Cell communication analysis method, device, computer equipment and storage medium

Abstract

The application relates to a cell communication analysis method and device, computer equipment, a storage medium and a computer program product. The method comprises the following steps: acquiring group-in-cell communication information of each sample in a plurality of samples containing cells; wherein each sample in the plurality of samples is collected at different stages under the same biological condition; the group-in-cell communication information is used for representing a cell communication mode between different cell types in the sample; the group-in-cell communication information of the plurality of samples is combined to obtain multi-sample communication information; and feature extraction is performed on the multi-sample communication information to obtain group-between-cell communication information of the plurality of samples; wherein the group-between-cell communication information is used for representing a cell communication mode between the plurality of samples. The method can reveal the continuity change of the cell communication mode between different samples.

Description

Technical Field

[0001] This application relates to the field of bioinformatics, and in particular to a cell communication analysis method, apparatus, computer equipment, storage medium, and computer program product. Background Technology

[0002] Intercellular communication is a dynamic biological process involving the transmission of information by cells in various ways, and it is crucial for understanding tissue function and disease mechanisms. With the development of biomedicine, single-cell sequencing technology has emerged, enabling the revelation of cellular heterogeneity and complex intercellular interactions at the single-cell level. Among these technologies, single-cell transcriptome data provides strong data support for understanding intercellular communication.

[0003] In existing technologies, researchers typically input single-cell transcriptome data into single-cell communication analysis tools for processing to obtain corresponding intercellular communication information. However, existing cell communication analysis tools mainly focus on analyzing intercellular communication information within a single group and cannot perform comprehensive analysis across different time points or processing conditions, thus failing to reveal the continuous changes in cell communication patterns over time. Summary of the Invention

[0004] Therefore, it is necessary to provide a cell communication analysis method, apparatus, computer device, computer-readable storage medium, and computer program product to address the aforementioned technical problems.

[0005] In a first aspect, this application provides a cell communication analysis method. The method includes:

[0006] Intra-group cell communication information is obtained for each of multiple groups of cell-containing samples; wherein each of the multiple groups of samples is collected at different stages under the same biological conditions; the intra-group cell communication information is used to characterize the cell communication patterns between different cell types within the sample.

[0007] The intragroup cell communication information of multiple groups of the samples is merged to obtain multi-sample communication information; wherein, the multi-sample communication information is used to characterize the cell communication patterns between different cell types in different samples;

[0008] Feature extraction is performed on the multi-sample communication information to obtain the inter-group cell communication information of the multi-sample group; wherein, the inter-group cell communication information is used to characterize the cell communication pattern among the multi-sample group.

[0009] In one embodiment, the step of extracting features from the multi-sample communication information to obtain the inter-group cell communication pattern of each group of samples in the multiple groups includes:

[0010] The multi-sample communication information is decomposed using a nonnegative matrix factorization algorithm to obtain a first feature matrix and a second feature matrix. The first feature matrix is ​​used to characterize the cell communication patterns among samples, and the second feature matrix is ​​used to characterize the interaction strength between cell types under each cell communication pattern.

[0011] In one embodiment, the method further includes:

[0012] Based on the first feature matrix, a first feature image is constructed;

[0013] A second feature image is constructed based on the second feature matrix; wherein, layers of different colors in the first feature image and the second feature image represent different interaction intensities.

[0014] In one embodiment, the step of establishing the second feature image based on the second feature matrix includes:

[0015] Based on the second feature matrix, an initial second feature image is established;

[0016] Cluster analysis is performed on the second feature matrix to obtain clustering results; the clustering results are used to characterize the similarity of the interaction strength between cell types.

[0017] The clustering results are mapped to the initial second feature image to adjust the visualization layout of the initial second feature image.

[0018] In one embodiment, obtaining intragroup cell communication information for each of multiple groups of cell-containing samples includes:

[0019] Obtain single-cell transcriptome data for each sample in multiple sets of samples containing cells;

[0020] Clustering was performed on each group of single-cell transcriptome data to obtain the corresponding cell population;

[0021] The cell type of the cell population is determined based on the marker genes;

[0022] Based on the single-cell transcriptome data and cell type of each group, cell communication analysis was performed on the samples in each group to obtain intra-group cell communication information.

[0023] In one embodiment, the cell communication information includes at least: the strength and number of interactions between cell types, signaling pathway activity, or the interaction strength of receptor pairs.

[0024] Secondly, this application also provides a cell communication analysis device. The device includes:

[0025] The intragroup information acquisition module is used to acquire intragroup cell communication information for each of the multiple groups of cell-containing samples; wherein, each of the multiple groups of samples is collected at different stages under the same biological conditions; the intragroup cell communication information is used to characterize the cell communication patterns between different cell types within the sample.

[0026] The communication information merging module is used to merge the intra-group cell communication information of multiple groups of the samples to obtain multi-sample communication information; wherein, the multi-sample communication information is used to characterize the cell communication patterns between different cell types in different samples.

[0027] The feature extraction module is used to extract features from the multi-sample communication information to obtain the inter-group cell communication information of the multi-sample group; wherein, the inter-group cell communication information is used to characterize the cell communication pattern between the multi-sample group.

