An endometrial cancer prognosis analysis system, device, and medium

Abstract

The application relates to the biomedical technology field and discloses an endometrial carcinoma prognosis analysis system, equipment and medium, which comprises a cell typing module and a prognosis analysis module connected in sequence. The cell typing module is used for marker recognition on a to-be-detected cell sample of a target object, so that the to-be-detected cell sample is typed into multiple target cell subtypes, including one or more subtypes in tumor cell lineage typing. The tumor cell lineage typing includes a cilia type (FOXJ1), a luminal type (LCN2), a first gland type (SCGB2A1) and a second gland type (CST1). The prognosis analysis module is used for prognosis analysis according to the multiple target cell subtypes, so as to formulate a treatment plan for the target object. The beneficial effect is that the cell sample of the target object can be subjected to targeted cell typing based on expression gene differences, the accuracy of the prognosis analysis result is improved, and important reference bases are provided for clinical treatment guidance of the target object.

Description

Technical Field

[0001] This application relates to the field of biomedical technology, and in particular to a system, device and medium for prognostic analysis of endometrial cancer. Background Technology

[0002] In the diagnosis and treatment of endometrial cancer, doctors need to predict the patient's prognosis based on the distribution of cancer cell subtypes in the body in order to develop an accurate treatment plan. Cancer cell typing methods in related technologies typically focus on the molecular characteristics of cancer cells, or a combination of molecular and immunohistochemical characteristics. However, cell typing based on molecular characteristics suffers from high testing costs, while cell typing based on a combination of molecular and immunohistochemical characteristics has low accuracy in prognostic analysis, resulting in nearly half of patients not receiving effective clinical guidance. Summary of the Invention

[0003] This application provides a system, device, and medium for prognostic analysis of endometrial cancer. It extracts subtype markers based on differences in gene expression among cells, thereby enabling targeted cell typing of cell samples from target subjects, improving the accuracy of prognostic analysis results, and providing a reference for guiding clinical treatment of target subjects.

[0004] To achieve the above objectives, the main technical solutions adopted in this application include: In a first aspect, embodiments of this application provide an endometrial cancer prognostic analysis system, the system comprising a cell typing module and a prognostic analysis module connected in sequence: The cell typing module is used to identify biomarkers in the cell samples of the target object and to perform cell typing based on the biomarker identification results to obtain multiple target cell subtypes; the multiple target cell subtypes are sent to the prognostic analysis module; wherein, the multiple target cell subtypes include one or more of the tumor cell lineage typing, the tumor cell lineage typing includes ciliated type, luminal type, first gland type and second gland type, the subtype biomarker for the ciliated type is FOXJ1, the subtype biomarker for the luminal type is LCN2, the subtype biomarker for the first gland type is SCGB2A1, and the subtype biomarker for the second gland type is CST1; The prognostic analysis module is used to perform prognostic analysis on the target object based on the multiple target cell subtypes, and obtain the prognostic analysis results of the target object, so as to formulate a treatment plan for the target object.

[0005] The endometrial cancer prognostic analysis system proposed in this application includes a cell typing module and a prognostic analysis module connected in sequence. The cell typing module performs cell typing on the cell samples to be tested based on the marker identification results, classifying the cell samples into multiple target cell subtypes. The target cell subtypes are one or more of the tumor cell lineage typing, including ciliated, luminal, first glandular, and second glandular types. The subtype marker for ciliated is FOXJ1, the subtype marker for luminal is LCN2, the subtype marker for first glandular is SCGB2A1, and the subtype marker for second glandular is CST1. After obtaining the cell typing results, the prognostic analysis module performs prognostic analysis on the target objects based on the typing results to obtain the prognostic analysis results of the target objects. Compared to related technologies, this application starts from tumor heterogeneity, selecting subtype markers that reflect the differences in cell lineage based on the differences in expressed genes among cells. This focuses on the molecular heterogeneity of tumor parenchymal cells and the driving mechanism of disease progression through the dynamic and complex interactions between tumor cells and the immune microenvironment. This allows for specific cell typing of the target cell samples, improving the accuracy of tumor cell typing. Furthermore, the typing results can be correlated with the tumor cell microenvironment to perform risk stratification and survival prediction for the target population, providing a reference for clinical treatment guidance and enabling personalized diagnosis and prognostic analysis of endometrial cancer.

[0006] Optionally, the cell typing module further includes a binary system submodule, used for: The target cell sample is subjected to biomarker identification, and cell typing is performed based on the biomarker identification results to obtain multiple target cell subtypes; the multiple target cell subtypes are sent to the prognostic analysis module; wherein, the target object is an object belonging to the prognostic heterogeneity classification in the molecular typing system, the prognostic heterogeneity classification includes mismatch repair deficiency and nonspecific molecular profiles; the multiple target cell subtypes include one or more of the dichotomous lineage classification, the dichotomous lineage classification includes ciliated type and luminal type, the subtype biomarker of the ciliated type is FOXJ1, and the subtype biomarker of the luminal type is LCN2.

[0007] Optionally, the system further includes a system construction module, which includes a gene sequencing submodule for performing single-cell gene sequencing on a reference cell sample of a reference object to obtain gene sequencing data of the reference cell sample for cell typing of the reference cell sample.

[0008] Optionally, the reference cell sample includes target type cells, which include reference tumor cells and reference normal cells; the system construction module further includes a reference typing submodule, used for: Based on gene sequencing data, unsupervised clustering was performed on the reference tumor cells and the reference normal cells to obtain multiple reference cell subpopulations of the target type of cells. Based on the differentially expressed genes of each of the multiple reference cell subpopulations, biomarkers were identified for each of the multiple reference cell subpopulations to obtain subtype biomarkers for each of the multiple reference cell subpopulations. Expression analysis was performed on the subtype markers to obtain the expression ratio of the subtype markers, and the multiple reference cell subpopulations were annotated as reference cell subtypes according to the expression ratio.

[0009] Optionally, the system construction module further includes a cell classification submodule, used for: The reference cell sample is classified into phenotypic lineages based on the gene sequencing data, and target cell types are selected from the phenotypic lineage classification results. Cell sorting is performed on the target type cells to classify them into target tumor cells and target normal cells.

