply to the brain tissue. In this case, cells in the infarct core and the ischemic penumbra show different biological responses and survival strategies. The ischemic penumbra is a potential recovery area of ​​brain tissue after ischemic stroke. As shown in Table 1, the apoptosis-related pathways (Regulation of neuron apoptotic process The high enrichment and activation of negative regulation of apoptoticsignaling pathway indicate that the cell death process in the ischemic penumbra area can be controlled, reflecting the mechanism by which cells strive to maintain survival. At the same time, these pathways slow down programmed cell death, providing a time window for clinical intervention; at the same time, some studies have reported that when leukocytes are recruited to the inflammatory site, they may also cause damage while producing the effect of protecting tissues. However, the positive regulation pathway enriched in interleukin-1 production in Table 1 reflects the protective effect of tissues, that is, the removal of dead cells and tissues through immune regulation can promote the recovery and repair of ischemic penumbra cells. In addition, the ischemic penumbra is enriched in pathways related to lipid metabolism and atherosclerosis. Some studies have reported that the metabolic transformation of multiple lipids is closely related to the expansion of cerebral infarction area and neurological damage, suggesting that abnormal lipid metabolism in ischemic stroke may play a role in cell survival, inflammatory response and subsequent repair.

[0160] Programmed cell death, as well as inflammation and autoimmune responses, play a key role in the cell death mechanism. Through the analysis of the relevant pathways enriched in the infarct core area, the main enrichment was the positive regulation of apoptotic process-related pathways, which is exactly the opposite of the apoptotic process in the ischemic penumbra. The activation of positive regulation of apoptotic process indicates that cell death, especially programmed cell death, will occur in the infarct core area after stroke, indicating that there is cell damage and death in the infarct core area due to hypoxia and insufficient energy supply. In addition, the enrichment and activation of necrotic death (Necroptosis) and spinal cord injury (Spinal cord injury) pathways indicate that there is not only programmed cell death in the infarct core area, but also cell death involving inflammation and autoimmune response mechanisms.

[0161] Table 1 Summary of gene pathway enrichment analysis results for each injury area

[0162] The analysis of spatial functional differences between the infarct core and ischemic penumbra is crucial for understanding the spatial heterogeneity and pathophysiological process of stroke, and provides important information for the design and selection of treatment strategies. For example, the protection and recovery potential of the ischemic penumbra suggests that early intervention (such as restoration of blood flow, anti-inflammatory therapy) may be beneficial in minimizing brain tissue damage and promoting functional recovery. The study of the infarct core helps to understand the irreversible process of brain tissue damage and provide strategies for reducing inflammation and preventing the spread of damage.

[0163] 2. Pathway enrichment analysis in different brain regions.

[0164] As shown in Table 2, the enrichment pathway analysis of genes in the cortical region found that the negative regulation of cell population proliferation and the regulation of smooth muscle cell proliferation pathways were highly enriched, indicating that the cortical region attempts to limit excessive cell proliferation and maintain tissue structure stability after ischemia to reduce pathological changes, such as scar formation, thereby affecting the recovery process. The activation and enrichment of biological pathways for the regulation of neuron apoptotic process and the negative regulation of apoptotic signaling pathway indicate that the cortical region may activate protective mechanisms after stroke to reduce the death of neurons through apoptotic pathways. A study showed that after ischemic stroke, apoptosis of neurons in the cortical region increased significantly, which also confirmed that the regulation of these pathways may be crucial for reducing damage and promoting recovery.

[0165] In addition, the positive regulation of angiogenesis pathway is highly enriched in the gene enrichment pathways in the hippocampus, indicating that the hippocampus promotes the formation of new blood vessels after ischemia, thereby improving local blood flow changes, which is beneficial to the recovery and functional reconstruction of the damaged area, which is different from the regulation of blood vessels in the cortical area. The enrichment of the positive regulation of cytokine-mediated signaling pathway biological pathway reflects that there may be more active inflammatory responses and immune regulation processes in the hippocampus, which may be a double-edged sword for the repair process after injury, which is beneficial for the removal of dead cells and tissue repair, but may also aggravate damage. At the same time, the enrichment and regulation of the cellular homeostasis pathway indicate that the hippocampus may play an important role in maintaining the stability of the cellular intracellular environment, which is particularly important for resisting the damage caused by ischemia.