[0028] In one embodiment, the feature extraction module is further configured to:

[0029] The multi-sample communication information is decomposed using a nonnegative matrix factorization algorithm to obtain a first feature matrix and a second feature matrix. The first feature matrix is ​​used to characterize the cell communication patterns among samples, and the second feature matrix is ​​used to characterize the interaction strength between cell types under each cell communication pattern.

[0030] In one embodiment, the device further includes:

[0031] An image rendering module is used to create a first feature image based on the first feature matrix;

[0032] The image drawing module is further configured to establish a second feature image based on the second feature matrix; wherein, layers of different colors in the first feature image and the second feature image represent different interaction intensities.

[0033] In one embodiment, the image rendering module is further configured to:

[0034] Based on the second feature matrix, an initial second feature image is established;

[0035] Cluster analysis is performed on the second feature matrix to obtain clustering results; the clustering results are used to characterize the similarity of the interaction strength between cell types.

[0036] The clustering results are mapped to the initial second feature image to adjust the visualization layout of the initial second feature image.

[0037] In one embodiment, the initial information acquisition module includes:

[0038] The data acquisition submodule is used to acquire single-cell transcriptome data for each sample in multiple groups of samples containing cells;

[0039] The cell clustering submodule is used to cluster each group of single-cell transcriptome data to obtain the corresponding cell population;

[0040] The cell type acquisition submodule is used to determine the cell type of the cell population based on the identifier gene;

[0041] The communication information acquisition submodule is used to perform cell communication analysis on the samples in each group based on the single-cell transcriptome data and cell type, and obtain the intra-group cell communication information of each group.

[0042] In one embodiment, the cell communication information includes at least: the strength and number of interactions between cell types, signaling pathway activity, or the interaction strength of receptor pairs.

[0043] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the cell communication analysis method as described in any one of the embodiments of this disclosure.

[0044] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the cell communication analysis method as described in any one of the embodiments of this disclosure.

[0045] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the cell communication analysis method as described in any one of the embodiments of this disclosure.

[0046] The aforementioned cell communication analysis methods, devices, computer equipment, storage media, and computer program products acquire intra-group cell communication information from multiple groups of cell-containing samples, merge the intra-group cell communication information of each group to obtain multi-sample communication information, and extract features from the multi-sample communication information to obtain the inter-group cell communication pattern of each group. Multi-group sample communication information can intuitively describe the cell communication patterns of different cell communication characteristics in different groups of samples, ensuring the comparability of cell communication patterns between different groups of samples and helping to eliminate outlier data, thereby improving the accuracy of inter-group cell communication information. By extracting features from this multi-sample communication information, it is ensured that the inter-group cell communication information is obtained based on a comprehensive analysis of multi-sample cell communication patterns, revealing the continuous changes in cell communication patterns between different samples. Attached Figure Description

[0047] Figure 1 This is a flowchart illustrating a cell communication analysis method in one embodiment;

[0048] Figure 2 This is a schematic diagram illustrating the merging of intragroup cell communication information in one embodiment.

[0049] Figure 3 This is a schematic diagram of the process for creating a feature image in one embodiment;

[0050] Figure 4 This is a first schematic diagram of a first feature image in one embodiment;

[0051] Figure 5 This is a flowchart illustrating the process of creating a second feature image in one embodiment;

[0052] Figure 6 This is a first schematic diagram of the second feature image in one embodiment;

[0053] Figure 7 This is a second schematic diagram of the first feature image in one embodiment;

[0054] Figure 8 This is a second schematic diagram of the second feature image in one embodiment;

[0055] Figure 9 This is a schematic diagram of the process for obtaining intra-group cell communication information in one embodiment;

[0056] Figure 10 This is a flowchart illustrating the implementation of a cell communication analysis method in one embodiment;

[0057] Figure 11 This is a structural block diagram of a cell communication analysis device in one embodiment;

[0058] Figure 12 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0059] To make the objectives, technical solutions, 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.

[0060] In one embodiment, such as Figure 1 As shown, a cell communication analysis method is provided. This embodiment illustrates the application of this method to a terminal. It is understood that this method can also be applied to a server, and further to a system including both a terminal and a server, and is implemented through interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0061] Step S100: Obtain intragroup cell communication information for each of the multiple groups of cell-containing samples; wherein, each of the multiple groups of samples was collected at different stages under the same biological conditions; the intragroup cell communication information is used to characterize the cell communication patterns between different cell types within the sample.

[0062] In one exemplary embodiment, the sample may include a combination of cells with a majority of cells, such as a single lung cell; wherein the cells may include B cells, T cells, etc.

[0063] In one exemplary embodiment, multiple sets of samples can be obtained by collecting an initial sample at different stages under the same biological conditions. The biological conditions may include disease states, drug treatments, or developmental stages; the different stages may include the duration of culture under different environments (e.g., hypoxia, low pressure, low temperature, high temperature). For example, multiple sets of samples may be cultured under hypoxic conditions for 0 days, 3 days, 7 days, 14 days, 21 days, etc. The initial sample may include, but is not limited to, single-cell samples or tissue samples; for example, analyzing mouse lung single-cell samples to obtain communication information between cell types within the mouse lung single-cell sample.

[0064] In one exemplary embodiment, the intragroup cell communication information may include the interaction strength and number of different cell types in a single group of samples, signaling pathway activity, or interaction strength of receptor-ligand pairs; wherein, the number of interactions between different cell types can be obtained by counting the number of interactions between different cell types that are greater than a preset threshold, which is not limited here.