[0010] Optionally, the system construction module further includes a prognostic analysis submodule, used for: Immunohistochemical staining was performed on the subtype markers of each of the reference cell subtypes to obtain histochemically stained sections containing the subtype markers; the histochemically stained sections were then scanned and analyzed to determine the expression levels of each subtype marker. Based on the expression levels of the biomarkers and the prognostic association information of the reference cell subtypes, a prognostic analysis is performed on the reference subjects to obtain the prognostic analysis results of the reference subjects, which are then used to construct the tumor cell lineage typing.

[0011] Optionally, the system construction module further includes a prognostic correlation submodule, used for: Based on the gene sequencing data and the clinical characteristics of the reference subject, clinically associated cell analysis was performed on the reference cell sample to obtain the distribution of clinically associated cells in the reference cell sample. A comparative analysis was performed on the distribution of clinically associated cells based on the reference cell subtypes to determine the proportion of clinically associated cells for each reference cell subtype. Based on the proportion of the cell subtypes, the prognostic correlation information of each of the reference cell subtypes, and the proportion of clinically associated cells, a prognostic analysis is performed on the reference subjects to obtain the prognostic analysis results of the reference subjects.

[0012] Optionally, the gene sequencing submodule is further used for: The reference cell sample was processed into a single cell sample to obtain a single cell sample. Transcriptome sequencing was performed on the single-cell samples to obtain transcriptome data; whole-exome sequencing was performed on the single-cell samples to obtain whole-exome data; T-cell receptor sequencing was performed on the single-cell samples to obtain T-cell receptor library data; and spatial proteome detection was performed on the single-cell samples to obtain spatial proteome data. The gene sequencing data were obtained based on the transcriptome data, the whole exome data, the T cell receptor library data, and the space proteome data.

[0013] Secondly, embodiments of this application provide a computer device, including: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to implement the system described in any of the above embodiments.

[0014] Thirdly, embodiments of this application provide a computer-readable storage medium storing computer instructions, which are used to cause a computer to implement the system described in any one of the above embodiments. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the specific embodiments of this application or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0016] Figure 1 This is a module connection diagram of the endometrial cancer prognostic analysis system provided in the embodiments of this application; Figure 2a The classification results of the Cancer Cell Line Encyclopedia in the embodiments of this application; Figure 2b This is the typing result of the single-cell transcriptome dataset of endometrioid carcinoma in the embodiments of this application; Figure 3 This is a statistical diagram illustrating the expression levels of subtype markers in the second sample cohort in this application embodiment; Figure 4a This is a schematic diagram illustrating the survival prediction of ciliated objects in the embodiments of this application; Figure 4b This is a schematic diagram illustrating the survival prediction of a lumen-type object in an embodiment of this application; Figure 4c This is a schematic diagram illustrating the survival prediction of the first glandular type object in the embodiments of this application; Figure 4dThis is a schematic diagram illustrating the survival prediction of the second glandular type object in the embodiments of this application; Figure 5a This is a schematic diagram of the progression-free survival of each subtype in the WHZJU-EC cohort in the embodiments of this application; Figure 5b This is a schematic diagram illustrating the overall survival of each subtype in the WHZJU-EC cohort in this application embodiment; Figure 6a This is a schematic diagram of the progression-free survival of the dichotomous lineage subtype in the WHZJU-EC cohort in the embodiments of this application; Figure 6b This is a schematic diagram of the overall lifespan of the binary system in the nonspecific molecular spectrum of the WHZJU-EC cohort in the embodiments of this application; Figure 7a This is a schematic diagram of the progression-free survival of ciliated objects in the bipolar lineage typing in this application embodiment; Figure 7b This is a schematic diagram of the overall lifespan of ciliated objects in the bipolar phylogenetic classification of this application embodiment; Figure 7c This is a schematic diagram of the progression-free survival of a lumen-type object in the bipolar classification of this application. Figure 7d This is a schematic diagram of the overall lifespan of the lumen-type object in the bipolar classification of this application; Figure 8 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0018] In the diagnosis and treatment of endometrial cancer, doctors need to predict the patient's prognosis based on the distribution of cancer cell subtypes in the body in order to develop an accurate treatment plan. Cancer cell typing methods in related technologies typically focus on the molecular characteristics of cancer cells, or a combination of molecular and immunohistochemical characteristics. However, cell typing based on molecular characteristics suffers from high testing costs, while cell typing based on a combination of molecular and immunohistochemical characteristics has low accuracy in prognostic analysis, resulting in nearly half of patients not receiving effective clinical guidance.

[0019] To address the aforementioned issues, this application provides an endometrial cancer prognostic analysis system, device, and medium, comprising a cell typing module and a prognostic analysis module connected in sequence. The cell typing module is used to identify biomarkers in the cell samples of the target patient to classify the cell samples into multiple target cell subtypes, including one or more subtypes in the tumor cell lineage classification, which includes ciliated (FOXJ1), luminal (LCN2), first glandular (SCGB2A1), and second glandular (CST1) types. The prognostic analysis module is used to perform prognostic analysis based on the multiple target cell subtypes to develop a treatment plan for the target patient.

[0020] The endometrial cancer prognostic analysis system proposed in this application includes a cell typing module and a prognostic analysis module connected in sequence. The cell typing module performs cell typing on the cell samples to be tested based on the marker identification results, classifying the cell samples into multiple target cell subtypes. The target cell subtypes are one or more of the tumor cell lineage typing, including ciliated, luminal, first glandular, and second glandular types. The subtype marker for ciliated is FOXJ1, the subtype marker for luminal is LCN2, the subtype marker for first glandular is SCGB2A1, and the subtype marker for second glandular is CST1. After obtaining the cell typing results, the prognostic analysis module performs prognostic analysis on the target objects based on the typing results to obtain the prognostic analysis results of the target objects.

[0021] Compared to related technologies, this application starts from tumor heterogeneity, selecting subtype markers that reflect the differences in cell lineage based on the differences in expressed genes among cells. This focuses on the molecular heterogeneity of tumor parenchymal cells and the driving mechanism of disease progression through the dynamic and complex interactions between tumor cells and the immune microenvironment. This allows for specific cell typing of the target cell samples, improving the accuracy of tumor cell typing. Furthermore, the typing results can be correlated with the tumor cell microenvironment to perform risk stratification and survival prediction for the target population, providing a reference for clinical treatment guidance and enabling personalized diagnosis and prognostic analysis of endometrial cancer.

