| Objective: Screening prognostic genes for esophageal squamous cell carcinoma(ESCC)patients using GEO and TCGA databases,as well as cell communication analysis to comprehend the mode of cell connection in the tumor microenvironment,we hope to provide a theoretical framework for prognosis prediction,targeted therapy,and tumor metastasis mechanism for ESCC patients.Methods:(1)From the extensive gene expression database,the microarray data sets for the human esophageal squamous cell carcinoma(GSE17351,GSE20347,GSE38129,and GSE77861)were obtained and downloaded.Both the batch correction approach and the sequencing method were used to assess the discrepancies.Use the "venn" diagram to identify the differentially expressed genes(DEGs),and then use enrichment analysis to comprehend the biological activities and processes of the DEGs.The Cytoscape program is used to visualize the STRING database’s protein-protein interaction(PPI)association file,and Cyto Hubba is used to find important genes.Expression and survival analyses are verified using the GSE23400 data set and the Kaplan-Meier plotter database.Download the clinical and RNA sequencing data for ESCC patients from the Cancer Genome Atlas(TCGA)database,perform univariate and multivariate Cox regression analysis on the key genes from the GEO data set to identify prognostication-related genes,create the risk score model,and assess the model.The nomogram and calibration curve were built,together with clinicopathological traits and risk-score.Gene set enrichment analysis(GSEA)was utilized to examine the KEGG pathway,which mostly implicated the model gene and risk-score.(2)All samples from the TCGA queue were examined to determine the relative quantity of invading immune cells using the CIBERSORT algorithm.The link between the number of immune cells and the risk scores of six platforms was investigated using Spearman correlation analysis,and the TIMER database was utilized to study the infiltration relationship between model genes and immune cells.The variations in the tumor immune microenvironment between high-risk and low-risk groups were examined using the "estimate" R package,and the differences in immune cell infiltration and immune-related pathways were examined using single-sample gene set enrichment analysis(ss GSEA).Download ESCC patient maf mutation data from the TCGA database to create a gene mutation map and examine the differences in gene mutation traits across high-risk groups.Using the "p RRophic" R package,the difference in IC50 values of 20 chemotherapeutic medicines in various risk categories was examined.(3)Cell annotation and cell communication analysis are performed using the t-distribution domain embedding technique,the "Single R" R packet,and the "cellchat" R packet.Results:(1)Sequencing and batch correction methods were used to identify 878 and 810 differently expressed genes,respectively.The "Venn" map was used to cross 558 of these genes to find the common differentially expressed genes.To exclude non-interacting isolated genes,a PPI network was built,and a network diagram with 239 nodes and 1019 edges was created.Cytohubba discovered a total of 18 important genes.The GSE23400 data set’s expression verification results revealed that there were 18 genes whose expression levels varied between ESCC and normal esophageal tissues,and all of them were up-regulated in ESCC tissues;the KaplanMeier plotter database’s survival verification results revealed 12 genes that were significantly associated with the prognosis of ESCC patients.BUB1 B and ASPM were shown to be the best characteristics for building the risk-score prognostic model of ESCC patients by univariate and Lasso-Cox regression analyses.The outcomes of the survival study demonstrated that the OS of patients in the low-risk group was higher,and the prognostic model’s area under the ROC curve for one year,three years,and five years was,respectively,0.688,0.715,and 0.714.The risk-score prognosis model might be utilized as a stand-alone prognostic factor,according to the results of the multivariate Cox regression analysis.Clinical characteristics and risk score were used to form the area under the ROC curve over a 1-year period: risk-score was 0.688,age was 0.588,sex was 0.546,grade was 0.567,T stage was 0.564,and N stage was 0.579.With an M-stage of 0.538,the risk-score prognosis model is considered to be quite accurate.The nomogram has high stability,as seen by the nomogram calibration curve.(2)The results of immune cell infiltration demonstrated a significantly positive correlation between the levels of T cell follicular helper cells and CD8+T cells and a strongly negative correlation between the levels of macrophage M0 and regulatory T cell infiltration.An analysis of the TIMER database revealed a negative correlation between the expression level of the BUB1 B gene and dendritic cell and neutrophil infiltration,and a similar correlation between the expression level of the ASPM gene and dendritic cell infiltration.The majority of immune cells were positively connected with risk-score,according to Spearman correlation analysis.According to the results of the ss GSEA analysis,there were differences between the high-risk and low-risk groups in three immune cells and one immune-related pathway.The immunological score was greater in the high-risk group,according to the immune microenvironment study.ESCC patients in the high risk group had a greater mutation frequency than those in the low risk group,according to somatic mutation study,which revealed variations in tumor mutation burden(TMB)between the two risk groups.According to the findings of the drug sensitivity study,there were differences in the IC50 values of 12 medicines between high-and low-risk groups.(3)Eight distinct cell subsets,comprising epithelial,endothelial,B,histiocyte,T,smooth muscle,neuron,and monocyte cells,were found in the ESCC tissue.The MIF signal pathway plays a significant role in cell communication through the examination of receptor-ligand combinations and cell-cell contacts.Conclusion: BUB1 B and ASPM have the potential to serve as molecular biomarkers in the early diagnosis of ESCC and may be involved in the carcinogenesis of ESCC.The MIF signal pathway may be crucial for the transmission of signals between ESCC cells.This work aids in the molecular mechanism of ESCC clarification and encourages early detection and targeted ESCC therapy. |
PDF Full Text Request