[0166] Table 2 Summary of gene pathway enrichment analysis results for each physiological brain region

[0167] In another exemplary embodiment, after the above step 107, the stroke spatial knowledge graph database is also displayed. In order to effectively display the constructed knowledge graph, a special website is developed in the present application embodiment. The front end of the server is developed using HBuilder X 3.7.11, and the back end is implemented using AppServ 9.3.0, Apache 2.4.4, PHP 7.3.1, My SQL 8.0.1 and phpMyAdmin 4.9.1. The original data matrix, pre-processed data and intermediate results are stored in Notepad 8.4.8, and the pre-processing is performed on Python 3.7, Anaconda Navigator 2.3.2, Jupyter Notebook 6.4.12, R 4.2.2 and R Studio. The interactive visualization diagram is implemented using Apache ECharts 5.4. In the website display design, users can query specific genes through the search function, and the website will display detailed information related to the gene, including its distribution in the brain, the biological processes involved and the pathological state. In addition, the website provides a graphical knowledge graph view, allowing users to intuitively understand the spatial heterogeneity of stroke and the interactions between genes.

[0168] The design of the homepage of the StrokeDB website focuses on providing a user-friendly interactive interface. The main function bars include options such as search, browse, network, pathway, and statistics. In addition, the homepage shows the spatial heterogeneity of stroke to the greatest extent in visual display, including rich brain region information and damaged area information. Through this design, users can quickly obtain the basic information they need; through the integrated spatial expression data, users can also view the distribution of genes in brain regions through high-resolution images, and have an intuitive understanding of the complexity of stroke. The retrieval function of the database provides users with flexible and powerful data query capabilities. Users can use keywords to retrieve specific gene information, gene descriptions, and expression locations in stroke space, or they can search for specific brain regions to return detailed information on brain regions and related gene expression data. In addition, the website also includes a browsing function for gene lists and brain region lists, allowing users to easily compare the functions and expression patterns of different genes and gain a deep understanding of the transcriptional expression profiles and biological pathways of specific brain regions. For stroke-related protein-protein interaction information, the database provides a detailed protein interaction network (PPI) data display. This section not only includes basic information about PPIs, such as interacting protein pairs, interaction types, and functional annotations, but also provides relevant experimental support evidence and literature citations. With this detailed PPI information, researchers can explore the molecular mechanisms of stroke in more depth, providing a scientific basis for further experimental design and disease treatment. Finally, the data statistics page of StrokeDB summarizes and displays comprehensive information about the number of genes, detailed information on brain tissue samples, marker genes, protein interactions, and related pathway enrichment in the database, highlighting the wide application potential and data support capabilities of StrokeDB in the field of stroke research.

[0169] Based on the above-mentioned various method embodiments, this application uses the Tongyi Qianwen 2.1 model to conduct in-depth text mining on a large number of documents, and conducts pathway enrichment analysis on genes related to spatial regions, successfully constructs the stroke spatial knowledge map database StrokeDB, and presents it in the form of a website. By comparing the differential gene and pathway enrichment analysis between the ischemic penumbra and the infarct core area, it is found that the infarct core area and the ischemic penumbra exhibit different biological characteristics in terms of cell death, inflammatory response, cell protection and repair. In comparing the differences between different physiological brain regions in stroke diseases, it is found that the cortical region limits the scope of damage and promotes tissue stability by regulating cell death and proliferation processes, while the hippocampal region promotes damage repair and functional recovery by promoting angiogenesis and regulating inflammatory responses. In short, the method of the embodiment of the present application emphasizes the importance of spatial heterogeneity in the development of stroke disease, challenges the general conclusions drawn from the traditional Bulk research method, and also lays the foundation for the comparative study of the anti-acute stroke effects of Panax notoginseng and ginseng.