[0022] Reference Figure 1 As shown, this embodiment provides an endometrial cancer prognostic analysis system, which includes a cell typing module 100 and a prognostic analysis module 200 connected in sequence. The cell typing module 100 is used to identify markers in the cell samples of the target object and to perform cell typing based on the marker identification results to obtain multiple target cell subtypes; the multiple target cell subtypes are sent to the prognostic analysis module 200; wherein, the multiple target cell subtypes include one or more of the tumor cell lineage typing, the tumor cell lineage typing includes ciliated type, luminal type, first gland type and second gland type, the subtype marker for ciliated type is FOXJ1, the subtype marker for luminal type is LCN2, the subtype marker for first gland type is SCGB2A1, and the subtype marker for second gland type is CST1.

[0023] The prognostic analysis module 200 is used to perform prognostic analysis on the target object based on multiple target cell subtypes, and obtain the prognostic analysis results of the target object for the purpose of formulating a treatment plan for the target object.

[0024] Specifically, for target subjects requiring prognostic analysis of endometrial cancer surgery, during the diagnosis or surgical treatment process, cell samples are collected from the target subjects and sent to a gene sequencing module. For example, the cell samples can be collected by extracting a small amount of tissue from the tumor site during diagnosis or treatment via puncture biopsy or endoscopic biopsy, or by using the removed portion as a cell sample for testing after tumor resection during surgery.

[0025] Furthermore, the cell typing module 100 identifies biomarkers in the target cell sample and performs cell typing based on the biomarker identification results, classifying the cell sample into multiple target cell subtypes. It is understood that the number of target cell subtypes can be one or more, and in practice, the cell sample typically contains multiple target cell subtypes. In some embodiments, biomarker identification can be performed using methods such as immunohistochemistry (IHC) staining or multiplex immunofluorescence (mIHC) staining, without specific limitations here.

[0026] It should be noted that the target cell subtype in this embodiment may include one or more subtypes from the Dominant Cancer Lineage Subtyping (DCLS) classification. DCLS includes ciliated, luminal, and glandular subtypes. Ciliated cells are those that differentiate into ciliate-like structures, reaching a highly mature level of differentiation, and generally represent a good prognosis. The subtype marker for ciliated cells is FOXJ1. High expression of FOXJ1 usually indicates that the cells have reached a high degree of differentiation, at which point the cells typically possess lower proliferative activity and metastatic potential. Therefore, ciliated cells using FOXJ1 as a subtype marker generally have a good prognosis, indicating a lower risk of recurrence and better treatment outcomes. Luminal cells are those with glandular formation capabilities, reaching a moderate degree of differentiation, representing a relatively good prognosis. The subtype marker for luminal cells is LCN2. High expression of LCN2 usually indicates the presence of an intrinsic malignant drive within the cells, forming an immunosuppressive microenvironment. Therefore, luminal types with LCN2 as a subtype marker usually have poor prognoses, indicating that the target patient is at risk of metastasis or recurrence and requires further escalation of treatment.

[0027] Glandular type refers to cells with glandular formation and secretory function, representing a moderate to favorable prognostic outcome. Glandular types can be further divided into primary glandular type and secondary glandular type, corresponding to different subtype markers. The subtype marker for primary glandular type is SCGB2A1, and for secondary glandular type is CST1. It should be noted that SCGB2A1 is a marker of normal glandular secretory function, specifically highly expressed in glandular epithelium of some organs. High expression of SCGB2A1 usually indicates that cells have undergone a normal and mature glandular differentiation program and are capable of performing normal secretory functions. CST1 is a gene encoding a cysteine ​​protease inhibitor that regulates protease activity and participates in tissue remodeling and immune regulation. High expression of CST1 usually indicates abnormal activation during cellular acquisition of invasiveness and is closely related to behaviors such as epithelial-mesenchymal transition (EMT), cell invasion, metastasis, and immune escape. Understandably, the first glandular type, with SCGB2A1 as the subtype marker, has a good prognosis and can be precisely treated with drugs; the second glandular type, with CST1 as the subtype marker, has a worse prognosis, higher drug resistance, and may be immunosuppressed.

[0028] The tumor cell lineage typing used in this embodiment was derived from surgical samples of endometrial cancer patients treated at the Women's Hospital of Zhejiang University School of Medicine from March 2021 to May 2022. A total of 24 samples were collected, including 20 cases of endometrioid carcinoma, 3 cases of serous carcinoma, and 1 case of clear cell carcinoma. In addition to the surgical samples, cancerous tissue, adjacent tissue, and peripheral blood mononuclear cell samples were also collected to form the first sample cohort (SC-EC). Gene sequencing was performed on the first sample cohort to obtain high-resolution single-cell multi-omics data.

[0029] Considering the biological characteristic that endometrial cancer originates from glandular epithelial cells, all epithelial cells in the first sample cohort were selected, including tumor cells and normal epithelial cells. Unsupervised clustering was performed on all epithelial cells based on single-cell multi-omics data, and the most significantly differentially expressed genes in each cell cluster were identified to provide biological annotation for tumor cells and normal epithelial cells respectively. This resulted in the first sample cohort being divided into 8 tumor cell subtypes and 6 normal cell subtypes. The tumor cell subtypes included the ciliated type (using FOXJ1 as a subtype marker), the glandular type (using SCGB2A1 as a subtype marker), the luminal type (using LCN2 as a subtype marker), the luminal type (using ERBB4 as a subtype marker), the basal type (using KRT5 as a subtype marker), the epithelial-mesenchymal transition-like (EMT-like) type (using SULF1 as a subtype marker), the fibroblast-like type (using COL3A1 as a subtype marker), and the glandular type (using CST1 as a subtype marker). Normal cell subtypes include ciliated type (marked by FOXJ1), glandular type (marked by SCGB2A1), luminal type (marked by LCN2), luminal type (marked by WNT7A), luminal type (marked by CXCL1), and luminal type (marked by MMP7).

