Single cell transcriptome sequencing data annotation method

Abstract

The embodiment of the invention discloses a single cell transcriptome sequencing data annotation method. The method comprises the following steps: acquiring single cell transcriptome sequencing data associated with a target disease, and clustering a plurality of cells corresponding to the single cell transcriptome sequencing data to obtain a plurality of cell clusters; aiming at each marker gene corresponding to the target disease, according to single cell transcriptome sequencing data, obtaining the cell cluster gene expression quantity of the marker gene in each cell cluster; for each cell cluster and each cell type corresponding to the target disease, accumulating the cell cluster gene expression quantity of each marker gene corresponding to the cell type in all the marker genes in the cell cluster to obtain the total gene expression quantity; and aiming at each cell cluster, according to the total gene expression quantity corresponding to the cell cluster under each cell type, annotating a target cell type for the cell cluster from all the cell types. According to the technical scheme provided by the embodiment of the invention, the accuracy of cell type annotation can be ensured.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of bioinformatics, and in particular to a single-cell transcriptome sequencing data annotation method. Background Art

[0002] Single cell RNA sequencing (scRNA-seq) is a powerful technology that can reveal cellular heterogeneity in tissues with high resolution, bringing new perspectives to disease research.

[0003] Previously, single-cell transcriptome sequencing data were mainly analyzed through manual annotation methods, but they were inefficient and difficult to ensure accuracy. In response to this, automatic annotation methods emerged.

[0004] However, although the currently used automatic annotation methods can significantly improve the annotation efficiency, they still cannot guarantee the accuracy of the annotation, especially the accuracy when annotating cell types, which needs to be urgently addressed. Summary of the invention

[0005] The embodiment of the present invention provides a single-cell transcriptome sequencing data annotation method, which solves the problem that the accuracy of annotation for cell types cannot be guaranteed.

[0006] According to one aspect of the present invention, a method for annotating single-cell transcriptome sequencing data is provided, which may include:

[0007] For a target subject suffering from a target disease, single-cell transcriptome sequencing data of the target subject is obtained, and based on the single-cell transcriptome sequencing data, multiple cells corresponding to the single-cell transcriptome sequencing data are clustered to obtain multiple cell clusters;

[0008] For each marker gene corresponding to the target disease, the cell cluster gene expression level of the marker gene in each cell cluster is obtained based on the single-cell transcriptome sequencing data;

[0009] For each cell cluster and each cell type corresponding to the target disease, the cell cluster gene expression levels of each marker gene corresponding to the cell type in all marker genes in the cell cluster are accumulated to obtain the total gene expression level;

[0010] For each cell cluster, the target cell type is annotated for the cell cluster from all cell types according to the total gene expression level corresponding to each cell type of the cell cluster.

[0011] The technical solution of the embodiment of the present invention is to obtain single-cell transcriptome sequencing data of a target object suffering from a target disease, and cluster multiple cells corresponding to the single-cell transcriptome sequencing data based on the single-cell transcriptome sequencing data to obtain multiple cell clusters, so as to perform cell type annotation for the multiple cell clusters respectively; for each marker gene corresponding to the target disease, the cell cluster gene expression amount of the marker gene in each cell cluster is obtained according to the single-cell transcriptome sequencing data, and the cell cluster gene expression amount corresponds to the marker gene, and the marker gene corresponds to the target disease, so that the annotation result obtained by the subsequent cell type annotation based on the cell cluster gene expression amount is an annotation result closely related to the target disease; for each cell cluster and each cell type corresponding to the target disease, the cell cluster gene expression amount of each marker gene corresponding to the cell type in all marker genes in the cell cluster is accumulated to obtain the total gene expression amount, and the total gene expression amount can characterize the gene expression of a certain cell type under a certain cell cluster; thus, for each cell cluster, the target cell type can be annotated for the cell cluster from all cell types according to the total gene expression amount corresponding to each cell type of the cell cluster. Compared with annotating cell types by applying universal marker genes, the above technical solution annotates specific cell types related to the target disease by applying marker genes that are fully compatible with the target disease, thereby ensuring the accuracy of cell type annotation.

[0012] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0014] Figure 1 This is a flow chart of a single-cell transcriptome sequencing data annotation method provided by an embodiment of the present invention;

[0015] Figure 2 is a flowchart of another single-cell transcriptome sequencing data annotation method provided by an embodiment of the present invention;

[0016] Figure 3 is a schematic diagram of a quality control process in another single-cell transcriptome sequencing data annotation method provided by an embodiment of the present invention;

[0017] Figure 4 is a schematic diagram of cell clustering results in another single-cell transcriptome sequencing data annotation method provided by an embodiment of the present invention;

[0018] Figure 5 is a schematic diagram of cell type annotation results in another single-cell transcriptome sequencing data annotation method provided by an embodiment of the present invention;

[0019] Figure 6 is a schematic diagram of cell type verification results in another single-cell transcriptome sequencing data annotation method provided by an embodiment of the present invention;

[0020] Figure 7 is a structural block diagram of a single-cell transcriptome sequencing data annotation device provided according to an embodiment of the present invention;

[0021] Figure 8 It is a schematic diagram of the structure of an electronic device for implementing the single-cell transcriptome sequencing data annotation method of an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0023] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. The situations of "target", "original", etc. are similar and will not be repeated here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0024] It should be noted that the collection, collection, updating, analysis, processing, use, transmission, storage and other aspects of user personal information involved in the technical solution of the present invention are in compliance with the provisions of relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. Necessary measures are taken for user personal information to prevent illegal access to user personal information data and maintain the security of user personal information and network security.