[0170] Based on the same inventive concept, the embodiment of the present application also provides a device for constructing a stroke spatial knowledge graph database for implementing the method for constructing a stroke spatial knowledge graph database involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in the embodiments of the device for constructing one or more stroke spatial knowledge graph databases provided below can be referred to the limitations of the method for constructing a stroke spatial knowledge graph database above, and will not be repeated here.

[0171] In an exemplary embodiment, a device for constructing a stroke spatial knowledge graph database is provided, comprising the following modules.

[0172] The paper acquisition module is used to obtain papers related to ischemic stroke.

[0173] The gene identification module is used to use the Tongyi Qianwen 2.1 model to identify the genes and gene description information related to ischemic stroke recorded in each paper, and output them in the form of structured data; the structured data includes gene name, description information and source.

[0174] The region annotation module is used to analyze the papers that record various structured data using the Tongyi Qianwen 2.1 model, annotate the physiological regions and pathological regions of each structured data, and add the physiological region annotations and pathological region annotations to the structured data respectively to construct a structured data set.

[0175] The cleaning module is used to clean the structured data set, delete invalid and repeated structured data in the structured data set, obtain the cleaned structured data set, and output the cleaned structured data set in the form of an Excel table to obtain Excel table data.

[0176] The protein-protein interaction network construction module is used to construct a protein-protein interaction network based on Excel table data using the STRING database.

[0177] The distribution difference analysis module is used to analyze the distribution differences of genes in different physiological and pathological regions based on Excel table data using online bioinformatics tools, and to determine the functional annotations and pathway enrichment of each gene in the Excel table data.

[0178] The database construction module is used to construct a stroke spatial knowledge graph database based on Excel table data, protein-protein interaction network and functional annotations and pathway enrichment of each gene in Excel table data.

[0179] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 4 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for constructing a stroke spatial knowledge graph database is implemented.

[0180] Those skilled in the art will understand that Figure 4The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components. In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0181] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0182] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0183] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0184] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0185] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0186] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for constructing a stroke spatial knowledge graph database, characterized in that: include: Obtain papers related to ischemic stroke; The Tongyi Qianwen 2.1 model was used to identify the genes and gene description information related to ischemic stroke recorded in each paper, and output them in the form of structured data; the structured data included gene names, description information and sources; The Tongyi Qianwen 2.1 model was used to analyze the papers that recorded various structured data, and physiological region annotations and pathological region annotations were performed on each structured data. The physiological region annotations and pathological region annotations were added to the structured data to construct a structured data set. Cleaning the structured data set, deleting invalid and repeated structured data in the structured data set, obtaining a cleaned structured data set, and outputting the cleaned structured data set in the form of an Excel table to obtain Excel table data; According to the Excel table data, the protein-protein interaction network was constructed using the STRING database; Based on the Excel table data, online bioinformatics tools were used to analyze the distribution differences of genes in different physiological and pathological regions, and the functional annotations and pathway enrichment of each gene in the Excel table data were determined; A stroke spatial knowledge graph database was constructed based on the Excel table data, protein-protein interaction network, and functional annotations and pathway enrichment of each gene in the Excel table data.

2. The method for constructing a stroke spatial knowledge graph database according to claim 1, characterized in that: Obtain papers related to ischemic stroke, including: Obtain papers related to ischemic stroke published within a preset time period from the PubMed database; Set the file name of each paper to the format of "Stroke+serial number".

3. The method for constructing a stroke spatial knowledge graph database according to claim 1, characterized in that: The Tongyi Qianwen 2.1 model was used to identify the genes and gene descriptions related to ischemic stroke recorded in each paper, including: Input the command "Please extract all genes related to ischemic stroke and their description information from the paper" to the Tongyi Qianwen 2.1 model; Each paper was input into the Tongyi Qianwen 2.1 model one by one to obtain the genes and gene description information related to ischemic stroke recorded in each paper output by the Tongyi Qianwen 2.1 model.

