Patient layering method based on breast cancer single cell space proteome multi-cell enrichment mode

Abstract

The invention belongs to the technical field of breast cancer treatment, and particularly relates to a patient layering method based on a breast cancer single cell space proteome multicellular enrichment mode, which comprises the following steps: S1, breast cancer single cell space proteome image data cell segmentation and cell type identification; s2, constructing a cell network diagram, and simulating a tumor microenvironment; s3, detecting closely connected cell communities, and extracting cell community characteristics; s4, carrying out cell community feature clustering, and representing a tumor microenvironment distribution mode of the patient; and S5, calculating the similarity of the tumor microenvironment distribution modes of the patients, and layering the patients. By means of the method, the patients can be layered, and prognosis of the patient groups has significant difference.

Description

Technical Field

[0001] The present invention belongs to the technical field of breast cancer treatment, and in particular relates to a patient stratification method based on multi-cell enrichment patterns of single-cell spatial proteomes of breast cancer. Background Art

[0002] Cancer is a major public health problem worldwide. At present, the prognostic evaluation systems for breast cancer mainly include the following four systems: (1) TNM (Tumor Node Metastasis) staging system for breast cancer. (2) Histological grading of breast cancer. (3) Molecular classification of breast cancer. Based on the gene expression information of estrogen receptor ER (Estrogen Receptor), progesterone receptor PR (Progesterone Receptor), and human epidermal growth factor receptor 2 HER2 (Human Epidermal growth factor Receptor 2) in breast cancer. (4) Genetic detection methods for breast cancer. These methods mainly focus on the morphology, proliferation ability, and molecular expression of tumor cells in breast cancer. However, in addition to the malignant proliferation caused by gene mutations in cancer cells themselves, the occurrence and development of breast cancer is also closely related to the tumor microenvironment (TME). TME is highly heterogeneous, which is derived from differences in cell phenotype, cell-to-cell interaction, and cell spatial location, affecting the development and prognosis of tumors. Therefore, it is of great significance to accurately analyze the structure of TME and associate it with tumor development and prognosis.

[0003] In order to accurately analyze the structure of TME and establish a single-cell map of tumors, single-cell analysis has entered the era of multi-omics. The molecular information group of cells is used as a parameter set to describe the phenotype and function of cells. For example, single-cell sequencing technology detects the transcriptome information of single cells in dissociated tissues to analyze cell phenotypes and functions and predict cell-to-cell interactions, but lacks spatial information of single cells, cell-to-cell interactions, and tissue structural characteristics; spatial transcriptome sequencing technology obtains the gene expression information of single cells or adjacent cell groups in tissue samples and their relationship with their spatial position in the tissue, but spatial transcriptome technology, which is a hot research topic, cannot achieve single-cell resolution; spatial metabolome technology can obtain metabolome information in single cells, analyze TME from the spatial metabolic level, discover new phenotypes of cell metabolism, and analyze the mechanism of metabolic pathways remodeling the microenvironment, but it is currently impossible to accurately identify cell phenotypes; spatial proteome technology detects the expression of proteins in single cells and obtains the spatial location information of single cells. At the same time, protein markers can accurately define cell phenotypes and functions. Among the transcriptome, metabolome, and proteome, proteins are the main bearers of cell life activities. Their abnormalities are often closely related to diseases, and tissue space is the research level closest to biological reality. Therefore, obtaining the protein localization expression profile of single cells at the tissue spatial level is of great value for analyzing tissue microenvironment, diagnosis and prognosis, and precision medicine.

[0004] At present, single-cell spatial proteomics is an emerging technical field. There is no unified method for analyzing the data, especially the multi-cell enrichment pattern analysis of TME structure, which does not make full use of spatial topological information, resulting in insufficient research on the relationship between cancer spatial dimension analysis and prognosis, and fewer methods for stratifying patients based on the analysis of TME structure based on multi-cell enrichment patterns. Therefore, the present invention takes breast cancer as the background and proposes a patient stratification method based on the multi-cell enrichment pattern of single-cell spatial proteomics, which can achieve patient stratification and obtain some new biological insights. Summary of the invention

[0005] In order to comprehensively solve the above problems, especially to address the deficiencies in the prior art, the present invention provides a patient stratification method based on the multicellular enrichment pattern of single-cell spatial proteome of breast cancer, which can achieve stratification of breast cancer patients and obtain some new biological insights.