[0025] Figure 1 It is a flow chart of a single-cell transcriptome sequencing data annotation method provided by an embodiment of the present invention. This embodiment is applicable to the case where cell type annotation is performed based on single-cell transcriptome sequencing data, and is particularly applicable to the case where cell type annotation related to the target disease is performed based on single-cell transcriptome sequencing data related to the target disease. The method can be performed by a single-cell transcriptome sequencing data annotation device provided by an embodiment of the present invention, which can be implemented by software and / or hardware, and the device can be integrated on an electronic device, which can be various user terminals or servers.

[0026] See also Figure 1 The method of the embodiment of the present invention specifically includes the following steps:

[0027] S110. For a target subject suffering from a target disease, single-cell transcriptome sequencing data of the target subject is obtained, and based on the single-cell transcriptome sequencing data, multiple cells corresponding to the single-cell transcriptome sequencing data are clustered to obtain multiple cell clusters.

[0028] Among them, the target disease can be understood as a disease for which cell type annotation is expected. In an embodiment of the present invention, optionally, the disease can be, for example, achalasia, gastric ulcer, leukemia or diabetes, etc. This can be set according to actual needs and is not specifically limited here.

[0029] The target object can be understood as an object suffering from a target disease, and single-cell transcriptome sequencing data of the object is obtained, and for multiple cells corresponding to the single-cell transcriptome sequencing data, the multiple cells are clustered based on the single-cell transcriptome sequencing data to obtain multiple cell clusters.

[0030] On this basis, in combination with application scenarios that may be involved in the embodiments of the present invention, optionally, the clustering results of the multiple cells may be visualized so that researchers can intuitively obtain the clustering results.

[0031] S120. For each marker gene corresponding to the target disease, the cell cluster gene expression level of the marker gene in each cell cluster is obtained according to the single-cell transcriptome sequencing data.

[0032] Among them, the marker gene can be understood as a gene associated with the target disease obtained in advance, and the number of the gene can be one, two or more, which is related to the actual situation and is not specifically limited here. For example, taking achalasia as an example, its marker genes include CD8A, FCGR3A, MS4A7, PDGFRA, LUM, DCN, COL1A1, CD79A, CD79B, CD19 and MS4A1, etc.

[0033] For each marker gene among all marker genes, the cell cluster gene expression level of the marker gene in each cell cluster is obtained, that is, the gene expression level of a certain cell cluster can be understood as the gene expression level of the marker gene in the cell cluster corresponding to the gene expression level of the cell cluster.

[0034] S130. For each cell cluster and each cell type corresponding to the target disease, the cell cluster gene expression levels of each marker gene corresponding to the cell type in all marker genes in the cell cluster are accumulated to obtain the total gene expression level.

[0035] Among them, the cell type can be understood as a pre-obtained cell type associated with the target disease, and the number of the cell type can be one, two or more, which is related to the actual situation and is not specifically limited here. For example, taking achalasia as an example, the corresponding cell types may include CD8+T, FCGR3A+Mono, Fibroblast and B CELL, etc.

[0036] For each cell type, a marker gene X corresponding to the cell type among all marker genes can be obtained, and the number of the marker gene X can be one, two or more, which is related to the actual situation and is not specifically limited here. For example, taking the cell type B CELL as an example, the marker gene X corresponding to B CELL may include CD79A, CD79B, CD19 and MS4A1.

[0037] On this basis, further, for each cell cluster, the cell cluster gene expression amount corresponding to each marker gene X in the cell cluster can be accumulated to obtain the total gene expression amount, that is, the total gene expression amount can be understood as the gene expression amount corresponding to the cell type in the cell cluster. Exemplarily, here still taking B CELL as an example, the total gene expression amount corresponding to B CELL in the cell cluster = the cell cluster gene expression amount of CD79A in the cell cluster + the cell cluster gene expression amount of CD79B in the cell cluster + the cell cluster gene expression amount of CD19 in the cell cluster + the cell cluster gene expression amount of MS4A1 in the cell cluster.

[0038] At this point, the total gene expression level corresponding to each cell type in the cell cluster can be obtained. Assuming there are N cell clusters and O cell types, one cell cluster corresponds to O total gene expression levels.