[0065] In one exemplary embodiment, the acquisition of intragroup cell communication information may include obtaining the gene expression matrix and cell type of a single-cell sample, and inputting the gene expression matrix and cell type into a cell communication analysis tool to obtain intragroup cell communication information. In another exemplary embodiment, the acquisition of intragroup cell communication information may also include single-cell RNA sequencing, intercellular communication network analysis, flow cytometry, co-culture experiments, spatial transcriptomics, etc.

[0066] Step S200: Merge intragroup cell communication information of multiple groups of samples to obtain multi-sample communication information; wherein, the multi-sample communication information is used to characterize the cell communication patterns between cell types in different samples.

[0067] In one exemplary embodiment, merging the intragroup cell communication information of multiple groups of samples may include transforming the intragroup cell communication information of each group of samples into one-dimensional data (each data in the matrix represents the interaction strength of the corresponding cell type interaction pair within the sample, etc.), and merging the transformed one-dimensional data of each group of samples to obtain multi-sample communication information. Specifically, taking the intragroup cell communication data as the interaction strength between different cell types as an example, it can be done as follows: Figure 2 As shown, the three tables on the left represent intragroup cell communication data for samples from groups A, B, and C, respectively. In the intragroup cell communication data, 'a', 'b', and 'c' represent different cell types; the corresponding values ​​are the interaction strength between cell types. When merging the three tables on the left, since they belong to different groups but the sample groups are the same, they are merged into the multi-sample communication information on the right. The row coordinates represent cell type interaction pairs that have undergone cell communication, the column coordinates represent samples from different groups, and the values ​​represent the interaction strength of the cell type interaction pairs within the corresponding samples. Cells can include Endothelial, Pericyte, B, Fibroblast, T cells, etc., and cell type interaction pairs can specifically include "AT1_SMC", "B_T", "DC_B", etc.

[0068] In another exemplary embodiment, the intra-group cell communication data may also include the activity of different signaling pathways, and the combined multi-sample communication information can represent different signaling pathways, such as EPHB, KIT, PTPRM, PECAM1, CCL, CDH5 and other signaling pathways.

[0069] In another exemplary embodiment, the intra-group cell communication data may also include the interaction strength of ligands and receptors of different cell types, and the established multi-sample communication information can represent different ligand and receptor pairs, such as "LR1_LR2", etc.

[0070] Step S300: Feature extraction is performed on the multi-sample communication information to obtain inter-group cell communication information in the multi-group samples; wherein, the inter-group cell communication information is used to characterize the cell communication pattern between the multi-group samples.

[0071] In one exemplary embodiment, the inter-group cell communication information may include cell communication patterns among multiple groups of samples. It is understood that the multiple groups of samples may include initial samples collected at different stages under the same biological conditions. The cell communication patterns among the multiple groups of samples can be used to explain the continuous changes in cell communication patterns at different stages, i.e., the continuous changes in the cell communication patterns of the initial samples over time. Specifically, the feature extraction can obtain the interaction strength matrix of different samples in different cell communication patterns and the interaction strength matrix of different cell type interaction pairs in different cell communication patterns. It is understood that the cell communication patterns among multiple groups of samples can be determined through the interaction strength matrix of different samples in different cell communication patterns; and the interaction strength matrix of different cell type interaction pairs in different cell communication patterns can reveal which cell types interact more actively under specific cell communication patterns.

[0072] In one exemplary embodiment, the feature extraction can use nonnegative matrix factorization, principal component analysis, nonnegative factorization, etc., and feature extraction algorithms can be used to extract features from multi-sample communication information to obtain inter-group cell communication information, etc.

[0073] In the aforementioned cell communication analysis method, intra-group cell communication information is obtained from multiple groups of samples containing cells. This intra-group cell communication information is then merged to obtain multi-sample communication information. Feature extraction is performed on this multi-sample communication information to obtain the inter-group cell communication pattern for each sample group. This multi-sample communication information can intuitively describe the cell communication patterns of different cell communication characteristics across different sample groups, ensuring the comparability of cell communication patterns between different groups and helping to eliminate outlier data, thereby improving the accuracy of inter-group cell communication information. By extracting features from this multi-sample communication information, it is ensured that the inter-group cell communication information is obtained based on a comprehensive analysis of multi-sample cell communication patterns, revealing the continuous changes in cell communication patterns between different samples.

[0074] In one embodiment, the step of extracting features from the multi-sample communication information to obtain the inter-group cell communication pattern of each group of samples in the multi-sample set includes:

[0075] The multi-sample communication information is decomposed using a nonnegative matrix factorization algorithm to obtain a first feature matrix and a second feature matrix. The first feature matrix is ​​used to characterize the cell communication patterns among samples, and the second feature matrix is ​​used to characterize the interaction strength between cell types under each cell communication pattern.