[0006] In view of this, the present invention provides a patient stratification method based on the multi-cell enrichment pattern of single-cell spatial proteome of breast cancer, comprising the following steps:

[0007] S1: Cell segmentation and cell type identification of single-cell spatial proteomics image data of breast cancer

[0008] Use the deep learning method Mesmer or CellProfiler software combined with Ilastik software to perform cell segmentation on the image data to obtain a cell mask image (cell mask), and then superimpose the cell mask image on each protein channel image to obtain the expression of each protein in each cell to obtain a cell protein expression matrix. Then, based on the cell protein expression matrix, a clustering algorithm is used to cluster them to obtain a cell group protein expression heat map, and the cell phenotype of the cell group is identified based on the protein expression of the cell group.

[0009] S2: Constructing a cell network diagram to simulate the tumor microenvironment

[0010] Use the cell mask image mentioned in S1 and the image processing software package skimage or opencv to extract the coordinates of each cell's centroid, calculate the distance between each cell's centroid and other cell's centroids, set a radius value, and consider the cells within this value as the cell's neighbors, that is, connect the two cells with an edge, and do not connect the edges beyond the radius range to construct a cell network diagram.

[0011] S3: Detect tightly connected cell communities and extract cell community features

[0012] Using the cell network graph mentioned in S2, a community detection algorithm was used to detect densely connected cell communities. To make full use of the spatial topological information, the number of edges connecting the cell types contained in each cell community was extracted. The number of edges connecting the cell types represented the importance of the cell type in the cell community and could better reflect the arrangement of the cells in the cell community. The number of edges connecting each cell type was used as a characteristic representation of the cell community.

[0013] S4: Clustering of cell community features to characterize the distribution pattern of the patient's tumor microenvironment

[0014] Using the cell community features extracted by S3, a clustering algorithm was used to obtain k tumor microenvironment characteristic structures. The frequency of occurrence of each tumor microenvironment characteristic structure for each patient was counted as a representation of the patient's tumor microenvironment distribution pattern.

[0015] S5: Calculate the similarity of the distribution patterns of the patient's tumor microenvironment and stratify the patients

[0016] The tumor microenvironment distribution pattern of each patient obtained by S4 was used to calculate the similarity of the tumor microenvironment distribution patterns of each patient. Then, the patient was used as a node, and the similarity between patients was used as the weight of the edge connecting the patients to construct a graph. Then, the patients were divided into different groups through the community detection algorithm.

[0017] The beneficial effect of the present invention is that the present invention can fully utilize spatial topological information based on a large data set, extract multicellular enrichment patterns to analyze TME structural characteristics, and characterize the TME structural distribution pattern of the patient's tumor microenvironment, so as to accurately stratify patients with different prognoses. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is the overall operation flow chart of the present invention;

[0019] Figure 2 This is a schematic diagram of constructing a cell community network according to the present invention;

[0020] Figure 3 It is a schematic diagram of clustering of cell community features extracted by the present invention;

[0021] Figure 4 This is a schematic diagram of the present invention for characterizing the distribution pattern of the patient's tumor microenvironment and stratifying the patients;

[0022] Figure 5 It is a schematic diagram of the patient stratification results of the present invention. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present application to clearly describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments in the present application belong to the scope of protection of this application.