[0030] Based on the above typing results, 29 endometrial cancer cell lines from the Cancer Cell Line Encyclopedia (CCLE) were used as the first validation cohort. Unsupervised clustering analysis was performed based on top variable genes, and through high-expression gene feature analysis, the first validation cohort was divided into four tumor cell subtypes. A UMAP (Uniform Manifold Approximation and Projection) diagram was plotted based on the typing results, which can be referenced. Figure 2aAs shown in the figure, the tumor cell subtypes in the first validation cohort include ciliated subtypes with high FOXJ1 expression, epithelial-mesenchymal transition-like subtypes with high RAMP2 expression, luminal subtypes with high LCN2 expression, and luminal subtypes with high ERBB4 expression. It can be seen that using the same cluster analysis method on the validation cohort yields four tumor cell subtypes identical to those in the aforementioned classification system.

[0031] The single-cell transcriptome dataset of endometrioid carcinoma from a published paper was used as the second validation cohort. Analysis was performed using the same clustering analysis procedure, and the results were referenced. Figure 2b As shown, the validation results of the second validation cohort also identified some tumor cell subtypes that are identical to those in the aforementioned classification system. Based on the above consistency analysis, the previous eight tumor cell subtypes can be further divided into six subtypes: ciliated type (using FOXJ1 as a subtype marker), luminal type (using LCN2 as a subtype marker), second luminal type (using ERBB4 as a subtype marker), first glandular type (using SCGB2A1 as a subtype marker), second glandular type (using CST1 as a subtype marker), and epithelial-mesenchymal transition-like type (using SULF1 as a subtype marker).

[0032] In the above-mentioned typing system, the epithelial-mesenchymal transition (EMT) subtype contains several specific genes that are also present in fibroblasts. This makes it easy for some fibroblasts to be mixed in during the definition of the EMT subtype, affecting the accuracy of typing. Therefore, this subtype is not considered in the typing system. The second luminal subtype, which uses ERBB4 as a marker, accounts for a small proportion of all luminal cell types. The main luminal subtype is the one using LCN2 as a marker. Therefore, the second luminal subtype can be merged into the luminal subtype for simplification. Thus, a typing system containing four target cell subtypes is obtained, which is the tumor cell lineage typing proposed in this embodiment.

[0033] Furthermore, the paraffin-embedded samples from the first sample cohort were stained with subtype markers to assess the expression level of each subtype marker and determine the positive expression rate of each subtype marker in the samples. Statistical analysis revealed that the expression cutoff value for FOXJ1 in ciliated samples was 10.5%, for LCN2 in luminal samples it was 37.3%, for SCGB2A1 in the first glandular samples it was 25.4%, and for CST1 in the second glandular samples it was 23.9%. The positive expression rates of each subtype marker in the first sample cohort are shown in Table 1, demonstrating that basic subtyping of the first sample cohort can be performed based on tumor cell lineage typing.

[0034] Table 1. Partial typing results of the first sample cohort. To validate the cutoff values ​​mentioned above, a cohort of 180 endometrial cancer patients (WHZJU-EC) from the Women's Hospital of Zhejiang University School of Medicine between 2015 and 2020 was used as the second sample cohort to validate the cutoff values ​​for the four subtype markers. The validation process included: obtaining paraffin-embedded tissue samples from the patients; performing immunohistochemical staining of the four subtype markers on the paraffin-embedded tissue samples to detect the positive expression rate of each subtype marker; and classifying the second sample cohort according to the positive expression rate of the subtype markers and their corresponding cutoff values. The results can be referenced... Figure 3 As shown, it can be seen that, based on the aforementioned cutoff value, basic typing of the second sample cohort can be performed under tumor cell lineage typing.

[0035] Furthermore, after cell typing of the cell sample to be tested, the prognostic analysis module 200 performs prognostic analysis on the target object based on the expression distribution of multiple target cell subtypes in the cell sample to be tested and their corresponding prognostic conditions, and obtains the prognostic analysis results of the target object.

[0036] To validate the prognostic capabilities of tumor cell lineage subtyping, the TCGA-UCEC dataset for Uterine Corpus Endometrial Carcinoma (UCEC) from The Cancer Genome Atlas (TCGA) project was used as the third sample cohort. The Scissor algorithm was employed to identify Scissor (+ / -) specific cell populations and their differentially expressed genes for each tumor cell subtype in the third sample cohort, thereby establishing a characteristic gene set for prognostic prediction. The results can be referenced... Figures 4a to 4d As shown. Based on Figures 4a to 4d It can be seen that the characteristic gene sets derived from the ciliated, primary glandular, and secondary glandular types are associated with significantly improved survival outcomes, while the characteristic gene sets derived from the luminal type are significantly associated with poor prognosis.

[0037] Next, after classifying the second sample cohort based on tumor cell lineage typing, survival analysis was performed on each tumor cell subtype in the second sample cohort. The results can be found in [reference needed]. Figures 5a to 5b As shown. Based on Figures 5a to 5b It can be seen that the ciliated type is significantly associated with good clinical outcomes, while the luminal type is closely associated with poor prognosis. The prognostic analysis of the first and second glandular types is moderate. Combined with the validation results of the second and third sample cohorts, it is shown that the tumor cell lineage typing proposed in this application can predict clinical prognosis and has the potential to serve as a new method for classifying endometrial cancer.

[0038] The endometrial cancer prognostic analysis system proposed in this application includes a cell typing module 100 and a prognostic analysis module 200 connected in sequence. The cell typing module 100 performs cell typing on the cell sample to be tested based on the marker identification results, classifying the cell sample into multiple target cell subtypes. The target cell subtype is one or more of the tumor cell lineage typing, including ciliated, luminal, first glandular, and second glandular types. The subtype marker for ciliated is FOXJ1, the subtype marker for luminal is LCN2, the subtype marker for first glandular is SCGB2A1, and the subtype marker for second glandular is CST1. After obtaining the cell typing results, the prognostic analysis module 200 performs prognostic analysis on the target object based on the typing results to obtain the prognostic analysis result of the target object.

[0039] Compared with related technologies, this application starts from tumor heterogeneity and selects subtype markers that reflect the differences in cell lineage based on the differences in gene expression among cells. This focuses on the molecular heterogeneity of tumor parenchymal cells, enabling specific cell typing of the target cell samples and improving the accuracy of tumor cell typing. Furthermore, the typing results can be correlated with the tumor cell microenvironment to perform risk stratification and survival prediction for the target population, providing a reference for clinical treatment guidance and achieving personalized diagnosis and prognostic analysis of endometrial cancer.