4. The method for constructing a stroke spatial knowledge graph database according to claim 3, characterized in that: Input each paper into the Tongyi Qianwen 2.1 model one by one to obtain the genes and gene description information related to ischemic stroke recorded in each paper output by the Tongyi Qianwen 2.1 model, including: Use the natural language processing function of the Tongyi Qianwen 2.1 model to segment the text in the paper; The deep neural network gene recognition function of the Tongyi Qianwen 2.1 model is used to recognize each word obtained by word segmentation, and obtain words that can represent gene names related to ischemic stroke and their description information; the description information is obtained by combining descriptive words of gene names related to ischemic stroke; the deep neural network recognition function is implemented based on a pre-trained deep neural network sub-model, and the pre-trained deep neural network sub-model is obtained by training the deep neural network model using knowledge in the fields of medicine and biology.

5. The method for constructing a stroke spatial knowledge graph database according to claim 1, characterized in that: The Tongyi Qianwen 2.1 model is used to analyze the papers that record various structured data, and to annotate the physiological and pathological regions of each structured data, including: The natural language processing function of the Tongyi Qianwen 2.1 model is used to analyze the papers containing structured data i to obtain the words about the region to which the genes in the structured data i belong; i=1, 2, ..., I, where I is the number of structured data; The deep neural network region recognition function of the Tongyi Qianwen 2.1 model is used to identify words about the region to which the genes in the structured data i belong, determine the physiological region and pathological region to which the genes in the structured data i belong, and annotate the physiological region and pathological region of the structured data i.

6. The method for constructing a stroke spatial knowledge graph database according to claim 1, characterized in that: Clean the structured data set, delete invalid and duplicate structured data in the structured data set, and obtain the cleaned structured data set, including: Check the integrity of each structured data in the structured data set, delete incomplete structured data, and obtain the structured data set after the first cleaning; Matching the description information of each structured data in the structured data set after the first cleaning with the preset stroke keywords, deleting the structured data whose description information does not match the preset stroke keywords in the structured data set after the first cleaning, and obtaining the structured data set after the second cleaning; Performing unified and standardized processing on each structured data in the structured data set after the second cleaning to obtain a structured data set after standardized processing; Compare each structured data in the structured data set after the standardization processing, delete the repeated structured data in the structured data set after the standardization processing, and obtain the structured data set after the third cleaning; The gene name of each structured data in the structured data set after the third cleaning is matched with the standard gene set, and the structured data whose gene name in the structured data set after the third cleaning does not match the standard gene set is deleted to obtain the cleaned structured data set.

7. The method for constructing a stroke spatial knowledge graph database according to claim 1, characterized in that: The online bioinformatics tools specifically include: one or more of Metascape, DAVID and KEGG.

8. A device for constructing a stroke spatial knowledge graph database, characterized in that: The device for constructing a stroke spatial knowledge graph database applies the method for constructing a stroke spatial knowledge graph database according to any one of claims 1 to 7, and the device for constructing a stroke spatial knowledge graph database comprises: The paper acquisition module is used to obtain papers related to ischemic stroke; A gene identification module is used to use the Tongyi Qianwen 2.1 model to identify the genes and gene description information related to ischemic stroke recorded in each paper, and output them in the form of structured data; the structured data includes gene name, description information and source; The region annotation module is used to analyze the papers recording various structured data using the Tongyi Qianwen 2.1 model, annotate the physiological regions and pathological regions of each structured data, and add the physiological region annotations and pathological region annotations to the structured data to construct a structured data set; A cleaning module is used to clean the structured data set, delete invalid and repeated structured data in the structured data set, obtain the cleaned structured data set, and output the cleaned structured data set in the form of an Excel table to obtain Excel table data; The protein-protein interaction network construction module is used to construct a protein-protein interaction network based on Excel table data using the STRING database; The distribution difference analysis module is used to analyze the distribution differences of genes in different physiological and pathological regions based on the Excel table data using online bioinformatics tools, and determine the functional annotations and pathway enrichment of each gene in the Excel table data; The database construction module is used to construct a stroke spatial knowledge graph database based on Excel table data, protein-protein interaction network and functional annotations and pathway enrichment of each gene in Excel table data.

9. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for constructing a stroke spatial knowledge graph database as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, it implements the method for constructing a stroke spatial knowledge graph database as described in any one of claims 1 to 7.

Images