[0024] In the description of the present application, it should be noted that the terms used herein are only for describing specific embodiments, and are not intended to limit the exemplary embodiments according to the present application. For ease of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship. The technology, methods and equipment known to ordinary technicians in the relevant field may not be discussed in detail, but in appropriate cases, the technology, methods and equipment should be regarded as part of the authorization specification. In all examples shown and discussed here, any specific value should be interpreted as merely exemplary, rather than as a limitation. Therefore, other examples of the exemplary embodiments may have different values. It should be noted that similar numbers and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further discussed in subsequent drawings.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of one type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0026] It should be noted that, in the description of the present application, the orientation or positional relationship indicated by terms such as "front, back, up, down, left, right", "lateral, vertical, perpendicular, horizontal" and "top, bottom" are usually based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description. Unless otherwise stated, these orientation words do not indicate or imply that the device or element referred to must have a specific orientation or be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the scope of protection of the present application; the orientation words "inside and outside" refer to the inside and outside relative to the contour of each component itself.

[0027] It should be noted that, in the present application, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises one..." does not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be noted that the scope of the method and device in the embodiment of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0028] Embodiment 1:

[0029] like Figure 1-Figure 5 As shown, this embodiment provides a patient stratification method based on the multi-cell enrichment pattern of single-cell spatial proteome of breast cancer, comprising the following steps:

[0030] S1: Cell segmentation and cell type identification of single-cell spatial proteomics image data of breast cancer

[0031] For the deep learning method Mesmer, it is necessary to input the nuclear channel image and the cell membrane or cytoplasm channel image. For single-cell spatial proteomics images, there are often multiple nuclear channels and multiple cell membrane or cytoplasm channel images. The pixel values ​​of multiple nuclear channel images, cell membrane or cytoplasm channel images are averaged to obtain the fused nuclear channel, cell membrane or cytoplasm channel, which are input into the Mesmer method to output a cell mask image. For the CellProfiler software combined with the Ilastik software method, the cell nucleus is first detected based on the Ilastik software, and then the nuclear area is expanded based on the propagation method to cover the cell membrane and cytoplasm signals, and the signal intensity gradient is stopped at the edge of the cell to generate a probability map, which is then input into the CellProfiler software to generate a cell mask image.

[0032] After obtaining the cell mask, the expression value of a protein on a single cell is defined as the average pixel value of the image area surrounded by the cell mask. In this way, a cell protein expression matrix is ​​obtained, and then the clustering algorithm FlowSOM is used to cluster the cells according to their protein expression to obtain a cell group protein expression matrix, and then the cell group type is identified based on the expression of each cell group protein.

[0033] S2: Constructing a cell network diagram to simulate the tumor microenvironment

[0034] Using the cell mask image obtained by S1, use the Python package skimage or opencv to extract the coordinates of each cell's centroid, calculate the distance between each cell's centroid and other cell's centroids, set a radius value, and consider the cells within this value to be the "neighbors" of the cell, that is, connect the two cells with an edge, and do not give edge connection if the radius range is exceeded. That is, the cells are regarded as nodes, and if the cells are neighbors, there are edge connections between them to construct a cell network diagram.

[0035] S3: Detect tightly connected cell communities and extract cell community features

[0036] The cell network diagram obtained using S2 was used to identify densely connected cell communities using the community detection algorithm random walktrap. In order to better utilize the spatial topological information of cells, the number of connected edges of the cell types contained in each cell community was extracted to characterize the cell community, i.e., the multi-cellular enrichment pattern.

[0037] S4: Clustering of cell community features to characterize the distribution pattern of the patient's tumor microenvironment

[0038] Using the characterization vector of each multicellular enrichment pattern obtained by S3, and using the Canberra distance as the distance measurement method, hierarchical clustering was performed to cluster similar multicellular enrichment patterns together to obtain k tumor microenvironment characteristic structures. The frequency of occurrence of the characteristic microenvironment structure of each patient was statistically analyzed to characterize the distribution pattern of the patient's tumor microenvironment, namely:

[0039]

[0040] S5: Calculate the similarity of the distribution patterns of the patient's tumor microenvironment and stratify the patients