[0040] As an embodiment of this application, the cell typing module 100 further includes a binary system submodule, used for: identifying markers in the cell samples of the target object, and performing cell typing based on the marker identification results to obtain multiple target cell subtypes; sending the multiple target cell subtypes to the prognostic analysis module 200; wherein, the target object is an object belonging to the prognostic heterogeneity classification in the molecular typing system, the prognostic heterogeneity classification includes mismatch repair defects and nonspecific molecular profiles; the multiple target cell subtypes include one or more of the binary lineage classifications, the binary lineage classifications include ciliated type and luminal type, the subtype marker for ciliated type is FOXJ1, and the subtype marker for luminal type is LCN2.

[0041] Specifically, molecular subtyping systems are based on the molecular characteristics of endometrial cancer. Compared to molecular subtyping systems, tumor cell lineage typing shows greater prognostic heterogeneity and cannot accurately predict prognosis for individuals with deficient mismatch repair (dMMR) and near-silent near-poly (NSNP) molecular profiles, while tumor cell lineage typing struggles to predict outcomes in unknown cohorts. Considering that both subtyping systems can mitigate the difficulties caused by tumor heterogeneity in endometrial cancer subtyping from different perspectives, this embodiment combines the two systems. Tumor cell lineage typing is used to further subtype those subtypes with ambiguous prognostic results in the molecular subtyping system, thus simultaneously considering both molecular and genetic characteristics and reducing the impact of epithelial cell lineage differences on endometrial cancer subtyping.

[0042] It should be noted that, referring to Figures 5a to 5b It can be seen that in tumor cell lineage typing, the prognostic analysis results of the first glandular type (using SCGB2A1 as the subtype marker) and the second glandular type (using CST1 as the subtype marker) are similar. Furthermore, the prognostic analysis results of the first glandular type partially overlap with the ciliated type, and the prognostic analysis results of the second glandular type partially overlap with the luminal type. This leads to a relatively ambiguous prognostic analysis result for both the first and second glandular types, making them unsuitable for practical prognostic analysis. Based on this, this embodiment further simplifies tumor cell lineage typing, proposing a binary lineage typing system including ciliated and luminal types, where the subtype marker for the ciliated type is FOXJ1, and the subtype marker for the luminal type is LCN2. According to... Figures 5a to 5b The results show that there are significant differences in prognostic analysis between ciliated and luminal types. Therefore, further subdividing the cell samples of subjects with prognostic heterogeneity into ciliated and luminal types can effectively differentiate the prognostic manifestations of prognostic heterogeneity and improve the clinical guidance effect.

[0043] Furthermore, in this embodiment, the target object is first classified into cells based on a molecular typing system to obtain the molecular subtype of the target object within the molecular typing system. If the target object is classified as mismatch repair deficient or without a specific molecular profile, the target object is used as the target object in this embodiment. Immunohistochemical staining and other methods are used to identify biomarkers in the cell samples of the target object, identifying the positive expression rates of two subtype biomarkers corresponding to the dichotomous typing in the sample, to obtain the expression level of each subtype biomarker. After obtaining the biomarker identification results, the expression level of each biomarker and its corresponding cutoff value are compared to classify the cell samples into multiple target cell subtypes.

[0044] To verify the prognostic analysis capability of bipartite phylogenetic typing, taking the non-specific molecular profile in prognostic heterogeneity typing as an example, bipartite phylogenetic typing was used to perform cytotyping and prognostic analysis on 89 subjects with non-specific molecular profiles in the second sample cohort. The prognostic analysis method was to predict prognostic performance using the Scissor algorithm. The results can be referenced. Figures 6a to 6b As shown. Based on Figures 6a to 6b It can be seen that the ciliated subtype marker FOXJ1 and the luminal subtype marker LCN2 can significantly distinguish the survival curves of subjects with no specific molecular profile, thus confirming that FOXJ1 and LCN2 can serve as risk stratification markers for subjects with no specific molecular profile and guide clinical diagnosis and treatment strategies for subjects with no specific molecular profile.

[0045] Furthermore, this embodiment also used tissue samples from 47 patients with non-specific molecular profile endometrial cancer collected from Xiangya Hospital of Central South University between 2015 and 2020 as the fourth sample cohort (XY-EC cohort). For the fourth sample cohort, immunohistochemical staining was performed on each tissue sample to stain for two biomarkers, analyzing the expression levels of FOXJ1 and LCN2. Based on the expression levels of each subtype biomarker and their corresponding cutoff values, the subjects in the fourth sample cohort were divided into high-expression and low-expression groups. The high-expression group was defined as having a subtype biomarker expression level greater than or equal to the corresponding cutoff value, while the low-expression group was defined as having a subtype biomarker expression level less than the corresponding cutoff value. Survival analysis was performed on the fourth sample cohort based on the high-expression and low-expression groups for each subtype biomarker. The results can be referenced... Figures 7a to 7d As shown, the prognostic performance of subjects with high FOXJ1 expression was better than that of subjects with low FOXJ1 expression, while the overall survival (OS) of subjects with high LCN2 expression was significantly lower than that of subjects with low LCN2 expression. Based on these results, it can be concluded that FOXJ1 and LCN2, as subtype markers for dichotomous prognostic lineages, can accurately stratify the risk of heterogeneous prognostic individuals, and have high potential clinical application value.

[0046] As one embodiment of this application, the system further includes a system construction module, which includes a gene sequencing submodule for performing single-cell gene sequencing on a reference cell sample of a reference object to obtain gene sequencing data of the reference cell sample for cell typing of the reference cell sample.

[0047] Specifically, reference cell samples can be cell samples with known cell typing results, used as the data foundation for typing system research. After obtaining reference cell samples for the reference subjects, gene sequencing is performed on the reference cell samples at the single-cell level, allowing for independent analysis of each sample cell and obtaining gene sequencing data for each sample cell to characterize the gene expression characteristics of each sample cell. It is understood that tumor cells have their own unique gene characteristics, and the interactions between the gene characteristics of different cells are correlated with the prognostic performance of the subjects.

[0048] As one embodiment of this application, the reference cell sample includes target type cells, which include reference tumor cells and reference normal cells. The system construction module also includes a reference typing submodule, used for: performing unsupervised clustering on the reference tumor cells and reference normal cells according to gene sequencing data to obtain multiple reference cell subpopulations of the target type cells; identifying markers for the multiple reference cell subpopulations according to their differentially expressed genes to obtain subtype markers for each of the multiple reference cell subpopulations; performing expression analysis on the subtype markers to obtain the expression ratio of the subtype markers, and annotating the multiple reference cell subpopulations as reference cell subtypes according to the expression ratio.