[0039] S140. For each cell cluster, according to the total gene expression amount corresponding to each cell type of the cell cluster, annotate the target cell type for the cell cluster from all cell types.

[0040] Among them, for each cell cluster, the total gene expression values ​​corresponding to each cell type of the cell cluster can be obtained according to S130, for example, the above example can obtain 0 total gene expression values ​​corresponding to the cell cluster. On this basis, according to these total gene expression values, the target cell type can be determined from all cell types, and the target cell type is used as the cell type annotation result of the cell cluster.

[0041] On this basis, in combination with application scenarios that may be involved in the embodiments of the present invention, optionally, the annotation results of the multiple cell clusters may be visualized so that researchers can intuitively obtain the annotation results.

[0042] According to the above analysis, the embodiment of the present invention is to perform cell type annotation based on the accumulation of gene expression. Specifically, for each cell cluster, the cell cluster gene expression of the marker gene of each cell type for the cell cluster can be summed up, and the total gene expression of each cell type for the cell cluster can be obtained by direct accumulation. The higher the total gene expression, the higher the degree of match between the cell cluster and the cell type. Thus, the most likely cell type can be automatically annotated for the cell cluster by comparing the total gene expression corresponding to the different cell types under the cell cluster.

[0043] The technical solution of the embodiment of the present invention is to obtain single-cell transcriptome sequencing data of a target object suffering from a target disease, and cluster multiple cells corresponding to the single-cell transcriptome sequencing data based on the single-cell transcriptome sequencing data to obtain multiple cell clusters, so as to perform cell type annotation for the multiple cell clusters respectively; for each marker gene corresponding to the target disease, the cell cluster gene expression amount of the marker gene in each cell cluster is obtained according to the single-cell transcriptome sequencing data, and the cell cluster gene expression amount corresponds to the marker gene, and the marker gene corresponds to the target disease, so that the annotation result obtained by the subsequent cell type annotation based on the cell cluster gene expression amount is an annotation result closely related to the target disease; for each cell cluster and each cell type corresponding to the target disease, the cell cluster gene expression amount of each marker gene corresponding to the cell type in all marker genes in the cell cluster is accumulated to obtain the total gene expression amount, and the total gene expression amount can characterize the gene expression of a certain cell type under a certain cell cluster; thus, for each cell cluster, the target cell type can be annotated for the cell cluster from all cell types according to the total gene expression amount corresponding to each cell type of the cell cluster. Compared with annotating cell types by applying universal marker genes, the above technical solution annotates specific cell types related to the target disease by applying marker genes that are fully compatible with the target disease, thereby ensuring the accuracy of cell type annotation.

[0044] An optional technical solution is to annotate the target cell type for the cell cluster from all cell types according to the total gene expression corresponding to each cell type of the cell cluster, including:

[0045] Determine the maximum gene expression with the largest value from the total gene expression corresponding to each cell type in the cell cluster;

[0046] The cell type corresponding to the maximum gene expression level among all cell types is annotated as the target cell type of the cell cluster.

[0047] The above technical solution takes into account that the higher the total gene expression level, the higher the matching degree between the cell type corresponding to the total gene expression level and the cell cluster. Therefore, by using the cell type as the cell type annotation result of the cell cluster, the accuracy of the cell type annotation is further guaranteed.

[0048] Another optional technical solution, the above-mentioned single-cell transcriptome sequencing data annotation method further includes:

[0049] For each cell type, the annotation result of the cell cluster that uses the cell type as the target cell type is verified based on the cell gene expression level in each cell of each marker gene corresponding to the cell type among all marker genes.

[0050] Among them, for each cell type, the cell gene expression level of each marker gene Y corresponding to the cell type in all marker genes is obtained in each cell, and the cell gene expression level can characterize the cells where each marker gene Y is concentrated. Therefore, the annotation accuracy of the cell cluster with the cell type as the annotation result can be verified based on the cell gene expression level. For example, if the cells in the concentration are exactly the cells in the cell cluster, it means that the annotation result of the cell cluster is accurate, otherwise there is a deviation.

[0051] The above technical solution further ensures the accuracy of cell type annotation by verifying the annotation results of cell types.

[0052] Figure 2 It is a flowchart of another single-cell transcriptome sequencing data annotation method provided by an embodiment of the present invention. This embodiment is optimized based on the above-mentioned technical solutions. In this embodiment, optionally, based on single-cell transcriptome sequencing data, multiple cells corresponding to the single-cell transcriptome sequencing data are clustered to obtain multiple cell clusters, which may include: for all genes corresponding to the single-cell transcriptome sequencing data, multiple target genes are determined from all genes according to the single-cell transcriptome sequencing data; for each target gene and multiple cells corresponding to the single-cell transcriptome sequencing data, the cell gene expression amount of the target gene in multiple cells is scaled so that the cell gene expression amount of different target genes falls within a preset range; the single-cell transcriptome sequencing data is updated according to the obtained scaling processing result, and multiple cells are clustered according to the single-cell transcriptome sequencing data to obtain multiple cell clusters. Wherein, the explanation of the terms identical or corresponding to the above-mentioned embodiments is not repeated here.

