Molecular typing method, device, electronic equipment and storage medium

Abstract

This application provides a molecular typing method, apparatus, electronic device, and storage medium, including acquiring cell type abundance data corresponding to each analysis unit in the spatial transcriptome data of the tissue sample to be tested; spatially clustering the analysis units based on the cell type abundance data to obtain multiple clusters; determining the target cluster based on the highest proportion of cell types corresponding to each cluster, wherein the target cluster represents the tumor region of the tissue sample to be tested; determining the first expression value of N specified genes in the target cluster, and constructing a first gene expression matrix based on the first expression value; and performing correlation determination based on the first gene expression matrix and a preset typing centroid matrix to obtain the typing result corresponding to the target cluster. This application can achieve more refined molecular typing, determine the tumor region in the tissue sample to be tested, and accurately provide the typing result of the tumor region, which is of great significance for improving the accuracy and effectiveness of tumor treatment.

Description

Technical Field

[0001] This application relates to the field of biological detection technology, and in particular to a molecular typing method, apparatus, electronic device and storage medium. Background Technology

[0002] Tumor molecular typing is a method of classifying tumors based on their molecular characteristics, providing a basis for precise cancer diagnosis, prognostic stratification, and tumor staging. However, current techniques for tumor molecular typing are based on whole-piece tissue samples and cannot accurately distinguish or explore molecular typing differences between different regions of a pathological section. Summary of the Invention

[0003] The embodiments of this application aim to provide a molecular typing method, apparatus, electronic device, and storage medium, which are intended to perform more refined molecular typing on tissue samples and accurately provide spatial typing results of tumor regions.

[0004] In a first aspect, embodiments of this application provide a molecular typing method, the method comprising:

[0005] Obtain cell type abundance data for each analysis unit in the spatial transcriptome data corresponding to the tissue sample to be tested;

[0006] Based on the cell type abundance data corresponding to each of the analysis units, spatial clustering is performed on each of the analysis units to obtain multiple clusters;

[0007] At least one target cluster is determined based on the highest percentage of cell types corresponding to each of the aforementioned clusters, wherein the target cluster is used to represent the tumor region of the tissue sample to be tested;

[0008] Determine the first expression value of N specified genes in each target cluster, and construct a first gene expression matrix based on the first expression value, where N is an integer greater than 1;

[0009] Correlation determination is performed based on the first gene expression matrix and the preset genotyping centroid matrix to obtain the genotyping result corresponding to each target cluster, wherein the genotyping centroid matrix includes the centroid values ​​of N specified genes under each preset genotyping.

[0010] In this embodiment, based on the cell type abundance data corresponding to each analysis unit in the spatial transcriptome data of the tissue sample to be tested, spatial clustering is performed on each analysis unit to obtain multiple clusters. Then, based on the highest proportion of cell types corresponding to each cluster, target clusters representing tumor regions are screened from all clusters, thereby distinguishing tumor regions and non-tumor regions in the tissue sample to be tested. For the target clusters representing tumor regions, a first gene expression matrix is ​​constructed based on the first expression values ​​of N specified genes in each target cluster, and a genotyping centroid matrix is ​​obtained. The genotyping centroid matrix includes the centroid values ​​of N specified genes under each preset genotype. The correlation between the first gene expression matrix and the genotyping centroid matrix is ​​determined to obtain the genotyping result corresponding to each target cluster. Compared with related technologies that perform molecular genotyping based on the entire tissue sample, this embodiment achieves more refined molecular genotyping, which can realize the division of spatial tumor regions in tissues and accurately provide molecular genotyping results for tumor regions, which is of great significance for improving the accuracy and effectiveness of tumor treatment.

[0011] In one possible implementation, determining at least one target cluster based on the highest percentage of cell types corresponding to each of the said clusters includes:

[0012] For each cluster, the proportion of each cell type in the cluster is determined based on the cell type abundance data corresponding to the analysis unit in the cluster, and the highest proportion cell type corresponding to the cluster is determined based on the proportion of each cell type in the cluster.

[0013] The cluster whose highest percentage of cell types is the preset target cell type is selected as the target cluster.

[0014] In one possible implementation, the step of determining the correlation based on the first gene expression matrix and a preset genotyping centroid matrix to obtain the genotyping result corresponding to each target cluster includes:

[0015] The first gene expression matrix is ​​correlated with the preset subtyping centroid matrix to obtain a correlation matrix, wherein the correlation matrix is ​​used to describe the similarity between each target cluster with respect to different preset subtypings;

[0016] Based on the correlation matrix, the typing result corresponding to the target cluster is determined.

[0017] In one possible implementation, the first gene expression matrix includes the first expression values ​​of N specified genes in each target cluster; the first gene expression matrix is ​​correlated with the preset genotyping centroid matrix to obtain a correlation matrix, including:

[0018] For each target cluster, the distance between the first expression value of N specified genes corresponding to the target cluster and the centroid value corresponding to the preset subtype is calculated to obtain the similarity between the target cluster and the preset subtype;

[0019] The correlation matrix is ​​determined based on the similarity between the target cluster and the preset subtype.

[0020] In one possible implementation, determining the first expression values ​​of N specified genes in the target cluster includes:

[0021] For each specified gene, the expression values ​​of the specified gene at all analysis units in the target cluster are merged to obtain the first expression value of the specified gene in the target cluster.

[0022] In one possible implementation, after obtaining the first expression value of the specified gene in the target cluster, the method further includes:

[0023] Obtain a known tissue sample and a second expression value of each specified gene in the known tissue sample, wherein the known tissue sample is a tissue sample with a known subtype;

[0024] For each specified gene, the first expression value corresponding to the specified gene is normalized based on the second expression value corresponding to the specified gene.

[0025] In one possible implementation, cell type abundance data corresponding to each analysis unit in the spatial transcriptome data of the tissue sample to be tested is obtained, including:

[0026] Obtain single-cell transcriptome data from the tissue sample to be tested;

[0027] Based on the spatial transcriptome data and the single-cell transcriptome data, the cell type abundance data corresponding to each analysis unit is determined by deconvolution.