[0041] Using the tumor microenvironment distribution pattern of each patient obtained by S4, the cosine similarity of the tumor microenvironment distribution pattern between two patients was calculated, that is:

[0042]

[0043] Get the similarity matrix:

[0044]

[0045] According to the similarity value, the nearest k neighbors of each patient were obtained, and the Intersection over Union (IOU) matrix was calculated, which is the number of intersections of the nearest k neighbors between two patients divided by the number of unions. The IOU matrix was used to construct a network diagram between patients, and then the Louvain community detection algorithm was used to group the patients. Finally, the multivariate log-rank test was used to compare the survival outcomes of each group based on the overall survival of the patients and whether the death event occurred. The results showed that there were statistically significant differences between the groups. Combined with clinical information, it was found that patients could be accurately stratified with different prognoses.

[0046] The embodiments of the present application are described above in conjunction with the accompanying drawings. In the absence of conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. A patient stratification method based on multicellular enrichment patterns of single-cell spatial proteomes of breast cancer, characterized in that: The following steps are involved: S1, cell segmentation and cell type identification of breast cancer single-cell spatial proteomics image data; S2. Construct a cell network diagram to simulate the tumor microenvironment; S3, detect tightly connected cell communities and extract cell community features; S4, clustering of cell community features to characterize the distribution pattern of the patient's tumor microenvironment; S5. Calculate the similarity of the distribution patterns of the patients’ tumor microenvironment and stratify the patients.

2. A patient stratification method based on multicellular enrichment pattern of single-cell spatial proteome of breast cancer according to claim 1, characterized in that: The S1 specifically includes the following steps: Use the deep learning method Mesmer or CellProfiler software combined with Ilastik software to perform cell segmentation on the image data to obtain a cell mask image, which is then superimposed on each protein channel image to obtain the expression of each protein in each cell to obtain a cell protein expression matrix. Then, based on the cell protein expression matrix, a clustering algorithm is used to cluster them to obtain a cell population protein expression heat map, and the cell type of the cell population is identified based on the protein expression of the cell population.

3. A patient stratification method based on multicellular enrichment pattern of single-cell spatial proteome of breast cancer according to claim 2, characterized in that: The S2 specifically includes the following steps: Using the cell mask image mentioned in S1, the coordinates of each cell’s centroid are extracted using an image processing software package, the distance between each cell’s centroid and other cell’s centroids is calculated, a radius value is set, and cells within this value are considered to be neighbors of the cell, that is, an edge is connected between the two cells, and no edge connection is given to cells outside the radius range, and a cell network diagram is constructed.

4. A patient stratification method based on multicellular enrichment pattern of single-cell spatial proteome of breast cancer according to claim 1, characterized in that: The S3 specifically includes the following steps: Using the cell network graph mentioned in S2, a community detection algorithm is used to detect tightly connected cell communities. For each cell community, the number of edges connecting the cell types contained therein is extracted. The number of edges connecting the cell types represents the importance of the cell type in the cell community and can better reflect the arrangement of the cell type in the cell community. The number of edges connecting each cell type is used as a characteristic representation of the cell community.

5. A patient stratification method based on multicellular enrichment pattern of single-cell spatial proteome of breast cancer according to claim 1, characterized in that: The S4 specifically comprises the following steps: Using the cell community features extracted by S3, a clustering algorithm was used to obtain k tumor microenvironment characteristic structures. The frequency of occurrence of each tumor microenvironment characteristic structure for each patient was counted as a representation of the patient's tumor microenvironment distribution pattern.

6. A patient stratification method based on multicellular enrichment pattern of single-cell spatial proteome of breast cancer according to claim 1, characterized in that: The S5 specifically includes the following steps: The tumor microenvironment distribution pattern of each patient obtained by S4 was used to calculate the similarity of the tumor microenvironment distribution patterns of each patient. Then, the patients were used as nodes and the similarity between patients was used as the weight of the edges connecting the patients to construct a graph. Then, the patients were divided into different groups through the community detection algorithm.

Images