[0049] The expression ratio can be obtained by identifying the positive expression rate of any subtype marker in the reference cell subpopulation, thus characterizing the expression level of each subtype marker in the reference cell subpopulation. It is understood that any reference cell subpopulation may contain one or more subtype markers, each with a different expression ratio. Based on the clustering combinations of multiple subtype markers, the reference cell subpopulation can be annotated as the corresponding reference cell subtype.

[0050] Specifically, in the reference cell sample, the target cell type is divided into reference tumor cells and reference normal cells using a reference typing submodule. Unsupervised clustering is then performed on the reference tumor cells and reference normal cells respectively, resulting in multiple reference cell subpopulations. It can be understood that these multiple reference cell subpopulations include tumor cell subpopulations and normal cell subpopulations. The tumor cell subpopulations can be obtained by unsupervised clustering of reference tumor cells, and the normal cell subpopulations can be obtained by unsupervised clustering of reference normal cells. Each reference cell subpopulation has a similar genomic expression profile.

[0051] In some embodiments, the unsupervised clustering process of reference tumor cells may include: determining the expression variation of each gene in the gene sequencing data among reference tumor cells based on expression fluctuations; identifying genes with high expression differences as Top Variable Genes based on the expression differences of each gene; and performing unsupervised clustering of reference tumor cells based on the Top Variable Genes to obtain multiple tumor cell subpopulations based on Top Variable Gene typing. Similarly, reference normal cells can also be unsupervised clustered in a similar manner to obtain multiple normal cell subpopulations.

[0052] Furthermore, gene characterization analysis of reference cell subpopulations is performed using a reference genotyping submodule to identify one or more genes with higher expression levels in any reference cell subpopulation compared to other reference cell subpopulations, which are then designated as differentially expressed genes for that reference cell subpopulation. For any reference cell subpopulation, based on the expression specificity and biological importance of each gene, one or more marker genes are selected from the differentially expressed genes of that reference cell subpopulation as subtype markers for that reference cell subpopulation.

[0053] Furthermore, the expression of subtype markers for any reference cell subpopulation is analyzed using the reference typing submodule to determine the expression ratio of each subtype marker in that subpopulation, generating a marker expression list for that subpopulation. Based on this list, the similarity in marker expression ratios between that subpopulation and known cell subtypes is calculated. The subpopulation is then biologically annotated as the corresponding reference cell subtype based on the similarity calculation results. Understandably, the subtype marker with the highest expression level in the reference cell subtype can be used as the primary marker, serving as a crucial basis for biological annotation and effectively improving the interpretability of the cell typing results.

[0054] As one embodiment of this application, the system construction module further includes a cell classification submodule, used for: classifying the reference cell sample according to gene sequencing data and selecting target type cells from the phenotypic classification results; and sorting the target type cells into target tumor cells and target normal cells.

[0055] Specifically, the cell classification submodule obtains the marker genes for each sample cell in the reference cell sample based on gene sequencing data. Phenotypic lineage classification is then performed on each sample cell based on these marker genes, classifying the sample cells into different cell types. Biological annotation is then performed on the sample cells based on the classification results to obtain the phenotypic lineage classification results. Target cell types are selected from the phenotypic lineage classification results, allowing for prognostic analysis of specific cell types. It should be noted that when performing prognostic analysis for endometrial cancer, the reference cell sample can include multiple cell types, including but not limited to lymphoid lineage cells, myeloid cells, stromal cells, and epithelial cells. Lymphoid lineage cells can include CD4+ T cells, CD8+ T cells, Treg cells (regulatory T cells), γδ T cells, NK cells (natural killer cells), and B cells. Myeloid cells can include monocytes, macrophages, dendritic cells (DCs), and mast cells. Stromal cells can include endothelial cells and fibroblasts. The target cell type can be one or more of the aforementioned cell types. The selection of the target cell type can be based on factors such as the disease scenario of the reference subject, the cell abundance of the corresponding cells in the reference cell sample, the specificity of cell markers, or the method of obtaining the reference cell sample. Since endometrial cancer is a malignant tumor primarily originating from endometrial epithelial cells, its main pathological process occurs within the tumor epithelial cells, and the treatment of endometrial cancer also primarily targets these cells. Therefore, in the context of endometrial cancer, the target cell type can be epithelial cells.

[0056] After identifying the target cell type, the cell classification submodule performs cell sorting to identify the malignancy status of the target cell type and distinguish between reference tumor cells and reference normal cells. In some embodiments, cell sorting methods may include copy number variation (CNV) assessment, gene expression profile identification, and somatic mutation identification.

[0057] As one embodiment of this application, the system construction module further includes a prognostic analysis submodule, used for: performing immunohistochemical staining on the subtype markers of each reference cell subtype to obtain histochemically stained sections containing the subtype markers; performing scanning analysis on the histochemically stained sections to determine the marker expression levels of each subtype marker; and performing prognostic analysis on the reference object based on the marker expression levels and the prognostic association information of each reference cell subtype to obtain the prognostic analysis results of the reference object for constructing tumor cell lineage typing.

[0058] Among them, prognostic association information can be the prognostic performance represented by each reference cell subtype.

[0059] Specifically, tissue sections of the reference object are obtained, and immunohistochemical staining of the subtype markers of each reference cell subtype is performed on the tissue sections through the prognostic analysis submodule, resulting in histochemically stained sections containing the stained subtype markers.

[0060] Furthermore, a microscopic imaging instrument is used to scan and image the histochemically stained sections to obtain staining imaging results. A prognostic analysis submodule is then used to perform positive expression analysis of subtype markers based on the staining imaging results to determine the marker expression level for each subtype. In some embodiments, the positive expression analysis process may include: for any histochemically stained section, selecting 10 hotspot fields in the staining imaging results of that section; performing marker positivity rate analysis on any subtype marker in each of these hotspot fields to obtain the positive percentage for each hotspot field; and averaging the positive percentages across all hotspot fields to obtain the marker expression level for any subtype.