[0028] Secondly, embodiments of this application provide a molecular typing device, comprising:

[0029] The first determination module is used to obtain cell type abundance data corresponding to each analysis unit in the spatial transcriptome data of the tissue sample to be tested.

[0030] The clustering module is used to perform spatial clustering on each of the analysis units based on the cell type abundance data corresponding to each analysis unit, and obtain multiple clusters.

[0031] The second determining module is used to determine at least one target cluster based on the highest proportion of cell types corresponding to each of the clusters, wherein the target cluster represents the tumor region of the tissue sample to be tested;

[0032] The third determining module is used to determine the first expression value of N specified genes in each target cluster, and to construct a first gene expression matrix based on the first expression value, wherein N is an integer greater than 1;

[0033] The genotyping module is used to determine the correlation based on the first gene expression matrix and the preset genotyping centroid matrix to obtain the genotyping result corresponding to each target cluster, wherein the genotyping centroid matrix includes the centroid values ​​of N specified genes under each preset genotyping.

[0034] Thirdly, embodiments of this application provide an electronic device, including:

[0035] Memory, used to store programs;

[0036] A processor for executing a program stored in the memory, wherein when the processor executes the program stored in the memory, the processor is configured to perform the method described in any of the first aspects above.

[0037] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions for performing the method described in any of the first aspects above.

[0038] The solutions provided in the second to fourth aspects above are used to implement or cooperate with the molecular typing method provided in the first aspect above, and therefore can achieve the same or corresponding beneficial effects as the first aspect, which will not be elaborated here.

[0039] It should be understood that the above general description and the following detailed description are merely exemplary and do not limit this application. Attached Figure Description

[0040] Figure 1 A schematic diagram of an implementation environment for the molecular typing method provided in this application embodiment;

[0041] Figure 2 A schematic flowchart of a molecular typing method provided in an embodiment of this application;

[0042] Figure 3 An optional schematic diagram of bin50 provided in an embodiment of this application;

[0043] Figure 4 A schematic diagram of the spatial distribution of cell types in a tissue, provided in an embodiment of this application;

[0044] Figure 5 Clustering heatmaps provided for embodiments of this application;

[0045] Figure 6 This is a schematic diagram of the cluster distribution obtained by spatial clustering of the analysis units in the spatial transcriptome data corresponding to the tissue sample to be tested, as provided in the embodiments of this application.

[0046] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0047] To make the objectives, technical methods, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0048] It should be noted that the meaning of "multiple" (or "more than") in the description of the embodiments of this application refers to two or more, and "greater than," "less than," "exceeding," etc. are understood to exclude the number itself, while "above," "below," "within," etc. are understood to include the number itself. If "first," "second," etc. are used in the description, they are only for the purpose of distinguishing technical features and should not be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated or the order of the technical features indicated.

[0049] In this application embodiment, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent the existence of A alone, the simultaneous existence of A and B, or the existence of B alone. A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of singular or plural items. For example, at least one of a, b, and c can represent: the existence of a alone, the existence of b alone, the existence of c alone, the simultaneous existence of a and b, the simultaneous existence of a and c, the simultaneous existence of b and c, or the simultaneous existence of a, b, and c, where a, b, and c can be single or multiple.

[0050] First, a brief introduction to the terminology used in the embodiments of this application will be given.

[0051] Tumor molecular subtyping refers to a method of classifying tumors based on their molecular characteristics, such as gene expression, protein levels, metabolites, and epigenetics. This classification approach helps to deepen our understanding of the biological characteristics of tumors and provides a basis for precision medicine.

[0052] Spatial transcriptomics is an experimental technique used to study the spatial distribution of cells on tissue sections. Traditionally, transcriptomics focuses primarily on gene expression at the whole-cell level. Spatial transcriptomics allows for the simultaneous detection of gene expression levels at hundreds or thousands of locations on a tissue section, thus creating a spatial map of gene expression. The technique involves placing a tissue section on a slide with spatial barcodes, allowing the barcode primers to bind and capture neighboring messenger ribonucleic acid (mRNA) from the tissue. The captured mRNA then undergoes reverse transcription, resulting in complementary deoxyribonucleic acid (cDNA) containing the spatial barcodes. By analyzing the sequences of these spatial barcodes in the sequencing results, the transcribed sequence of each mRNA can be mapped back to its starting position on the tissue section.

[0053] Single-cell RNA sequencing (scRNA-seq) is a technique for analyzing gene expression at the single-cell level.

[0054] bin: refers to a small region into which a tissue slice is divided, used to capture and analyze gene expression data within that region; bin1 is the smallest unit of analysis, which is a 500 nm x 500 nm square, while binM represents a region composed of M x M bin1 units.

[0055] Spot: In spatial transcriptomics, a spot is a marker on a tissue slice. It is a small region in the tissue. Each spot contains a specific spatial barcode to identify its location on the tissue slice and to help analyze the spatial distribution of gene expression data.

[0056] Tumor molecular subtyping is a method of classifying tumors based on their molecular characteristics, which can provide a basis for accurate cancer diagnosis, prognostic stratification, and tumor staging.

[0057] Taking breast cancer subtyping as an example, related technologies classify tumors into four intrinsic subtypes based on gene expression levels in breast cancer samples: luminal A (LumA), luminal B (LumB), HER2-enriched (HER2), and basal-like (Basal). These different subtypes have been shown to have different prognoses in untreated and tamoxifen-treated patients. Therefore, different treatment strategies can be developed based on different subtypes, and the risk of distant metastasis of breast cancer can be assessed based on subtype and tumor proliferation index.

[0058] Currently, the most common method for detecting molecular subtypes of breast cancer is four-molecule immunohistochemistry (IHC), which estimates the molecular subtype of breast cancer by semi-quantitatively detecting the expression levels of ER, PR, HER2, and Ki67 proteins. While IHC is relatively simple, it suffers from inaccurate subtyping, potentially misleading clinical treatment plans and leading to adverse outcomes. Furthermore, current techniques for tumor molecular subtyping are based on whole-piece tissue samples, failing to accurately distinguish or explore molecular subtyping differences between different regions of pathological sections.