[0053] See also Figure 2 The method of this embodiment may specifically include the following steps:

[0054] S210. For a target subject suffering from a target disease, single-cell transcriptome sequencing data of the target subject is obtained, and for all genes corresponding to the single-cell transcriptome sequencing data, multiple target genes are determined from all genes according to the single-cell transcriptome sequencing data.

[0055] Among them, the target gene can be understood as a characteristic gene (or a hypervariable gene) among all genes with a large difference in cellular gene expression between different cells, which can provide sufficient information for distinguishing different cell types. In other words, the difference in cellular gene expression of the target gene between different cells is greater than the difference in cellular gene expression of genes other than the target gene among all genes among different cells.

[0056] Based on single-cell transcriptome sequencing data, multiple target genes were identified from all genes.

[0057] S220, for each target gene and multiple cells corresponding to the single-cell transcriptome sequencing data, scaling processing is performed on the cellular gene expression amounts of the target gene in the multiple cells, so that the cellular gene expression amounts of different target genes fall within a preset range.

[0058] Among them, according to the above description, the difference in the cellular gene expression levels of the target gene between different cells is large. Therefore, the cellular gene expression levels of the target gene in different cells can be scaled here so that the cellular gene expression levels of different target genes fall within the preset range, thereby eliminating the differences in cellular gene expression levels of different target genes and ensuring the accuracy of subsequent cell clustering.

[0059] S230, updating the single-cell transcriptome sequencing data according to the obtained scaling processing result, and clustering the multiple cells according to the single-cell transcriptome sequencing data to obtain multiple cell clusters.

[0060] S240. For each marker gene corresponding to the target disease, the cell cluster gene expression level of the marker gene in each cell cluster is obtained based on the single-cell transcriptome sequencing data.

[0061] S250, for each cell cluster and each cell type corresponding to the target disease, the cell cluster gene expression levels of each marker gene corresponding to the cell type in all marker genes in the cell cluster are accumulated to obtain the total gene expression level.

[0062] S260. For each cell cluster, according to the total gene expression amount corresponding to each cell type of the cell cluster, annotate the target cell type for the cell cluster from all cell types.

[0063] The technical solution of the embodiment of the present invention determines multiple target genes from all genes and scales the cellular gene expression levels of each target gene so that they fall within the same range, thereby eliminating the differences in cellular gene expression levels of different target genes and ensuring the accuracy of subsequent cell clustering.

[0064] On this basis, an optional technical solution is to identify multiple target genes from all genes based on single-cell transcriptome sequencing data, including:

[0065] For each gene, the expression difference coefficient of the gene between multiple cells is obtained according to the cell gene expression amount of the gene in multiple cells respectively;

[0066] According to the expression difference coefficients corresponding to all genes respectively, multiple target genes are determined from all genes.

[0067] Among them, according to the above description, the target gene is a gene whose cellular gene expression levels between different cells are relatively different. Therefore, for each gene, the expression difference coefficient of the gene between multiple cells can be obtained according to the cellular gene expression levels of the gene in multiple cells; further, based on the expression difference coefficients corresponding to different genes, the target gene is determined.

[0068] The above technical solution achieves the effect of accurate screening of target genes by utilizing the expression difference coefficient.

[0069] Another optional technical solution, before determining multiple target genes from all genes based on the single-cell transcriptome sequencing data, the single-cell transcriptome sequencing data annotation method further includes:

[0070] The single-cell transcriptome sequencing data is standardized, and the single-cell transcriptome sequencing data is updated according to the obtained standardized processing results.

[0071] The above technical solution eliminates the impact caused by differences in sequencing depth by standardizing the single-cell transcriptome sequencing data, thereby ensuring the accuracy of target gene screening.

[0072] On this basis, optionally, before the single-cell transcriptome sequencing data is standardized, the single-cell transcriptome sequencing data annotation method further includes:

[0073] According to the single-cell transcriptome sequencing data, the sequencing quality of each cell is obtained, and multiple cells are filtered according to the obtained total sequencing quality;

[0074] According to the obtained filtering results, the single-cell transcriptome sequencing data is updated.

[0075] Among them, in this technical solution, the sequencing quality can be expressed in a variety of ways, for example, it can be expressed by at least one of the number of gene features, the amount of RNA, and the proportion of mitochondrial genes.

[0076] Based on the single-cell transcriptome sequencing data, the sequencing quality of each cell is obtained, and then multiple cells can be filtered based on the sequencing qualities to filter out low-quality cells among the multiple cells.

[0077] The above technical solution removes low-quality cells through quality control (QC), and correspondingly removes sequencing data related to the low-quality cells in the single-cell transcriptome sequencing data, thereby further improving the accuracy of cell type annotation.