[0076] In one exemplary embodiment, a nonnegative matrix factorization algorithm is used to decompose the multi-sample communication information into two nonnegative matrices. The product of these two nonnegative matrices can represent the multi-sample communication information, etc. It is understood that the two nonnegative matrices represent the basis matrix (feature matrix) and the coefficient matrix (weight matrix), respectively. The goal of the nonnegative matrix factorization algorithm can include finding the two optimal matrices such that their product is as close as possible to the multi-sample communication information, while ensuring the nonnegativity of the two matrices. In this way, the resulting structure is more sparse and interpretable. Specifically, the nonnegative matrix factorization algorithm can be represented by V(m×n)≈W(m×r)×H(r×n), where V represents multi-sample communication information; W represents the basis matrix; H represents the coefficient matrix; m represents the number of samples; n represents the number of cell communication features, i.e., the number of cell type interaction pairs; and r represents the number of cell communication feature patterns, which is usually much smaller than m and n. By performing nonnegative matrix factorization on the multi-sample communication information V, it is decomposed into the basis matrix W and the coefficient matrix H. The basis matrix W is the first feature matrix, and the coefficient matrix H is the second feature matrix, etc.

[0077] In one exemplary embodiment, the row identifier information of the first feature matrix can be used to characterize different cell communication feature patterns; the column identifier information of the first feature matrix can be used to characterize samples from different groups; the row identifier information of the second feature matrix can be used to characterize different cell communication feature patterns, and the column identifier information of the second feature matrix can be used to characterize different cell type interaction pairs.

[0078] In this embodiment, nonnegative matrix factorization is used to decompose the multi-sample communication information into a first feature matrix and a second feature matrix. Nonnegative matrix factorization improves the interpretability of the decomposition results, revealing cell communication patterns among multiple samples and significantly enhancing the ability to identify these patterns. Furthermore, it can capture previously difficult-to-discover cell communication information, such as pathways, intercellular communication strength, and receptor-ligand pair interaction strength.

[0079] In one embodiment, such as Figure 3 As shown, the method further includes:

[0080] Step S401: Establish a first feature image based on the first feature matrix.

[0081] Step S402: Based on the second feature matrix, establish a second feature image; wherein, in the first feature image and the second feature image, layers of different colors represent different interaction intensities.

[0082] In one exemplary embodiment, the first feature image may be as follows: Figure 4As shown, the horizontal axis labels (Pattern1, Pattern2, Pattern3) represent different cell communication patterns, and the column axis labels (0d, 3d, 7d, 14d, 21d, 28d) represent different times when the samples in each group were cultured under the same biological environment, such as 0 days, 3 days, 7 days, 14 days, 21 days and 28 days respectively.

[0083] In this embodiment, a first feature image and a second feature image are obtained by plotting the first feature matrix and the second feature matrix. This enables the visualization of the continuous changes in cell communication patterns between samples over time, allowing for a faster and more accurate grasp of the overall trend and local details of cell communication patterns.

[0084] In one embodiment, such as Figure 5 As shown, based on the second feature matrix, a second feature image is constructed, including:

[0085] Step S411: Establish an initial second feature image based on the second feature matrix.

[0086] Step S412: Perform cluster analysis on the second feature matrix to obtain clustering results; the clustering results are used to characterize the similarity of the interaction strength between cell types.

[0087] In one exemplary embodiment, cluster analysis of the second feature matrix can be performed using Euclidean distance to calculate the similarity of interaction strengths between cell types, and further clustering can be performed on the second feature matrix. Specifically, the interaction strength of each cell type interaction pair in the second feature matrix under different cell communication modes can be input sequentially, and the similarity of interaction strengths between cell types can be calculated using Euclidean distance. In another exemplary embodiment, similarity algorithms such as cosine similarity, Manhattan distance, Gerard similarity coefficient, and Pearson correlation coefficient can also be used to calculate the similarity of the second feature matrix.

[0088] Step S413: Map the clustering results to the initial second feature image to adjust the visualization layout of the initial second feature image.

[0089] In one exemplary embodiment, mapping the clustering results to an initial second feature image to adjust the visualization layout of the initial second feature image may include adjusting the order of cell type interaction pairs in the initial second feature image using the clustering results, moving cell type interaction pairs clustered into the same category together. It is understood that moving cell type interaction pairs with high similarity together can more intuitively demonstrate the similarities and differences between different cell type interaction pairs. In another exemplary embodiment, clustering markers may be added to the initial second feature image to connect cell type interaction pairs of the same category. In actual use, the two cell type interaction pairs with the highest similarity may be connected sequentially, ultimately connecting all cell type interaction pairs together for display.

[0090] In one exemplary embodiment, cell type interaction pairs with a similarity greater than a preset threshold can be grouped together, that is, features of similar feature patterns can be grouped together, which can more intuitively reveal the similarity and differences between different cell type interaction pairs; in another exemplary embodiment, the similarity of the interaction strength of each cell type interaction pair can be sorted according to the similarity, and the cell type interaction pairs in the initial second feature image can be sorted accordingly according to the sorting to obtain the second feature matrix, etc.

[0091] In one exemplary embodiment, similarity calculations can also be performed on the first feature matrix to cluster similar samples together, revealing the similarities and differences between different samples. Specifically, clustering can be performed on the first feature matrix using the interaction strength of each sample in each cell communication mode as input.