[0059] To address the technical problems existing in related technologies, this application provides a molecular typing method. Based on the cell type abundance data corresponding to each analysis unit in the spatial transcriptome data of the tissue sample to be tested, spatial clustering is performed on the analysis units to obtain multiple clusters. Then, based on the highest proportion of cell types corresponding to each cluster, target clusters representing tumor regions are screened from all clusters, thereby distinguishing tumor regions and non-tumor regions in the tissue sample to be tested. For the target clusters representing tumor regions, a first gene expression matrix is ​​constructed based on the first expression values ​​of N specified genes in each target cluster, and a typing centroid matrix is ​​obtained. The typing centroid matrix includes the centroid values ​​of N specified genes under each preset typing. The correlation between the first gene expression matrix and the typing centroid matrix is ​​determined to obtain the typing results corresponding to each target cluster. Compared with related technologies that perform molecular typing based on the entire tissue sample, this application can achieve more refined molecular typing, realize the division of spatial tumor regions in the tissue, and accurately provide molecular typing results for tumor regions, which is of great significance for improving the accuracy and effectiveness of tumor treatment.

[0060] It should be noted that, for ease of description, this application embodiment will take the molecular subtyping of breast cancer as an example to introduce the specific implementation process of the molecular subtyping method provided in this application embodiment. It should be understood that the molecular subtyping method provided in this application embodiment is not limited to the molecular subtyping of breast cancer, but can also be applied to the molecular subtyping of other cancers, and this application embodiment does not impose any limitations on this.

[0061] The implementation environment of the molecular typing method provided in the embodiments of this application is described below.

[0062] Figure 1 This is a schematic diagram of an implementation environment for the molecular typing method provided in this application, including a terminal device 101 and a server 102. The terminal device 101 and the server 102 can communicate with each other via wired or wireless means.

[0063] In this embodiment, the object and the terminal device 101 can interact with each other. For example, the object can input relevant data of the tissue sample to be tested on the terminal device 101. The terminal device 101 can also present the typing result of the tissue sample to be tested to the object, such as presenting the probability that the tissue sample belongs to a certain subtype. The server 102 has data processing functions.

[0064] In some embodiments of this application, server 102 stores an application program for implementing the molecular typing method of the embodiments of this application. In the embodiments of this application, the user can input relevant data of a tissue sample to be tested on terminal device 101. The relevant data may include spatial transcriptome data and single-cell transcriptome data of the tissue sample to be tested. Terminal device 101 sends the relevant data of the tissue sample to be tested to server 102. Based on the relevant data of the tissue sample to be tested, server 102 determines the cell type abundance data corresponding to each analysis unit in the spatial transcriptome data of the tissue sample to be tested. For example, if the relevant data of the tissue sample to be tested includes spatial transcriptome data and single-cell transcriptome data, server 102 determines the cell type abundance data corresponding to each analysis unit based on the spatial transcriptome data and single-cell transcriptome data of the tissue sample to be tested. Then, server 102 performs spatial clustering on each analysis unit based on the cell type abundance data corresponding to each analysis unit to obtain multiple clusters. Then, server 102 determines at least one target cluster based on the highest proportion of cell types corresponding to each cluster. This target cluster is used to represent the tumor region of the tissue sample to be tested. Next, server 102 determines the first expression value of N (N greater than 1) specified genes in each target cluster and constructs a first gene expression matrix based on the first expression value. Then, server 102 performs correlation determination based on the first gene expression matrix and a preset genotyping centroid matrix to obtain the genotyping result corresponding to each target cluster. In this way, server 102 completes the genotyping processing of the tissue sample to be tested based on the relevant data of the tissue sample. Optionally, server 102 sends the processing result of the tissue sample to be tested to terminal device 101, and terminal device 101 presents the processing result to the target. For example, the processing result includes the tumor region and non-tumor region division result of the tissue sample to be tested, and the genotyping result for the tumor region. Therefore, this embodiment of the application, based on the cell type abundance data corresponding to each analysis unit in the spatial transcriptome data, distinguishes the tumor region and non-tumor region of the tissue sample to be tested through clustering, and then performs molecular subtyping on the tumor region. Compared with related technologies that perform molecular subtyping based on the whole tissue sample, this embodiment of the application can achieve more refined molecular subtyping, realize the division of the spatial tumor region of the tissue and accurately give the molecular subtyping results of the tumor region, which is of great significance for improving the accuracy and effectiveness of tumor treatment.

[0065] In some embodiments of this application, the molecular typing method is primarily performed by a terminal device. In these embodiments, the user can input relevant data of the tissue sample to be tested onto the terminal device 101. This relevant data may include spatial transcriptome data and single-cell transcriptome data of the tissue sample. Based on the relevant data of the tissue sample, the terminal device 101 determines the cell type abundance data corresponding to each analysis unit in the spatial transcriptome data of the tissue sample. Next, based on the cell type abundance data corresponding to each analysis unit, the terminal device 101 performs spatial clustering on each analysis unit to obtain multiple clusters. Alternatively, the terminal device 101 can send the cell type abundance data corresponding to each analysis unit to the server 102, requesting the server 102 to perform spatial clustering on each analysis unit based on the cell type abundance data, and return the clustering results to the terminal device 101. This clustering result includes multiple clusters, and each cluster contains several analysis units. Then, based on the highest percentage of cell types corresponding to each cluster, the terminal device 101 determines at least one target cluster, which represents the tumor region of the tissue sample. Next, terminal device 101 determines the first expression value of N (N greater than 1) specified genes in each target cluster and constructs a first gene expression matrix based on the first expression value. Then, terminal device 101 performs correlation determination based on the first gene expression matrix and a preset genotyping centroid matrix to obtain the genotyping result corresponding to each target cluster. Alternatively, terminal device 101 can send the first gene expression matrix to server 102 to request server 102 to perform correlation determination based on the first gene expression matrix and the preset genotyping centroid matrix, obtain the genotyping result corresponding to each target cluster, and then return the genotyping result to terminal device 101. In this way, terminal device 101 completes the genotyping processing of the tissue sample to be tested based on the relevant data of the tissue sample. Optionally, terminal device 101 presents the processing result to the object; for example, the processing result includes the tumor region and non-tumor region division results of the tissue sample to be tested, and the genotyping result for the tumor region. Therefore, this embodiment of the application, based on the cell type abundance data corresponding to each analysis unit in the spatial transcriptome data of the tissue sample to be tested, distinguishes the tumor region and non-tumor region of the tissue sample to be tested through clustering, and then performs molecular subtyping on the tumor region. Compared with related technologies that perform molecular subtyping based on the whole tissue sample, this embodiment of the application can achieve more refined molecular subtyping, realize the division of the tissue spatial tumor region and accurately give the molecular subtyping result of the tumor region, which is of great significance for improving the accuracy and effectiveness of tumor treatment.