[0061] Furthermore, the prognostic analysis submodule determines the proportion of the corresponding reference cell subtype in the reference cell sample based on the expression level of each subtype marker. Then, based on the subtype proportion and the prognostic association information of each reference cell subtype, a prognostic analysis is performed on the reference subjects to obtain the prognostic analysis results. It can be understood that a higher proportion of reference cell subtypes representing a good prognosis indicates a better prognosis for the reference subject, allowing for more conservative treatment; conversely, a higher proportion of reference cell subtypes representing a poor prognosis indicates a worse prognosis for the reference subject, requiring more aggressive adjuvant therapy.

[0062] Furthermore, based on the ability of reference cell subtypes to differentiate the prognostic performance of reference subjects, several subtypes with the highest differentiation ability are selected from all reference cell subtypes to construct tumor cell lineage typing.

[0063] As an embodiment of this application, the system construction module further includes a prognostic association submodule, used for: performing clinical association cell analysis on reference cell samples based on gene sequencing data and the clinical characteristics of reference subjects to obtain the distribution of clinical association cells in the reference cell samples; performing a comparative analysis on the distribution of clinical association cells based on reference cell subtypes to determine the proportion of clinical association cells for each reference cell subtype; and performing prognostic analysis on the reference subjects based on the proportion of cell subtypes and the prognostic association information and proportion of clinical association cells for each reference cell subtype to obtain the prognostic analysis results of the reference subjects.

[0064] The distribution of clinically associated cells can be the distribution of clinically associated cells that are correlated with different clinical prognoses in the reference cell sample, used to represent the probability that the reference subject exhibits different clinical prognoses. For example, clinically associated cells can be Scissor cells, which can be cells obtained by clinical manifestation correlation identification through the Scissor algorithm, including Scissor(+) cells, Scissor(-) cells, and Scissor(0) cells. Scissor(+) cells can be cells whose gene sequencing data shows a significant positive correlation with good clinical manifestation, and the enrichment of Scissor(+) cells can indicate that the reference subject has a good prognosis. Scissor(-) cells can be cells whose gene sequencing data shows a significant positive correlation with poor clinical manifestation, and the enrichment of Scissor(-) cells can indicate that the reference subject has a poor prognosis. Scissor(0) cells can be cells whose gene sequencing data does not show a significant correlation with clinical manifestation.

[0065] Specifically, the clinical manifestations of the reference subject during treatment are obtained as clinical features. A prognostic association analysis submodule performs association analysis between the gene sequencing data of the reference cell sample and the clinical features to obtain the correlation between the gene sequencing data and the clinical features. Based on the obtained correlation, the sample cells in the reference cell sample are identified to obtain the distribution of clinically associated cells. In some embodiments, the association analysis and correlation identification may be performed using the Scissor algorithm to conduct clinically associated cell analysis on the reference cell sample to perform cell prognostic scoring.

[0066] Furthermore, through the prognostic association analysis submodule, a comparative analysis of the distribution of clinically associated cells was conducted using multiple reference cell subtypes to determine the proportion of clinically associated cells in each reference cell subtype. It can be understood that the proportion of clinically associated cells includes both positive and negative cell proportions. The positive cell proportion can be the percentage of Scissor(+) cells in the reference cell subtype; a higher positive cell proportion indicates a good prognosis for the reference subject. The negative cell proportion can be the percentage of Scissor(-) cells in the reference cell subtype; a higher negative cell proportion indicates a poor prognosis for the reference subject.

[0067] Furthermore, through the prognostic association analysis submodule, combining the proportion of cell subtypes and the proportion of clinically associated cells, the contribution of each reference cell subtype to the prognostic performance of the reference subject is analyzed based on the prognostic association information of multiple reference cell subtypes. This allows for a prognostic analysis tailored to the specific circumstances of the reference subject, yielding the prognostic analysis results. It is understood that a positive cell proportion typically appears in reference cell subtypes that also represent a favorable prognosis, while a negative cell proportion typically appears in reference cell subtypes that also represent a poor prognosis. In some embodiments, a high negative cell proportion may occur in reference cell subtypes representing a favorable prognosis, and a high positive cell proportion may occur in reference cell subtypes representing a poor prognosis. This indicates the presence of a prognostic mechanism in the reference subject that differs from conventional cell typing, resulting in an abnormal prognostic performance.

[0068] As one embodiment of this application, the gene sequencing submodule is further used for: processing a reference cell sample into a single cell to obtain a single cell sample; performing transcriptome sequencing on the single cell sample to obtain transcriptome data; performing whole-exome sequencing on the single cell sample to obtain whole-exome data; performing T-cell receptor sequencing on the single cell sample to obtain T-cell receptor library data; performing spatial proteome detection on the single cell sample to obtain spatial proteome data; and obtaining gene sequencing data based on the transcriptome data, whole-exome data, T-cell receptor library data, and spatial proteome data.

[0069] Specifically, after obtaining reference cell samples from the reference object, the reference cell samples are processed into single-cell samples using the gene sequencing submodule. For example, the single-cell processing may include: washing the reference cell samples multiple times with cold PBS (Phosphate Buffered Saline), placing the washed reference cell samples into a six-well plate, and cutting them into multiple small sample pieces, the volume of which may be 1 mm. 3 The sample pieces were placed in DMEM (Dulbecco's Modified Eagle Medium) solution containing collagenase P and DNAse I, and digested at 37°C for 30 minutes with periodic shaking. After digestion, the sample pieces were filtered, washed with cold PBS, and centrifuged at 300g for 5 minutes, repeated 3 times. The centrifuged cell samples were then treated with erythrocyte lysis buffer (Miltenyi), centrifuged again at 300g for 5 minutes, and resuspended in 0.04% BSA (Bovine Serum Albumin) buffer to obtain a cell suspension. The cell suspension was filtered to obtain multiple single-cell samples.

[0070] Furthermore, multiple gene sequencing operations are performed on multiple single-cell samples of the reference object using the gene sequencing submodule to obtain gene sequencing data for the reference cell samples. The multiple gene sequencing operations may include: performing transcriptome sequencing (RNA-seq) on each single-cell sample to detect the types and abundance of RNA molecules in the single-cell sample genome, obtaining transcriptome data for the single-cell sample; performing whole exome sequencing (WES) on each single-cell sample to sequence exon fragments in the single-cell sample, obtaining whole exome data for the single-cell sample; performing T-cell receptor sequencing (TCR-seq) on each single-cell sample to sequence T-cell receptor genes in the single-cell sample, obtaining T-cell receptor library data for the single-cell sample; and performing co-detection byindexing (CODEX) on each single-cell sample to detect and reconstruct protein expression information in the single-cell sample, obtaining spatial proteome data for the single-cell sample.