[0078] Another optional technical solution, after updating the single-cell transcriptome sequencing data according to the obtained scaling processing result, the single-cell transcriptome sequencing data annotation method further includes:

[0079] The single-cell transcriptome sequencing data is subjected to dimensionality reduction processing, and the single-cell transcriptome sequencing data is updated according to the obtained dimensionality reduction processing results.

[0080] The above technical solution, by performing dimensionality reduction processing on single-cell transcriptome sequencing data, extracts the main features in the single-cell transcriptome sequencing data that can best reflect the differences between cells and eliminates noise, which helps to further improve the accuracy of cell type annotation.

[0081] It should be noted that the above text explains the three concepts of cell gene expression, cell cluster gene expression and total gene expression. Their essence is gene expression. The above text only distinguishes the gene expression in different application scenarios in terms of naming, and it does not limit their actual meanings.

[0082] On this basis, in order to better understand the above-mentioned technical solutions as a whole, the following is an explanation with reference to specific examples. For example, taking achalasia as an example, the specific implementation steps are as follows:

[0083] Step 1: Identification of marker genes

[0084] Based on existing literature, databases and experimental data, a marker gene table related to achalasia is constructed, which records the cell types related to achalasia and the marker genes corresponding to various cell types. Exemplarily, the contents recorded in the marker gene table are displayed in the form of [cell type, marker gene], and some examples are shown below:

[0085] CD8+T,CD8A

[0086] FCGR3A+Mono,FCGR3A

[0087] FCGR3A+Mono,MS4A7

[0088] Fibroblast,PDGFRA

[0089] Fibroblast,LUM

[0090] Fibroblast,DCN

[0091] Fibroblast,COL1A1

[0092] B CELL,CD79A

[0093] B CELL,CD79B

[0094] B CELL,CD19

[0095] B CELL,MS4A1.

[0096] Step 2: Creation of Seurat Object

[0097] Single cell sample collection: Tissue samples were obtained from the esophageal tissue of patients with achalasia, and then the tissue samples were dissociated into single cells using single cell isolation technology.

[0098] Single-cell RNA sequencing: A standard single-cell RNA sequencing platform is used to perform RNA sequencing on isolated single cells to obtain the gene expression data of each cell to form single-cell transcriptome sequencing data.

[0099] The open source Seurat tool is used to store and manage raw single-cell transcriptome sequencing data and convert the single-cell transcriptome sequencing data into the Seurat object format.

[0100] Step 3: Quality Control (QC)

[0101] In this stage, the cell quality of each cell is determined by calculating the number of gene features (nFeature_RNA), the number of RNA (nCount_RNA), and the proportion of mitochondrial genes (percent.mt) in each cell, so as to perform quality control and remove low-quality cells. For example, Figure 3 As shown, it shows the number of gene features, RNA quantity and mitochondrial gene ratio of a single cell from left to right. In this example, cells with gene feature numbers higher than 6000, cells with gene feature numbers lower than 50, cells with RNA quantity higher than 40000 and cells with mitochondrial gene ratio higher than 100% can be filtered out.

[0102] In this step, low-quality cells are removed through quality control to reduce the interference of noise on subsequent analysis.

[0103] Step 4: Data Standardization

[0104] In order to eliminate the impact caused by differences in sequencing depth, data normalization can be performed. Exemplarily, the NormalizeData function in the Seurat package is used to normalize the gene expression matrix represented by the single-cell transcriptome sequencing data. Among them, NormalizeData uses the log-normalization method to normalize the cell gene expression, eliminating the differences in sequencing depth or different samples, thereby ensuring that the cell gene expression between different cells is comparable.

[0105] Step 5: Selection of characteristic genes (i.e. target genes mentioned above)

[0106] In the standardized single-cell transcriptome sequencing data, select the highly variable characteristic genes. The cell gene expression of these characteristic genes varies greatly between different cells, which can provide sufficient information for distinguishing different cell types. Exemplarily, characteristic genes can be selected in the following way:

[0107] By calculating the mean and variance of each gene, and screening out characteristic genes with large expression difference coefficients according to a set quantity threshold (e.g., 2000), these characteristic genes have significant expression differences between different cells, which can reflect the heterogeneity of cells. Characteristic genes can be used for subsequent clustering and dimensionality reduction analysis.

[0108] Step 6: Data Scaling and Principal Component Analysis (PCA)

[0109] The selected characteristic genes are scaled to adjust their cellular gene expression levels to the same range (eg, zero mean and unit variance), thereby eliminating the expression differences between different genes.

[0110] Next, principal component analysis (PCA) was used to reduce the dimensionality of the high-dimensional single-cell transcriptome sequencing data, extracting the main features that best reflect the differences between cells and eliminating noise.

[0111] The above steps, by standardizing and scaling the single-cell transcriptome sequencing data, eliminate the impact of sequencing depth and sample differences, and can better identify biologically significant cell types without interfering with annotation results due to technical noise.