[0066] This application does not limit the specific type of the terminal device 101. In some embodiments, the terminal device 101 may include, but is not limited to, mobile phones, computers, medical devices, etc. The device is often equipped with a display device, which may also be a monitor, display screen, touch screen, etc., and the touch screen may also be a touch screen, touch panel, etc.

[0067] In some embodiments, there may be one or more servers. When there are multiple servers, at least two servers are used to provide different services, and / or at least two servers are used to provide the same service, such as providing the same service in a load-balanced manner. This application embodiment does not limit this. The aforementioned servers may be independent physical servers, server clusters or distributed systems composed of multiple physical servers, or cloud servers providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. Servers may also become nodes in a blockchain.

[0068] In this embodiment, the terminal device 101 and the server 102 can be directly or indirectly connected through wired or wireless communication, and this application does not impose any restrictions on this.

[0069] It should be noted that the implementation environment of this application embodiment includes, but is not limited to, Figure 1 As shown.

[0070] The technical solutions of the embodiments of this application will be described in detail below through some examples. The following embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0071] Figure 2 This is a schematic flowchart of a molecular typing method provided in an embodiment of this application. The executing entity of this molecular typing method is a device with tumor molecular typing function, such as a molecular typing device. In some embodiments, the executing entity of this molecular typing method may be... Figure 1 The server in, or for Figure 1 Terminal devices in, or for Figure 1 The system consists of a server and terminal devices. For ease of description, this application uses an electronic device as an example for illustration.

[0072] like Figure 2 As shown in the embodiment of this application, a molecular typing method includes the following steps S210-S250:

[0073] Step S210: Obtain the cell type abundance data corresponding to each analysis unit in the spatial transcriptome data of the tissue sample to be tested.

[0074] It should be noted that the tissue sample to be tested is a tissue section, and the analysis unit in the spatial transcriptome data corresponding to the tissue sample can also be called a spot. It can be understood that a tissue section includes multiple spots, and each spot is an analysis unit of the tissue section, possessing corresponding coordinate information and gene expression information.

[0075] It should be noted that, in some embodiments of this application, before obtaining the cell type abundance data corresponding to each analysis unit in the spatial transcriptome data of the tissue sample to be tested, the molecular typing method further includes: dividing the tissue sample to be tested into multiple bin1 regions, each bin1 region having corresponding gene expression information; and taking a bin M region composed of M*M bin1 regions as an analysis unit, where M is an integer greater than 1. For example, dividing the tissue sample to be tested into multiple bin1 regions and merging 50*50 bin1 regions together can form a bin50 region, which can then be taken as an analysis unit. Optionally, each analysis unit uses its center coordinates as its identifier.

[0076] It is understood that the analysis unit in this embodiment can be regarded as a bin M expression matrix, composed of M*M bin1 expression matrices. Each bin1 expression matrix includes: gene ID, the coordinate corresponding to the gene ID, and the gene expression level corresponding to the gene ID at that coordinate. Compared with using bin1 expression matrices to construct analysis units, using bin M expression matrices in this embodiment can avoid many genes not being expressed in bin1, ensuring the richness of gene information carried by each analysis unit and providing support for subsequent genotyping. It should be understood that in specific implementations, the acquisition method and size of the analysis unit can be selected according to actual needs, and this embodiment does not impose any restrictions on this.

[0077] It should be noted that, in the embodiments of this application, the cell type abundance data corresponding to each analysis unit represents the relative abundance of different cell types at each analysis unit. For example, the cell type abundance data corresponding to an analysis unit includes the proportion of multiple preset cell types at that analysis unit.

[0078] As one possible implementation, obtaining cell type abundance data corresponding to each analysis unit in the spatial transcriptome data of the tissue sample to be tested includes: obtaining single-cell transcriptome data of the tissue sample to be tested; and determining the cell type abundance data corresponding to each analysis unit by deconvolution based on the spatial transcriptome data and the single-cell transcriptome data.

[0079] In other words, the method of combining spatial transcriptome data with single-cell transcriptome data for analysis determines the cell type abundance data corresponding to each analysis unit in the spatial transcriptome data. In practice, tools such as cell2location can be used to integrate single-cell transcriptome data (which provides transcriptional features of cell types) and spatial transcriptome data. By deconvolution, the composition and abundance of cell types at each analysis unit in the spatial transcriptome data can be predicted. This allows us to determine which cell type dominates at different spatial locations in tissue sections and to draw cell type distribution maps of tissue sections.

[0080] It should be understood that, in addition to the method of combining spatial transcriptome data with single-cell transcriptome data for analysis, other suitable methods can be selected according to actual needs to determine the cell type distribution in the tissue sample to be tested, and then determine the cell type abundance data corresponding to each analysis unit in the spatial transcriptome data. For example, a deep learning model can be used based on the spatial transcriptome gene expression data and histological images of the tissue sample to be tested to predict the cell type distribution of the tissue sample to be tested and to annotate the corresponding cell components at each analysis unit, thereby obtaining the cell type abundance data corresponding to each analysis unit. This application does not limit this approach.