[0092] In one exemplary embodiment, the second feature image can be as follows: Figure 6 As shown, different colored pixel blocks represent different interaction intensities. For example, different depths of color can be used to represent different interaction intensities, or different RGB colors can be designed to represent different interaction intensities. Figure 6 The horizontal axis (Pattern1, Pattern2, Pattern3, etc.) in the image can represent different cell communication patterns; the vertical axis (B_T, T_T, DC_B, etc.) in the image can represent different cell type interaction pairs; the corresponding layers of different depth colors can include the interaction strength between cell types under different cell communication patterns; Figure 6The lines on the left can represent the group relationships of various cell type interaction pairs (the group relationships can be obtained by clustering cell type interaction pairs based on similarity, etc.). In practical use, the first feature image and the second feature image can be combined to further analyze the cell communication patterns between cell types in each sample. The first feature image determines which samples have higher interaction strengths under which specific cell communication patterns, and the second feature image determines which cell type interaction pairs have higher interaction strengths under the corresponding cell communication patterns. In this way, it can be determined which cell type interaction pairs in which samples have unique cell communication patterns, etc. For example... Figure 4 In the pattern, "14d" is the darkest color, indicating the strongest interaction, which corresponds to... Figure 6 The darker colors of "Interstitial_Macro-T", "DC-T", and "TT" in "Pattern3" indicate that the interaction strength of these cell type interaction pairs reached its peak in the "14d" group samples, which also means that these cell type interaction pairs have unique cell communication patterns.

[0093] In one exemplary embodiment, when the cell communication information represents different signaling pathway activities, the first feature image can be as follows: Figure 7 As shown, the horizontal axis labels (Pattern1, Pattern2, Pattern3) represent different cell communication patterns, and the column coordinates (0d, 3d, 7d, 14d, 21d, 28d) represent the initial sample's culture time in the biological environment, specifically 0 days, 3 days, 7 days, 14 days, 21 days, and 28 days, respectively; the second feature image can be as follows... Figure 8 As shown, the horizontal axis (Pattern1, Pattern2, Pattern3, etc.) in the image can represent different cell communication patterns; the vertical axis (EPHB, KIT, PTPRM, etc.) in the image can represent different signaling pathways; images of different color depths can include the activity of different signaling pathways in different cell communication patterns; Figure 8 The lines on the left can represent the group relationships between various signaling pathways (these group relationships can be obtained by clustering based on the similarity of signaling pathway activity in different cell communication modes, etc.). In practical use, Figure 7 The activity of the "pattern 1" signaling pathway decreased in the "7d", "14d", and "21d" sample groups; while Figure 8The pathways with darker colors in "pattern 1" include "COLLAGEN" and "CXCL," indicating that the activity of these pathways gradually decreases at "7d," "14d," and "21d," but remains at a relatively high level, suggesting that these pathways have an important impact on cell function at these time points. Figure 7 In the text, "pattern2" is darker in "0d", corresponding to... Figure 8 The darker colored pathways include “OSM”, “CEACAM”, “IL2”, and “IL4”, indicating that these pathways have unique signaling patterns in the “0d” sample group.

[0094] In this embodiment, the similarity of interaction strength between cell types is obtained by calculating the similarity of the second feature matrix. Based on the similarity, the order of cell type pairs in the initial second feature image is adjusted. This allows cell type interaction pairs with higher similarity to cluster together, revealing the similarities and differences between different cell type pairs and making cell communication patterns clearer.

[0095] In one embodiment, such as Figure 9 As shown, intragroup cell communication information is obtained for each of multiple groups of cell-containing samples, including:

[0096] Step S101: Obtain single-cell transcriptome data for each sample in multiple groups of samples containing cells.

[0097] In one exemplary embodiment, the single-cell transcriptome data may include gene expression matrices, etc., and the single-cell transcriptome data can be obtained through RNA sequencing, microarray technology, querying public databases, etc.

[0098] In one exemplary embodiment, after obtaining single-cell transcriptome data, the data can be standardized and normalized so that the single-cell transcriptome data can be processed uniformly.

[0099] Step S102: Cluster the single-cell transcriptome data of each group to obtain the corresponding cell population.

[0100] In one exemplary embodiment, after obtaining single-cell transcriptome data, the single-cell transcriptome data can be dimensionality reduced. It is understood that dimensionality reduction can screen out cells with special significance, simplify data analysis, remove redundant information, and improve the efficiency and accuracy of subsequent processing.

[0101] Step S103: Determine the cell type of the cell population based on the identifier gene.

[0102] In one exemplary embodiment, the marker gene may include a marker gene, etc., and the cell type of the cell population is determined by the marker gene.

[0103] Step S104: Based on the single-cell transcriptome data and cell type of each group, perform cell communication analysis on the samples in each group to obtain intra-group cell communication information.

[0104] In one exemplary embodiment, single-cell transcriptome data and cell types of cell populations can be input into a cell communication analysis tool to obtain intra-group cell communication information, etc. Specifically, the cell communication analysis tool may include CellChat, CellphoneDB, etc.

[0105] In this embodiment, by acquiring single-cell transcriptome data corresponding to multiple groups of samples and clustering samples within the same group, cell populations can be effectively identified and segmented, thereby determining the cell type of each cell population based on marker genes. This not only helps to deepen the understanding of cell diversity and function but also enables the analysis of cell communication information within each group based on cell type, thereby revealing the interactions and regulatory mechanisms between cells in different groups.

[0106] In one embodiment, the cell communication information includes at least: the strength and number of interactions between cell types, signaling pathway activity, or the interaction strength of receptor-ligand pairs.