[0071] Please see Figure 8 , Figure 8 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application, such as... Figure 8 As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). Figure 8 Take a processor 10 as an example.

[0072] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0073] The memory 20 stores instructions executable by at least one processor 10 to enable at least one processor 10 to implement the system shown in the above embodiments.

[0074] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0075] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0076] The computer device also includes a communication interface 30 for communicating with other devices or communication networks.

[0077] This application also provides a computer-readable storage medium. The system described above according to this application can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the system described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that a computer, processor, microprocessor controller, or programmable hardware includes storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the system shown in the above embodiments is implemented.

[0078] This application provides a computer program product including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to implement the system of any embodiment of this application.

[0079] Although embodiments of this application have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of this application, and all such modifications and variations fall within the scope defined by the appended claims.

[0080] The systems, devices, modules, or sub-modules described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0081] For ease of description, the above devices are described in terms of function, divided into various sub-modules. Of course, in implementing this application, the functions of each sub-module can be implemented in one or more software and / or hardware.

[0082] Those skilled in the art will understand that embodiments of this application can be provided as a system or a computer program product. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0083] This application is described with reference to flowchart illustrations and / or block diagrams of devices (systems) and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0084] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0085] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0086] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

[0087] Although embodiments of this application have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of this application, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A prognostic analysis system for endometrial cancer, characterized in that, The system includes a cell typing module and a prognostic analysis module connected in sequence: The cell typing module is used to identify biomarkers in the cell samples of the target object and to perform cell typing based on the biomarker identification results to obtain multiple target cell subtypes; the multiple target cell subtypes are sent to the prognostic analysis module; wherein, the multiple target cell subtypes include one or more of the tumor cell lineage typing, the tumor cell lineage typing includes ciliated type, luminal type, first gland type and second gland type, the subtype biomarker for the ciliated type is FOXJ1, the subtype biomarker for the luminal type is LCN2, the subtype biomarker for the first gland type is SCGB2A1, and the subtype biomarker for the second gland type is CST1; The prognostic analysis module is used to perform prognostic analysis on the target object based on the multiple target cell subtypes, and obtain the prognostic analysis results of the target object, so as to formulate a treatment plan for the target object.

2. The system according to claim 1, characterized in that, The cell typing module also includes a binary system submodule, used for: The target cell sample is subjected to biomarker identification, and cell typing is performed based on the biomarker identification results to obtain multiple target cell subtypes; the multiple target cell subtypes are sent to the prognostic analysis module; wherein, the target object is an object belonging to the prognostic heterogeneity classification in the molecular typing system, the prognostic heterogeneity classification includes mismatch repair deficiency and nonspecific molecular profiles; the multiple target cell subtypes include one or more of the dichotomous lineage classification, the dichotomous lineage classification includes ciliated type and luminal type, the subtype biomarker of the ciliated type is FOXJ1, and the subtype biomarker of the luminal type is LCN2.

3. The system according to claim 1, characterized in that, The system also includes a system construction module, which includes a gene sequencing submodule for performing single-cell gene sequencing on a reference cell sample of a reference object to obtain gene sequencing data of the reference cell sample for cell typing of the reference cell sample.

4. The system according to claim 3, characterized in that, The reference cell sample includes target cell types, which include reference tumor cells and reference normal cells; the system construction module also includes a reference typing submodule, used for: Based on gene sequencing data, unsupervised clustering was performed on the reference tumor cells and the reference normal cells to obtain multiple reference cell subpopulations of the target type of cells. Based on the differentially expressed genes of each of the multiple reference cell subpopulations, biomarkers were identified for each of the multiple reference cell subpopulations to obtain subtype biomarkers for each of the multiple reference cell subpopulations. Expression analysis was performed on the subtype markers to obtain the expression ratio of the subtype markers, and the multiple reference cell subpopulations were annotated as reference cell subtypes according to the expression ratio.

5. The system according to claim 3, characterized in that, The system construction module also includes a cell classification submodule, used for: The reference cell sample is classified into phenotypic lineages based on the gene sequencing data, and target cell types are selected from the phenotypic lineage classification results. Cell sorting is performed on the target type cells to classify them into target tumor cells and target normal cells.

6. The system according to claim 4, characterized in that, The system construction module also includes a prognostic analysis submodule, used for: Immunohistochemical staining was performed on the subtype markers of each of the reference cell subtypes to obtain histochemically stained sections containing the subtype markers; the histochemically stained sections were then scanned and analyzed to determine the expression levels of each subtype marker. Based on the expression levels of the biomarkers and the prognostic association information of the reference cell subtypes, a prognostic analysis is performed on the reference subjects to obtain the prognostic analysis results of the reference subjects, which are then used to construct the tumor cell lineage typing.

7. The system according to claim 6, characterized in that, The system construction module also includes a prognostic correlation submodule, used for: Based on the gene sequencing data and the clinical characteristics of the reference subject, a clinical association analysis was performed on the reference cell sample to obtain the distribution of clinically associated cells in the reference cell sample; A comparative analysis was performed on the distribution of clinically associated cells based on the reference cell subtypes to determine the proportion of clinically associated cells for each reference cell subtype. Based on the proportion of the cell subtypes, the prognostic correlation information of each of the reference cell subtypes, and the proportion of clinically associated cells, a prognostic analysis is performed on the reference subject to obtain the prognostic analysis results of the reference subject.

8. The system according to claim 4, characterized in that, The gene sequencing submodule is also used for: The reference cell sample was processed into a single cell sample to obtain a single cell sample. Transcriptome sequencing was performed on the single-cell samples to obtain transcriptome data; whole-exome sequencing was performed on the single-cell samples to obtain whole-exome data. T-cell receptor sequencing was performed on the single-cell samples to obtain T-cell receptor library data; The single-cell samples were subjected to space proteomics analysis to obtain space proteomics data; The gene sequencing data were obtained based on the transcriptome data, the whole exome data, the T cell receptor library data, and the space proteome data.

9. A computer device, characterized in that, include: A memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, the processor executing the computer instructions to implement the system of any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing a computer to implement the system of any one of claims 1 to 8.

Images