[0081] Step S220: Based on the cell type abundance data corresponding to each analysis unit, perform spatial clustering on each analysis unit to obtain multiple clusters.

[0082] As one possible implementation, the Leiden algorithm can be used to spatially cluster each analysis unit based on the cell type abundance data corresponding to each analysis unit, thereby obtaining multiple clusters. It should be understood that, in addition to the Leiden algorithm, algorithms such as K-Means, DBSCAN, or Gaussian Mixture Models (GMM) can also be used to perform spatial clustering of analysis units in the spatial transcriptome data corresponding to the tissue sample to be tested, and this application embodiment does not limit this.

[0083] Understandably, spatial clustering of the analytical units in the spatial transcriptome data corresponding to the tissue sample under test, based on the cell type abundance data corresponding to each analytical unit, can group analytical units with similar cell compositions into a cluster. Here, a cluster represents a specific region of the tissue sample under test. In other words, after spatially clustering the analytical units in the spatial transcriptome data corresponding to the tissue sample under test based on the cell type abundance data corresponding to each analytical unit, the tissue sample under test can be divided into multiple regions, and the analytical units in each region have similar cell compositions.

[0084] Step S230: Determine at least one target cluster based on the highest percentage of cell types corresponding to each cluster, wherein the target cluster represents the tumor region of the tissue sample to be tested.

[0085] In this embodiment, after dividing the analysis units in the spatial transcriptome data corresponding to the tissue sample to be tested into multiple clusters, the highest proportion of cell types corresponding to each cluster is used to determine whether each cluster is a target cluster. If it is a target cluster, the region represented by the target cluster is taken as the tumor region. In this way, by determining whether each cluster is a target cluster, tumor and non-tumor regions can be separated from the tissue sample to be tested, and molecular subtyping can be performed on the tumor region to obtain more refined subtyping results.

[0086] As one possible implementation, the process of determining at least one target cluster based on the highest proportion of cell types corresponding to each cluster can be as follows: for each cluster, the proportion of each cell type in the cluster is determined based on the cell type abundance data corresponding to the analysis unit in the cluster, and the highest proportion of cell type in the cluster is determined based on the proportion of each cell type in the cluster; the cluster with the highest proportion of cell type being the preset target cell type is taken as the target cluster.

[0087] For example, in breast cancer tissue sections, the tumor area usually has the highest proportion of epithelial cells. Therefore, in the process of molecular subtyping of breast cancer, for each cluster obtained by clustering, the proportion of each cell type in the cluster can be determined based on the cell type abundance data corresponding to the analysis unit in the cluster. Based on the proportion of each cell type in the cluster, the cell type with the highest proportion in the cluster can be determined, and the cluster with the highest proportion of epithelial cells can be taken as the target cluster.

[0088] Step S240: Determine the first expression value of N specified genes in each target cluster, and construct a first gene expression matrix based on the first expression value, where N is an integer greater than 1.

[0089] It is understandable that the specified gene is a known gene with a corresponding gene ID.

[0090] It is understood that the first expression value of a specified gene in the target cluster is the gene expression level of the specified gene within the region represented by the target cluster. For example, the first gene expression matrix includes the first expression value corresponding to each gene ID under the target cluster.

[0091] It is understandable that the number of target clusters can be one or more. For example, if the number of target clusters obtained through step S230 is 5 and the number of specified genes is 50, then the first expression values ​​of the 50 specified genes under different target clusters can be seen in Table 1 below.

[0092] Table 1

[0093] Target cluster 1 Target cluster 2 Target cluster 3 Target cluster 4 Target cluster 5 Gene ID1 <![CDATA[x 1,1 ]]> <![CDATA[x 1,2 ]]> <![CDATA[x 1,3 ]]> <![CDATA[x 1,4 ]]> <![CDATA[x 1,5 <!-- 8 -->]]> Gene ID2 <![CDATA[x 2,1 ]]> <![CDATA[x 2,2 ]]> <![CDATA[x 2,3 ]]> <![CDATA[x 2,4 ]]> <![CDATA[x 2,5 ]]> …… …… …… …… …… …… Gene ID50 <![CDATA[x 50,1 ]]> <![CDATA[x 50,2 ]]> <![CDATA[x 50,3 ]]> <![CDATA[x 50,4 ]]> <![CDATA[x 50,5 ]]>

[0094] As one possible implementation, the process of determining the first expression value of N specified genes in a target cluster can be to combine the expression values ​​of the specified gene at all analysis units in the target cluster for each specified gene to obtain the first expression value of the specified gene in the target cluster.

[0095] In other words, the gene expression level of a specified gene is determined at each analysis unit within the region represented by the target cluster, and then the gene expression levels at all analysis units are combined to obtain the first expression value of the specified gene in the target cluster.

[0096] As one possible embodiment, after obtaining the first expression value of the specified gene in the target cluster, the molecular typing method further includes:

[0097] Obtain known tissue samples and the second expression value of each specified gene in the known tissue samples, wherein the known tissue samples are tissue samples with known subtypes;

[0098] Then, for each specified gene, the first expression value corresponding to the specified gene is normalized based on the second expression value corresponding to the specified gene.

[0099] In other words, after obtaining the first expression value of a specified gene in the target cluster through merging, the first expression value is normalized to obtain a normalized first expression value. The first gene expression matrix includes the normalized first expression values ​​of N specified genes in the target cluster.

[0100] It should be noted that the number of known tissue samples can be multiple, and they should cover various preset subtypes. For example, breast cancer has four subtypes (LumA, LumB, Her2, Basal), so the known tissue samples should cover these four subtypes. In a specific example, the known tissue samples include 7 LumA samples, 9 LumB samples, 10 Her2 samples, and 3 Basal samples, totaling 29 samples. After obtaining the known tissue samples, the second expression values ​​of N specified genes in the known tissue samples are determined. Let the number of known tissue samples be K. Then, the second expression values ​​of the N specified genes in K known tissue samples can form an N*K expression matrix.