[0112] Step 7: UMAP dimensionality reduction and neighbor graph construction

[0113] After obtaining the PCA results, the UMAP method was used to further reduce the high-dimensional data to two dimensions and visualize them. Then, a cell neighbor graph was constructed based on these reduced-dimensional single-cell transcriptome sequencing data to help understand the similarities between different cells and provide a basis for subsequent clustering analysis.

[0114] Step 8: Louvain cluster analysis

[0115] The Louvain clustering algorithm is used to cluster the standardized and dimensionally reduced single-cell transcriptome sequencing data to achieve preliminary classification of cell types. The Louvain clustering algorithm can effectively cluster cells with similar gene expression levels into a cluster, helping to identify different cell populations.

[0116] In this example, clustering is optionally performed at multiple resolutions (ie, multiple numbers of clusters) to observe the division of different cell clusters, and finally select the most suitable resolution for clustering.

[0117] The DotPlot function in the Seurat package was used to display the cell cluster gene expression levels of the marker genes in each cell cluster. The cell cluster gene expression level was represented by the average value of the cell gene expression levels of the marker genes in each cell in the corresponding cell cluster.

[0118] Furthermore, data is extracted from the generated DotPlot to form a cell cluster identifier (id), marker gene (feature.groups) and the average value of the cell gene expression after normalization and scaling (avg.exp.scaled) for each cell cluster (i.e., cell cluster gene expression).

[0119] Step 9: TSNE Dimensionality Reduction Visualization

[0120] In order to better display the distribution of cell clusters, the tSNE dimensionality reduction method can be used to map the multi-dimensional cell gene expression into a two-dimensional space for visualization. For example, see Figure 4 , each dot represents a cell, and the color of the dot represents the cell cluster to which the cell belongs (i.e., one of 0-21 in the figure).

[0121] Step 10: Automatic annotation of cell types

[0122] Automatic annotation is performed for each cell cluster to assign the most likely cell type to each cell cluster. The specific implementation process is as follows:

[0123] For a cell cluster C j , whose cell type is T k The total gene expression level S j,k It can be expressed as:

[0124]

[0125] Among them, S j,k Cell cluster C j T k The corresponding total gene expression; n k Is cell type T k The number of corresponding marker genes; avg.exp.scaled i,j Yes k The ith marker gene (i.e., marker gene i) among the marker genes is expressed in cell cluster C. j The gene expression levels of cell clusters in .

[0126] In order to identify cell clusters C j The final annotated cell type (i.e. the target cell type described above) T max , select the total gene expression S j,k The largest cell type is T k :T max = arg max k S j，k , where T max Cell cluster C j The annotation result, that is, the cell cluster C j The most likely corresponding cell type. The cell type with the highest total gene expression is selected as the annotation result.

[0127] Step 11: Visualize the annotation results

[0128] The tSNE dimensionality reduction diagram shows the automatically annotated cell clusters. For example, see Figure 5 , different cell types are displayed in different colors in the graph, allowing researchers to intuitively see the distribution of each cell cluster and the annotated cell type.

[0129] Step 12: Verify the expression of marker genes

[0130] Each cell type has a specific marker gene that can be used to verify the accuracy of the annotation results. For example, here we take the cell type B CELL as an example. Figure 6 The cell gene expression levels of some marker genes corresponding to the cell type in each cell are shown in an exemplary manner, and different colors correspond to different cell gene expression levels. Figure 6 It can be seen that the cells where these four marker genes are concentrated are exactly the cells in the cell cluster annotated as B CELL, which verifies the accuracy of the B CELL annotation.

[0131] The above example can quickly and accurately analyze single-cell transcriptome sequencing data related to achalasia, identify cell types and their gene expression characteristics closely related to achalasia, and further reveal the cellular heterogeneity of achalasia.

[0132] Figure 7 This is a structural block diagram of a single-cell transcriptome sequencing data annotation device provided in an embodiment of the present invention, and the device is used to execute the single-cell transcriptome sequencing data annotation method provided in any of the above embodiments. The device and the single-cell transcriptome sequencing data annotation method of the above embodiments belong to the same inventive concept. For details not described in detail in the embodiment of the single-cell transcriptome sequencing data annotation device, please refer to the embodiment of the single-cell transcriptome sequencing data annotation method. Figure 7The device may specifically include: a cell clustering module 310 , a cell cluster gene expression quantity obtaining module 320 , a total gene expression quantity obtaining module 330 , and a cell type annotation module 340 .