[0107] In one exemplary embodiment, the number of interactions between cell types may include the number of interactions between different cell type pairs where the interaction strength is greater than a preset threshold. It is understood that cell communication information may include intragroup cell communication information corresponding to each group of samples or intergroup cell communication information between multiple groups of samples. Specifically, intragroup cell communication information and intergroup cell communication information can be correlated; that is, intragroup cell communication information may include at least one of the above three types, and intergroup cell communication information is also the category of cell communication information corresponding to intragroup cell communication information. It is understood that the interaction strength and number between cell types, signaling pathway activity, or receptor-ligand pair interaction strength can all characterize the cell communication patterns between cell types within multiple groups of samples. Therefore, intragroup cell communication information may include only one type, and intergroup cell communication information can be obtained through this one type and used to characterize the cell communication patterns between multiple groups of samples.

[0108] In one exemplary embodiment, the interaction strength and quantity between the various cell types, the activity of signaling pathways, or the interaction strength of receptor-ligand pairs can all be obtained by analyzing each group of samples using cell communication analysis tools.

[0109] In this embodiment, the categories of cell communication information include at least one of the following: the strength and quantity of interactions between different cell types, the activity of signaling pathways, or the interaction strength of receptor-ligand pairs. This makes intragroup cell communication information more diverse, provides multiple pathways for obtaining intergroup cell communication information, and facilitates the acquisition of intergroup cell communication information.

[0110] In one exemplary embodiment, the cell communication analysis method can be as follows: Figure 10 The flowchart shown is implemented as follows:

[0111] Step S901: Acquire multi-time series data; specifically, this may include acquiring single-cell transcriptome data (including gene expression matrices) of samples from different groups treated under the same experimental conditions at different times; for example, single-cell samples of mouse lungs cultured for 0, 3, 7, 14, 21, and 28 days under low-pressure and low-oxygen conditions may be acquired; a total of 6 groups of samples. In actual use, the entire left lung of a mouse may be dissociated into a single-cell suspension, and single-cell transcriptome sequencing technology may be used to obtain sample data from different groups, etc.

[0112] Step S902, data preprocessing; specifically, this may include standardizing and normalizing multiple time series data to make the time series data comparable, etc. Specifically, six groups of samples can be integrated into one file, and the sample number ("0d", "3d", "7d", "14d", "21d", "28d") can be marked in the file to indicate which cells belong to which sample, etc., and low-quality cells, low-expressed genes, and mitochondrial genes, etc., can be filtered out; and the data in the processed file can be standardized and normalized, etc. Filtering low-quality cells, low-expressed genes, and mitochondrial genes can be done using the subset function, etc.; data standardization and normalization can be done using the NormalizeData() and ScaleData() functions, etc.

[0113] Step S903, dimensionality reduction and clustering; specifically, dimensionality reduction is used to screen cells and genes that meet preset conditions from multi-time series data; for example, screening genes in the gene expression matrix that are greater than a preset threshold; clustering can include multi-time series data, and single-cell transcriptome data can be clustered to obtain multiple cell populations; specifically, dimensionality reduction can be achieved using the RunPCA() function; and clustering can be performed using the FindNeighbors() function and the FindClusters() function;

[0114] Step S904, cell type annotation; specifically, cell types can be determined using known marker genes, and the cell type to which each cell population belongs is its annotation information, etc.

[0115] Step S905, cell communication analysis for a single group; specifically, this may include analyzing the sequence data of each group of samples to obtain the inter-group cell communication pattern corresponding to the sequence data of each group of samples; the cell communication results may include one of the following: the interaction strength and number between cell types, the activity of signaling pathways, or the interaction strength of receptor-ligand pairs; for example, each sequence data and annotation information may be input into a cell communication analysis tool (e.g., CellChat, CellphoneDB, etc.) to analyze and obtain the cell communication results for a single group, etc.

[0116] Step S906 integrates cell communication results from multiple groups. Specifically, this may include merging the cell communication results corresponding to the sequence data at each stage to obtain merged cell communication results from multiple groups. For example, if the cell communication result is signaling pathway activity, the integrated cell communication result may include returning a data frame containing the communication probability of the corresponding signaling pathway for each sample. This data frame contains three columns: pathway name, communication probability, and sample number. The data frame can be converted into a matrix where the row names are sample numbers, the column names are pathway names, and the values ​​are communication probabilities. This yields a matrix integrating the communication probabilities of signaling pathways from six samples. If the signaling communication result is the interaction strength between cell types for each sample, the row names in the integrated matrix can be sample numbers, and the column names can be cell type interaction pairs, such as "celltype1-celltype2". For example, "AT1_SMC" represents the signal sent by cell type AT1 to cell type SMC, and the matrix value is the interaction strength (communication probability). This yields a matrix integrating the interaction strength between cell types from six samples.

[0117] Step S907, Non-negative matrix decomposition; specifically, a non-negative matrix decomposition algorithm can be used to analyze the merged multi-group cell communication results, decomposing them into two non-negative matrices, and the product of the two matrices should be as close as possible to the merged multi-group cell communication results, etc.; it can be understood that the two matrices are the base matrix (feature matrix) and the coefficient matrix (weight matrix), respectively; wherein, the row identifier information of the base matrix can include different cell communication patterns; the column identifier information of the base matrix can include single-cell samples from different groups; the row identifier information of the coefficient matrix can include different cell communication patterns; the column identifier information of the coefficient matrix can include cell type interaction pairs in which cell communication occurs within the group; the data of the two matrices can be respectively represented as the cell communication patterns between each sample (when each sample is collected from different stages under the same biological conditions, it can reflect the continuous change of the cell communication patterns of the samples over time) and which cell types interact more actively under different cell communication patterns, etc.