[0101] It should be noted that for each specified gene, the expression values ​​of the specified gene at all analysis units in the target cluster are merged to obtain the first expression value (original expression value) of the specified gene in the target cluster. If there are J target clusters, then the original expression values ​​of N specified genes in J target clusters can form an N*J expression matrix.

[0102] By integrating the above N*J expression matrix and N*K expression matrix, the first expression value (original expression value) of N specified genes in the target cluster and the second expression value of N specified genes in known tissue samples can be integrated to obtain an N*(J+K) expression matrix.

[0103] Then, based on the N*(J+K) expression matrix, the expression values ​​within the matrix are normalized. This allows for the normalization of the expression values ​​of a specified gene, obtaining a normalized first expression value. This normalized first expression value is then used to construct a first gene expression matrix. It should be understood that normalizing the first expression value of a specified gene within the target cluster can reduce noise and bias in the tested tissue sample data, improving the stability of subsequent genotyping processing for the target cluster.

[0104] It should be noted that, in practice, the expression values ​​of the specified genes in the target cluster can be normalized without directly constructing the first gene expression matrix based on the original first expression values ​​obtained by merging. This can improve the efficiency of determining the typing results of different tumor regions in the tissue sample to be tested.

[0105] Step S250: Based on the first gene expression matrix and the preset genotyping centroid matrix, perform correlation determination to obtain the genotyping results corresponding to the target cluster.

[0106] The genotyping centroid matrix includes the centroid values ​​of N specified genes under each preset genotyping.

[0107] It is understandable that by determining the correlation between the first gene expression matrix and the preset genotyping centroid matrix, that is, determining the correlation between the expression values ​​of N specified genes in the target cluster and the centroid values ​​under the preset genotyping of the genotyping centroid matrix, the preset genotyping with high correlation is taken as the genotyping corresponding to the target cluster.

[0108] As one possible implementation, correlation determination is performed based on the first gene expression matrix and a preset genotyping centroid matrix to obtain the genotyping result corresponding to each target cluster. The specific process may include:

[0109] The correlation matrix is ​​obtained by performing correlation calculation between the first gene expression matrix and the preset subtyping centroid matrix. The correlation matrix is ​​used to describe the similarity between each target cluster and different preset subtypings.

[0110] Then, based on the correlation matrix, the typing results corresponding to the target cluster are determined.

[0111] For example, the above correlation calculation can be performed using Spearman's Rank Correlation Coefficient. Of course, other correlation algorithms can also be used, and this application does not impose any limitations on these embodiments.

[0112] As one possible embodiment, the first gene matrix includes the first expression values ​​of N specified genes in each target cluster. The process of calculating the correlation between the first gene expression matrix and a preset genotyping centroid matrix to obtain a correlation matrix may include:

[0113] For each target cluster, the distance between the first expression value of N specified genes under the target cluster and the centroid value under the preset subtype is calculated to obtain the similarity between the target cluster and the preset subtype; then, based on the similarity between the target cluster and the preset subtype, the correlation matrix is ​​determined.

[0114] For example, for a first target cluster and a first preset subtype, the similarity between the first target cluster and the first preset subtype can be determined by calculating the distance between the first expression value and the corresponding centroid value of N specified genes. Based on the principle of similarity, the similarity of each target cluster with respect to different preset subtypes can be determined.

[0115] The molecular subtyping method provided in this application will be further explained below using the molecular subtyping of breast cancer as an example.

[0116] First, the tissue sample to be tested is divided into multiple bin1 regions. A bin50 region, consisting of 50x50 bin1 regions, is used as an analysis unit. It should be understood that each bin1 region contains corresponding gene expression information, including the gene ID, the coordinates corresponding to the gene ID, and the expression level of the gene ID at those coordinates. A bin1 is a 500 nm x 500 nm square, and a bin50 combines the expression of 50x50 bin1 regions, forming a 25 μm x 25 μm window. Figure 3 The diagram shown is an optional schematic of bin50 provided in an embodiment of this application.

[0117] Then, based on spatial transcriptome data and single-cell transcriptome data, each analysis unit is annotated through deconvolution, such as annotating the cell type abundance data corresponding to each analysis unit and drawing a spatial distribution map of cell types. Figure 4 The diagram shown is a schematic representation of the spatial distribution of cell types in a tissue according to an embodiment of this application, illustrating the spatial distribution of nine cell types in a tissue section.

[0118] Table 2 shows a schematic diagram of the annotation data corresponding to multiple analysis units of the tissue sample to be tested. In Table 2, the first column is the center coordinates corresponding to the analysis unit, and the first row is the name of the cell type. Table 2 includes the annotation data corresponding to the analysis unit. Figure 4 The nine corresponding cell types are shown, along with the abundance values ​​for each cell type at different analysis units.

[0119] Table 2

[0120]

[0121] Then, based on the cell type abundance data corresponding to each analysis unit, spatial clustering is performed on the analysis units in the spatial transcriptome data corresponding to the tissue sample to be tested to obtain multiple clusters.

[0122] Then, based on the fact that breast cancer is an epithelial-derived tumor, tumor regions were defined as classes with high expression of either Cancer Epithelial or Normal Epithelial. Specifically, after clustering, cell type statistics were performed on each cluster to obtain, as shown below. Figure 5 The clustering heatmap shown identifies clusters highly expressed in Cancer Epithelial or Normal Epithelial as target clusters representing tumor regions. For example... Figure 6 The diagram shown is a schematic diagram of the cluster distribution obtained by spatial clustering of the analysis units in the spatial transcriptome data corresponding to the tissue sample to be tested, according to an embodiment of this application. Figure 6 The example shows how clustering divides the analysis units in the tissue sample into several clusters: 0, 1, 3, 7, 8, and NA. Among them, 0, 1, 3, 7, and 8 are target clusters representing tumor regions.