[0133] Among them, the cell clustering module 310 can be used to obtain single-cell transcriptome sequencing data of a target object suffering from a target disease, and cluster multiple cells corresponding to the single-cell transcriptome sequencing data based on the single-cell transcriptome sequencing data to obtain multiple cell clusters;

[0134] The cell cluster gene expression amount obtaining module 320 is used to obtain the cell cluster gene expression amount of the marker gene in each cell cluster according to the single-cell transcriptome sequencing data for each marker gene corresponding to the target disease;

[0135] A total gene expression amount obtaining module 330 is used to accumulate the cell cluster gene expression amount of each marker gene corresponding to the cell type in all marker genes in the cell cluster for each cell cluster and each cell type corresponding to the target disease, to obtain the total gene expression amount;

[0136] The cell type annotation module 340 can be used to annotate a target cell type for each cell cluster from all cell types according to the total gene expression amount corresponding to each cell type of the cell cluster.

[0137] Optionally, the cell type annotation module 340 may include:

[0138] A maximum gene expression amount determination unit is used to determine the maximum gene expression amount with the largest value from the total gene expression amounts corresponding to each cell type of the cell cluster;

[0139] The cell type annotation unit is used to annotate the cell type corresponding to the maximum gene expression amount among all cell types as the target cell type of the cell cluster.

[0140] Optionally, the cell clustering module 310 may include:

[0141] The target gene determination submodule is used to determine multiple target genes from all genes corresponding to the single-cell transcriptome sequencing data according to the single-cell transcriptome sequencing data;

[0142] A scaling processing submodule is used to perform scaling processing on the cellular gene expression amounts of the target gene in the multiple cells respectively for each target gene and the multiple cells corresponding to the single-cell transcriptome sequencing data, so that the cellular gene expression amounts of different target genes fall within a preset range;

[0143] The cell clustering submodule can be used to update the single-cell transcriptome sequencing data according to the obtained scaling processing results, and cluster multiple cells according to the single-cell transcriptome sequencing data to obtain multiple cell clusters.

[0144] On this basis, an optional target gene determination submodule may include:

[0145] An expression difference coefficient obtaining unit is used to obtain the expression difference coefficient of a gene between multiple cells according to the cell gene expression amounts of the gene in multiple cells respectively for each gene;

[0146] The target gene determination unit is used to determine a plurality of target genes from all the genes according to the expression difference coefficients corresponding to all the genes respectively.

[0147] Alternatively, the above device may further include:

[0148] The standardization processing module is used to standardize the single-cell transcriptome sequencing data before determining multiple target genes from all genes based on the single-cell transcriptome sequencing data, and update the single-cell transcriptome sequencing data according to the obtained standardization processing results.

[0149] On this basis, optionally, the above device may further include:

[0150] A filtering processing module can be used to obtain the sequencing quality of each cell based on the single-cell transcriptome sequencing data, and filter multiple cells based on the obtained total sequencing quality;

[0151] The filtering processing module is used to update the single-cell transcriptome sequencing data according to the obtained filtering processing results.

[0152] Alternatively, the above device may further include:

[0153] The dimensionality reduction processing module can be used to perform dimensionality reduction processing on the single-cell transcriptome sequencing data after updating the single-cell transcriptome sequencing data according to the obtained scaling processing results, and update the single-cell transcriptome sequencing data according to the obtained dimensionality reduction processing results.

[0154] Optionally, the above device may further include:

[0155] The annotation result verification module is used to verify the annotation result of the cell cluster that takes the cell type as the target cell type according to the cell gene expression level of each marker gene corresponding to the cell type in all marker genes in each cell.

[0156] Optionally, the above device may further include:

[0157] A clustering result visualization module, used to visualize the clustering results of multiple cells; and / or,

[0158] An annotation result visualization module is used to visualize the annotation results of multiple cell clusters.

[0159] Optionally, based on any of the above devices, the target disease is achalasia.

[0160] The single-cell transcriptome sequencing data annotation device provided by the embodiment of the present invention obtains the single-cell transcriptome sequencing data of the target object for the target object suffering from the target disease through the cell clustering module, and clusters the multiple cells corresponding to the single-cell transcriptome sequencing data based on the single-cell transcriptome sequencing data to obtain multiple cell clusters, so as to perform cell type annotation for the multiple cell clusters respectively; through the cell cluster gene expression amount obtaining module, for each marker gene corresponding to the target disease, according to the single-cell transcriptome sequencing data, the cell cluster gene expression amount of the marker gene in each cell cluster is obtained, the cell cluster gene expression amount corresponds to the marker gene, and the marker gene corresponds to the target disease, so that The annotation result obtained by the subsequent cell type annotation based on the gene expression of the cell cluster is an annotation result closely related to the target disease; through the total gene expression acquisition module, for each cell cluster and each cell type corresponding to the target disease, the cell cluster gene expression of each marker gene corresponding to the cell type in all marker genes in the cell cluster is accumulated to obtain the total gene expression, which can characterize the gene expression of a certain cell type under a certain cell cluster; thus, through the cell type annotation module, for each cell cluster, the target cell type can be annotated for the cell cluster from all cell types according to the total gene expression corresponding to each cell type of the cell cluster. Compared with the annotation of cell types by applying universal marker genes, the above device, by applying marker genes that are fully compatible with the target disease, performs the annotation process of specific cell types related to the target disease, thereby ensuring the accuracy of cell type annotation.