[0118] Step S908, feature pattern visualization; specifically, this may include plotting the basis matrix and coefficient matrix using heatmap visualization technology. It is understood that different shades of color can be used to represent different interaction intensities. In practical applications, clustering can be performed on the interaction intensities of different samples in the basis matrix under different cell communication modes, or on the interaction intensities of different cell type interaction pairs in the coefficient matrix under different cell communication modes. This clustering allows for a clearer observation of the relationships between different samples; clustering different cell type interaction pairs allows highly similar cell type interaction pairs to be grouped together, making the feature patterns clearer.

[0119] In one exemplary embodiment, the cell communication analysis method can be used to analyze the continuous changes in cell communication patterns over time in mouse lung single-cell samples under low-pressure and low-oxygen conditions. Specific steps may include:

[0120] Sample preparation: Single-cell samples of mouse lungs were obtained after being cultured under low-pressure and low-oxygen conditions for 0, 3, 7, 14, 21, and 28 days, for a total of 6 groups of samples; Specifically, the entire left lung of a mouse was dissociated into single-cell suspensions, and sample data of different groups were obtained by single-cell transcriptome sequencing technology.

[0121] Data integration and preprocessing, along with cell type annotation, are performed. Six groups of single-cell samples are integrated into a single file, labeled with sample numbers ("0d", "3d", "7d", "14d", "21d", "28d"), indicating which cells belong to which sample. Low-quality cells, low-expression genes, and mitochondrial genes are filtered (using functions like Seurat). The filtered data is then standardized (using functions like NormalizeData) and normalized (using functions like ScaleData). The standardized and normalized data is then clustered (using functions like RunPCA) and clustered (using functions like FindNeighbors and FindClusters). Finally, the data is annotated based on the marker genes for each cluster to obtain cell type annotation information.

[0122] To obtain cell communication analysis results for a single sample, the gene expression matrix and cell type annotation information of each sample can be input into the cell communication analysis tool to obtain the results. Specifically, the communication network for each sample can be constructed using the cell communication analysis tool. For example, a cellchat object can be constructed using cellchat (specifically, the createCellChat function can be used); the gene expression matrix is ​​input, and the corresponding annotated files for the sample are extracted from the file. The standardized gene expression matrix is ​​stored in the system, the cell type annotation information is input as the parameter "meta", and the cell communication network database is set up. Overexpressed ligands or receptors in the cell group and their interactions are identified. The communication probability is further calculated and the communication network is inferred. By performing the above calculation process on the six groups of samples, the cell communication network results for each group can be obtained.

[0123] This process integrates cell communication results from multiple samples; it merges the cell communication network results of six samples into a single object containing multiple datasets. Specifically, it can include merging the six sample CellChat objects into a single object containing multiple datasets using the "mergeCellChat" function. The result returns a data frame containing the communication probability of the corresponding signaling pathway for each sample. This data frame includes the pathway name, communication probability, and sample number information. This data frame is then converted into a matrix with rows named as sample numbers, columns named as pathway names, and values ​​as communication probabilities, thus obtaining a matrix integrating the communication probabilities of the signaling pathways from the six samples. In another instance, the cell communication network results can be the interaction strength between cell types for each sample. The integrated matrix has rows named as sample numbers and columns named as celltype1-celltype2, for example, "AT1_SMC" represents the signal sent from cell type AT1 to cell type SMC, and the matrix values ​​are the interaction strengths (communication probabilities). This yields a matrix integrating the interaction strengths between cell types from the six samples.

[0124] Non-negative matrix factorization (NMF) identifies cell communication feature patterns. The optimal rank of the NMF model can be obtained first (using functions like NMF:nmfEstimateRank). The input parameter is the integrated cell communication results from multiple samples. The parameter rank can be set to the optimal rank. The integrated cell communication results from multiple samples are decomposed using a non-negative matrix to obtain two matrices: a basis matrix (or feature matrix) and a coefficient matrix (or weight matrix).

[0125] Heatmap visualization: The base matrix and coefficient matrix are visualized, specifically through the "Heatmap" function. Hierarchical clustering is performed on the heatmap to group features or samples with similar patterns together. Based on the different color blocks in the heatmap, the activity of feature patterns can be intuitively displayed.

[0126] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0127] Based on the same inventive concept, this application also provides a cell communication analysis device for implementing the cell communication analysis method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more embodiments of the cell communication analysis device provided below can be found in the limitations of the cell communication analysis method described above, and will not be repeated here.

[0128] In one embodiment, such as Figure 11 As shown, a cell communication analysis device 100 is provided, including: an intragroup information acquisition module 101, a communication information merging module 102, and a feature extraction module 103, wherein:

[0129] The intragroup information acquisition module is used to acquire intragroup cell communication information for each of the multiple groups of cell-containing samples; wherein, each of the multiple groups of samples is collected at different stages under the same biological conditions; the intragroup cell communication information is used to characterize the cell communication patterns between different cell types within the sample.

[0130] The communication information merging module is used to merge the intra-group cell communication information of multiple groups of the samples to obtain multi-sample communication information; wherein, the multi-sample communication information is used to characterize the cell communication patterns between different cell types in different samples.