[0123] Then, molecular subtyping is performed on the target clusters representing tumor regions. Specifically, the first expression values ​​of N specified genes in each target cluster are determined, and a first gene expression matrix is ​​constructed based on the first expression values. In this embodiment, N=50, and the target clusters include 0, 1, 3, 7, and 8, a total of 5 categories. Thus, a first gene expression matrix with a dimension of 50*5 can be constructed based on the first expression values ​​of 50 specified genes in the 5 target clusters. Correlation determination is performed based on the first gene expression matrix and a preset subtyping centroid matrix to obtain the subtyping result corresponding to each target cluster. In this embodiment, the subtyping centroid matrix includes the centroid values ​​of 50 specified genes under each preset subtyping.

[0124] For example, Table 3 below shows the centroid matrix based on 50 specified genes, which includes the centroid values ​​of the 50 specified genes under the Basal, Her2, LumA, LumB and Normal genotypes.

[0125] Table 3

[0126]

[0127]

[0128] For example, the elements in the first gene expression matrix are the normalized expression values ​​of the specified genes in the target cluster, in order to reduce noise and bias in the sample data of the tissue to be tested. In specific implementation, the original expression values ​​of 50 specified genes in the target cluster are integrated and normalized with the basic dataset of spatial genomics sequencing (containing the expression matrices of 7 LumA, 9 LumB, 10 Her2, 3 Basal genotypes, and a total of 29 known samples) to obtain the normalized expression values ​​of the specified genes in the target cluster.

[0129] For example, the basic dataset mentioned above can be referred to as Table 4 below.

[0130] Table 4 (Part 1)

[0131]

[0132]

[0133] Table 4 (Part 2)

[0134]

[0135]

[0136]

[0137] Table 4 (Part 3)

[0138]

[0139]

[0140] For example, the specific process of the above correlation determination can be as follows: traverse each target cluster, take the currently traversed target cluster as the current cluster, and then perform the following processing for the current cluster: for each specified gene, calculate the correlation between the first expression value of the specified gene in the current cluster and the centroid value in the genotyping centroid matrix to obtain the first correlation value of the specified gene corresponding to various genotypes (Basal, Her2, LumA, LumB, and Normal); then, for each genotype, merge the first correlation values ​​corresponding to the 50 specified genes under that genotype to obtain the second correlation value, which represents the correlation value between the current cluster and the genotype; finally, based on the second correlation value between the current cluster and various genotypes, determine the genotyping result corresponding to the target cluster.

[0141] Taking the calculation of the correlation between target cluster 0 and Basal genotyping as an example, firstly, for each specified gene, the first expression value of the specified gene under target cluster 0 is correlated with the centroid value corresponding to the Basal genotyping in the genotyping centroid matrix, obtaining the first correlation value of the specified gene corresponding to the Basal genotyping; then, the first correlation values ​​of 50 specified genes corresponding to Basal genotyping are combined to obtain the second correlation value, which represents the correlation between target cluster 0 and Basal genotyping. This process can be repeated to calculate the correlation value between any target cluster and any genotyping, thus obtaining the correlation matrix.

[0142] For example, Table 5 below shows the correlation values ​​(i.e., second correlation values) of the target clusters representing tumor regions, 0, 1, 3, 7, and 8, under the Basal, Her2, LumA, LumB, and Normal classifications. For each target cluster, the classification with the highest correlation value is determined as the classification result corresponding to that cluster.

[0143] Table 5

[0144] Basal Her2 LumA LumB Normal Classification results 0 -0.0542857 0.22469388 -0.2237755 0.37846939 -0.5193878 LumB 1 0.00765306 0.19387755 -0.2507143 0.37938776 -0.5215306 LumB 3 -0.0054082 0.2294898 -0.282449 0.45704082 -0.5887755 LumB 7 -0.0035714 0.26877551 -0.3041837 0.49683673 -0.6031633 LumB 8 -0.0995918 0.20785714 -0.1989796 0.40020408 -0.5243878 LumB

[0145] The scheme in this application uses an expression matrix derived from the tumor region for genotyping, which can compensate for the shortcomings of mixed transcriptome sequencing sampling that may capture other regions. Furthermore, the scheme in this application can effectively distinguish tumor regions and clearly define spatial genotyping, thereby providing more accurate guidance for treatment.

[0146] The methods of the embodiments of this application have been described above, and the apparatus of the embodiments of this application is provided below.

[0147] This application also provides a molecular typing device that performs the molecular typing method provided in any of the above embodiments. The function can be implemented in hardware or by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above functions.

[0148] In one possible implementation, the molecular typing device includes:

[0149] The first determination module is used to obtain cell type abundance data corresponding to each analysis unit in the spatial transcriptome data of the tissue sample to be tested.

[0150] The clustering module is used to perform spatial clustering of each analysis unit based on the cell type abundance data corresponding to each analysis unit, and obtain multiple clusters.

[0151] The second determining module is used to determine at least one target cluster based on the highest proportion of cell types corresponding to each cluster, wherein the target cluster represents the tumor region of the tissue sample to be tested;

[0152] The third determining module is used to determine the first expression value of N specified genes in each target cluster, and to construct a first gene expression matrix based on the first expression value, where N is an integer greater than 1;

[0153] The genotyping module is used to determine the correlation based on the first gene expression matrix and the preset genotyping centroid matrix to obtain the genotyping result corresponding to each target cluster. The genotyping centroid matrix includes the centroid values ​​of N specified genes under each preset genotyping.

[0154] It should be noted that the information interaction and execution process between the above modules / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0155] It should also be noted that the device embodiments described above are merely illustrative. For example, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0156] See Figure 7 This application also provides an electronic device 300. The electronic device 300 can be a server or a terminal, and its internal structure includes, but is not limited to:

[0157] Memory 310 is used to store programs;

[0158] The processor 320 is used to execute the program stored in the memory 310. When the processor 320 executes the program stored in the memory 310, the processor 320 is used to execute the molecular typing method as in any of the preceding embodiments.

[0159] The processor 320 and memory 310 can be connected via a bus or other means.