[0161] The single-cell transcriptome sequencing data annotation device provided in the embodiment of the present invention can execute the single-cell transcriptome sequencing data annotation method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0162] It is worth noting that in the embodiment of the above-mentioned single-cell transcriptome sequencing data annotation device, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0163] Figure 8 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0164] like Figure 8 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0165] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0166] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The processor 11 executes the various methods and processes described above, such as a single-cell transcriptome sequencing data annotation method.

[0167] In some embodiments, the single-cell transcriptome sequencing data annotation method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the single-cell transcriptome sequencing data annotation method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform the single-cell transcriptome sequencing data annotation method by any other appropriate means (e.g., by means of firmware).

[0168] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0169] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine as a stand-alone software package and partially on a remote machine, or entirely on a remote machine or server.

[0170] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0171] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0172] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0173] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0174] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0175] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A single-cell transcriptome sequencing data annotation method, characterized in that: include: For a target subject suffering from a target disease, single-cell transcriptome sequencing data of the target subject is obtained, and based on the single-cell transcriptome sequencing data, a plurality of cells corresponding to the single-cell transcriptome sequencing data are clustered to obtain a plurality of cell clusters; For each marker gene corresponding to the target disease, according to the single-cell transcriptome sequencing data, the cell cluster gene expression amount of the marker gene in each of the cell clusters is obtained; For each of the cell clusters and each cell type corresponding to the target disease, the cell cluster gene expression levels of each of the marker genes corresponding to the cell type in all of the marker genes are accumulated in the cell clusters to obtain a total gene expression level; For each of the cell clusters, according to the total gene expression amounts corresponding to the cell cluster under each of the cell types, a target cell type is annotated for the cell cluster from all the cell types.

2. The method according to claim 1, characterized in that: The step of annotating a target cell type for the cell cluster from all the cell types according to the total gene expression amount corresponding to each cell type of the cell cluster comprises: Determine the maximum gene expression amount with the largest value from the total gene expression amounts corresponding to each cell type of the cell cluster; The cell type corresponding to the maximum gene expression level among all the cell types is annotated as the target cell type of the cell cluster.

3. The method according to claim 1, characterized in that The method of clustering a plurality of cells corresponding to the single-cell transcriptome sequencing data based on the single-cell transcriptome sequencing data to obtain a plurality of cell clusters comprises: For all genes corresponding to the single-cell transcriptome sequencing data, determining a plurality of target genes from all the genes according to the single-cell transcriptome sequencing data; For each of the target genes and the multiple cells corresponding to the single-cell transcriptome sequencing data, scaling processing is performed on the cellular gene expression amounts of the target genes in the multiple cells, so that the cellular gene expression amounts of different target genes fall within a preset range; The single-cell transcriptome sequencing data is updated according to the obtained scaling processing result, and the multiple cells are clustered according to the single-cell transcriptome sequencing data to obtain multiple cell clusters.

4. The method according to claim 3, characterized in that Determining a plurality of target genes from all the genes according to the single-cell transcriptome sequencing data comprises: For each of the genes, according to the cell gene expression levels of the gene in the multiple cells respectively, obtaining the expression difference coefficient of the gene between the multiple cells; According to the expression difference coefficients respectively corresponding to all the genes, a plurality of target genes are determined from all the genes.

5. The method according to claim 3, characterized in that: Before determining a plurality of target genes from all the genes according to the single-cell transcriptome sequencing data, the method further includes: The single-cell transcriptome sequencing data is standardized, and the single-cell transcriptome sequencing data is updated according to the obtained standardized processing result.

6. The method according to claim 5, characterized in that Before the single-cell transcriptome sequencing data is standardized, the method further includes: Obtaining the sequencing quality of each of the cells according to the single-cell transcriptome sequencing data, and filtering the multiple cells according to all the obtained sequencing qualities; According to the obtained filtering processing results, the single-cell transcriptome sequencing data is updated.

7. The method according to claim 3, characterized in that After updating the single-cell transcriptome sequencing data according to the obtained scaling processing result, the method further includes: Perform dimensionality reduction processing on the single-cell transcriptome sequencing data, and update the single-cell transcriptome sequencing data according to the obtained dimensionality reduction processing result.

8. The method according to claim 1, characterized in that: Also includes: For each of the cell types, the annotation result of the cell cluster that uses the cell type as the target cell type is verified based on the cell gene expression level of each of the marker genes corresponding to the cell type in all the marker genes in each of the cells.

9. The method according to claim 1, characterized in that: Also includes: Visualizing the clustering results of the plurality of cells; and / or, The annotation results of the plurality of cell clusters are visualized.

10. The method according to any one of claims 1 to 9, characterized in that The target disease is achalasia.

Images