[0131] The feature extraction module is used to extract features from the multi-sample communication information to obtain the inter-group cell communication information of the multi-sample group; wherein, the inter-group cell communication information is used to characterize the cell communication pattern between the multi-sample group.

[0132] In one embodiment, the feature extraction module is further configured to:

[0133] The multi-sample communication information is decomposed using a nonnegative matrix factorization algorithm to obtain a first feature matrix and a second feature matrix. The first feature matrix is ​​used to characterize the cell communication patterns among samples, and the second feature matrix is ​​used to characterize the interaction strength between cell types under each cell communication pattern.

[0134] In one embodiment, the device further includes:

[0135] An image rendering module is used to create a first feature image based on the first feature matrix;

[0136] The image drawing module is further configured to establish a second feature image based on the second feature matrix; wherein, layers of different colors in the first feature image and the second feature image represent different interaction intensities.

[0137] In one embodiment, the image rendering module is further configured to:

[0138] Based on the second feature matrix, an initial second feature image is established;

[0139] Cluster analysis is performed on the second feature matrix to obtain clustering results; the clustering results are used to characterize the similarity of the interaction strength between cell types.

[0140] The clustering results are mapped to the initial second feature image to adjust the visualization layout of the initial second feature image.

[0141] In one embodiment, the initial information acquisition module includes:

[0142] The data acquisition submodule is used to acquire single-cell transcriptome data for each sample in multiple groups of samples containing cells;

[0143] The cell clustering submodule is used to cluster each group of single-cell transcriptome data to obtain the corresponding cell population;

[0144] The cell type acquisition submodule is used to determine the cell type of the cell population based on the identifier gene;

[0145] The communication information acquisition submodule is used to perform cell communication analysis on the samples in each group based on the single-cell transcriptome data and cell type, and obtain the intra-group cell communication information of each group.

[0146] In one embodiment, the cell communication information includes at least: the strength and number of interactions between cell types, signaling pathway activity, or the interaction strength of receptor pairs.

[0147] Each module in the aforementioned cell communication analysis device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0148] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 12 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores cell communication data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When executed by the processor, the computer program implements a cell communication analysis method.

[0149] Those skilled in the art will understand that Figure 12 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0150] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0151] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0152] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0153] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A cell communication analysis method, characterized in that, The method includes: Intra-group cell communication information is obtained for each of multiple groups of cell-containing samples; wherein each of the multiple groups of samples is collected at different stages under the same biological conditions; the intra-group cell communication information is used to characterize the cell communication patterns between different cell types within the sample. The intragroup cell communication information of multiple groups of the samples is merged to obtain multi-sample communication information; wherein, the multi-sample communication information is used to characterize the cell communication patterns between different cell types in different samples; Feature extraction is performed on the multi-sample communication information to obtain the inter-group cell communication information of the multi-sample group; wherein, the inter-group cell communication information is used to characterize the cell communication pattern among the multi-sample group.

2. The method according to claim 1, characterized in that, The step of extracting features from the multi-sample communication information to obtain inter-group cell communication information for each group of samples in the multiple groups includes: The multi-sample communication information is decomposed using a nonnegative matrix factorization algorithm to obtain a first feature matrix and a second feature matrix. The first feature matrix is ​​used to characterize the cell communication patterns among samples, and the second feature matrix is ​​used to characterize the interaction strength between cell types under each cell communication pattern.

3. The method according to claim 2, characterized in that, The method further includes: Based on the first feature matrix, a first feature image is constructed; A second feature image is constructed based on the second feature matrix; wherein, layers of different colors in the first feature image and the second feature image represent different interaction intensities.

4. The method according to claim 3, characterized in that, The step of establishing the second feature image based on the second feature matrix includes: Based on the second feature matrix, an initial second feature image is established; Cluster analysis is performed on the second feature matrix to obtain clustering results; the clustering results are used to characterize the similarity of the interaction strength between cell types. The clustering results are mapped to the initial second feature image to adjust the visualization layout of the initial second feature image.

5. The method according to claim 1, characterized in that, The acquisition of intragroup cell communication information for each of the multiple groups of cell-containing samples includes: Obtain single-cell transcriptome data for each sample in multiple sets of samples containing cells; Clustering was performed on each group of single-cell transcriptome data to obtain the corresponding cell population; The cell type of the cell population is determined based on the marker genes; Based on the single-cell transcriptome data and cell type of each group, cell communication analysis was performed on the samples in each group to obtain intra-group cell communication information.

6. The method according to any one of claims 1-5, characterized in that, The cell communication information includes at least: the strength and number of interactions between different cell types, the activity of signaling pathways, or the interaction strength of receptor pairs.

7. A cell communication analysis device, characterized in that, The device includes: The intragroup information acquisition module is used to acquire intragroup cell communication information for each of the multiple groups of cell-containing samples; wherein, each of the multiple groups of samples is collected at different stages under the same biological conditions; the intragroup cell communication information is used to characterize the cell communication patterns between different cell types within the sample. The communication information merging module is used to merge the intra-group cell communication information of multiple groups of the samples to obtain multi-sample communication information; wherein, the multi-sample communication information is used to characterize the cell communication patterns between different cell types in different samples. The feature extraction module is used to extract features from the multi-sample communication information to obtain the inter-group cell communication information of the multi-sample group; wherein, the inter-group cell communication information is used to characterize the cell communication pattern between the multi-sample group.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images