[0160] The memory 310, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs, such as the molecular typing method described in any embodiment of this application. The processor 320 implements the molecular typing method as described in any of the preceding embodiments by running the non-transitory software program and instructions stored in the memory 310.

[0161] The memory 310 may include a program storage area and a data storage area. The program storage area may store the operating system and application programs required for at least one function; the data storage area may store the molecular typing method described above. Furthermore, the memory 310 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory 310 may optionally include memory remotely located relative to the processor 320, and these remote memories may be connected to the processor 320 via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0162] The non-transient software program and instructions required to implement the above-described molecular typing method are stored in memory 310. When executed by one or more processors 320, the molecular typing method provided in any embodiment of this application is executed.

[0163] This application also provides a computer-readable storage medium storing computer-executable instructions for performing the above-described molecular typing method.

[0164] In one embodiment, the storage medium stores computer-executable instructions that are executed by one or more control processors, such as one or more processors 320 in the electronic device 300, which can cause the one or more processors 320 to perform the molecular typing method provided in any embodiment of this application.

[0165] The embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0166] It will be understood by those skilled in the art that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, as is known to those skilled in the art, communication media typically include computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information transmission medium.

[0167] Furthermore, one embodiment of this application also provides a computer program product, including a computer program that, when executed by a processor, implements the molecular typing method as described in any of the preceding embodiments.

[0168] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0169] It should be noted that, in the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0170] Finally, it should be noted that the above embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The protection scope of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the scope of the technology disclosed in this application, or make equivalent substitutions for some of the technical features. Such modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be covered within the protection scope of this application.

Claims

1. A molecular typing method, characterized in that, The method includes: Obtain cell type abundance data for each analysis unit in the spatial transcriptome data corresponding to the tissue sample to be tested; Based on the cell type abundance data corresponding to each of the analysis units, spatial clustering is performed on each of the analysis units to obtain multiple clusters; At least one target cluster is determined based on the highest percentage of cell types corresponding to each of the aforementioned clusters, wherein the target cluster is used to represent the tumor region of the tissue sample to be tested; Determine the first expression value of N specified genes in each target cluster, and construct a first gene expression matrix based on the first expression value, where N is an integer greater than 1; Correlation determination is performed based on the first gene expression matrix and the preset genotyping centroid matrix to obtain the genotyping result corresponding to each target cluster, wherein the genotyping centroid matrix includes the centroid values ​​of N specified genes under each preset genotyping.

2. The method according to claim 1, characterized in that, The step of determining at least one target cluster based on the highest percentage of cell types corresponding to each of the aforementioned clusters includes: For each cluster, the proportion of each cell type in the cluster is determined based on the cell type abundance data corresponding to the analysis unit in the cluster, and the highest proportion cell type corresponding to the cluster is determined based on the proportion of each cell type in the cluster. The cluster whose highest percentage of cell types is the preset target cell type is selected as the target cluster.

3. The method according to claim 1, characterized in that, The step of determining the correlation based on the first gene expression matrix and the preset genotyping centroid matrix to obtain the genotyping result corresponding to each target cluster includes: The first gene expression matrix is ​​correlated with the preset subtyping centroid matrix to obtain a correlation matrix, wherein the correlation matrix is ​​used to describe the similarity between each target cluster with respect to different preset subtypings; Based on the correlation matrix, the typing result corresponding to the target cluster is determined.

4. The method according to claim 3, characterized in that, The first gene expression matrix includes the first expression values ​​of N specified genes in each target cluster; The correlation matrix is ​​obtained by performing correlation calculation between the first gene expression matrix and the preset genotyping centroid matrix, including: For each target cluster, the distance between the first expression value of N specified genes corresponding to the target cluster and the centroid value corresponding to the preset subtype is calculated to obtain the similarity between the target cluster and the preset subtype; The correlation matrix is ​​determined based on the similarity between the target cluster and the preset subtype.

5. The method according to claim 1, characterized in that, Determine the first expression value of N specified genes in the target cluster, including: For each specified gene, the expression values ​​of the specified gene at all analysis units in the target cluster are merged to obtain the first expression value of the specified gene in the target cluster.

6. The method according to claim 5, characterized in that, After obtaining the first expression value of the specified gene in the target cluster, the method further includes: Obtain a known tissue sample and a second expression value of each specified gene in the known tissue sample, wherein the known tissue sample is a tissue sample with a known subtype; For each specified gene, the first expression value corresponding to the specified gene is normalized based on the second expression value corresponding to the specified gene.

7. The method according to claim 1, characterized in that, Obtain cell type abundance data for each analysis unit in the spatial transcriptome data corresponding to the tissue sample to be tested, including: Obtain single-cell transcriptome data from the tissue sample to be tested; Based on the spatial transcriptome data and the single-cell transcriptome data, the cell type abundance data corresponding to each analysis unit is determined by deconvolution.

8. A molecular typing device, characterized in that, include: The first determination module is used to obtain cell type abundance data corresponding to each analysis unit in the spatial transcriptome data of the tissue sample to be tested. The clustering module is used to perform spatial clustering on each of the analysis units based on the cell type abundance data corresponding to each analysis unit, and obtain multiple clusters. The second determining module is used to determine at least one target cluster based on the highest proportion of cell types corresponding to each of the clusters, wherein the target cluster represents the tumor region of the tissue sample to be tested; The third determining module is used to determine the first expression value of N specified genes in each target cluster, and to construct a first gene expression matrix based on the first expression value, wherein N is an integer greater than 1; The genotyping module is used to determine the correlation based on the first gene expression matrix and the preset genotyping centroid matrix to obtain the genotyping result corresponding to each target cluster, wherein the genotyping centroid matrix includes the centroid values ​​of N specified genes under each preset genotyping.

9. An electronic device, characterized in that, include: Memory, used to store programs; A processor for executing a program stored in the memory, wherein when the processor executes the program stored in the memory, the processor is configured to perform the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The device stores computer-executable instructions for performing the method as described in any one of claims 1 to 7.

Images