Fig 1.
Expression and functional analysis of MRGs.
(A) Volcano map of differentially expressed MRGs in EC. The red dots represented highly expressed genes, and the green dots represented low-expressed genes. (B) The GO functional enrichment analysis of the differential MRGs. The size of the dots indicates the number of genes attributed to the corresponding category. The color of the dots represented the q value. (C) KEGG pathway analysis of the differentially expressed MRGs. The color of the bars represented the q value. (D) The survival curves of the molecular subtypes. The red curve represented Cluster I and the blue curve represented Cluster II. (E) The GO functional enrichment analysis of the differentially expressed genes between Cluster I and Cluster II. The size of the dots indicates the number of genes attributed to the corresponding category. The color of the dots represented the q value. (F) KEGG pathway analysis of the differentially expressed genes between Cluster I and Cluster II. The color of the bars represented the q value.
Fig 2.
Construction and validation of the risk model based on MRGs.
(A) Cross-validation for tuning the parameter selection in the LASSO regression. (B) LASSO regression of the 8 prognostic MRGs. (C) The 8 prognostic MRGs extracted by Univariate Cox regression analysis were shown in the forest map. (D-F) The K-M survival curves of train set(D), test set(E) and total set (F) based on the MRG-related risk model. The red curve represented the high-risk group, and the blue curve represented the low-risk group. (G-I) Time-dependent ROC curve analysis of train set(G), test set(H) and total set (I).
Fig 3.
Comparison of clinical features and MRG-related signature.
(A) Heatmap of the expression levels of the 8 MRGs contained in the MRG-related signature. (B) The pie chart showing the proportion of patients in the two risk groups for each clinical feature. (C-F) Differences in risk scores among the clinical features, including Age (C), Grade (D), Cluster (E), Stage (F), and LNM (G). (H) The AUC of risk score and clinical characteristics.
Fig 4.
Clinical value of risk score by independent prognostic analysis.
(A) PCA analysis of the two risk groups. (B) t-SNE analysis of the two risk groups. (C) The Univariate analysis of risk model and clinical features. (D) The Multivariate analysis of risk model and clinical features. (E) The Nomogram model based on risk model and clinical features. (F) The calibration curve of the risk model.
Fig 5.
Functional pathways and evaluation of mutation between the two MRG-related risk groups.
(A) The GSVA analysis of two risk subgroups. (B) GO analysis between high-risk and low-risk groups. (C) The top five significant enrichment pathways in the low-risk group by GSEA enrichment analysis. (D) The top five significant enrichment pathways in the high-risk group by GSEA enrichment analysis. (E) The level of TMB between high-risk and low-risk groups. (F) Survival analysis of distinct groups stratified by both TMB and signature. (G-H) The waterfall plot of somatic mutation features established with low (G) and high (H) risk scores.
Fig 6.
The differences in the chemotherapy response of common chemotherapy drugs between the high- and low-risk groups.
(A-F) Relationships between risk scores and IC50 level of Cisplatin (A), Crizotinib (B), Cytarabine(C), Docetaxel(D), Paclitaxel(E), Tamoxifen(F), Sorafenib(G) and Vinorelbine (H). (I) The Spearman’s correlation coefficients between drug susceptibility and expression levels of the 8 genes in the MRG-related risk model.
Fig 7.
Knockdown of suppresses the proliferation of EC cells.
(A-B) The expression of MACC1 in HESCs and EC cells. (C-D) The knockdown efficiency of MACC1 in EC cells. (E) CCK-8 assays of NC and the si2 groups to detect cell viability. The data marked with ns or asterisks were presented as mean ± SD (n = 3) and subjected to ANOVA analysis. ns: not significant, **P < 0.01, and ***P < 0.001 compared to the NC group at the respective time points (0h, 24h, 48h, 72h and 96h). (F) Colony formation assay. (G) The knockdown efficiency of MACC1 in EC cells after two weeks of siRNA transfection. (H) EdU staining were employed to assess cell proliferation. ns: not significant, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.
Fig 8.
Knockdown of suppresses the migration, invasion, and promotes the apoptosis of EC cells.
(A) Wound healing assay. (B) Cell migration and invasion measured through transwell assay. (C) Flow cytometry detected cell apoptosis. (D) Tumor volume, tumor weight and representative tumor images were shown to assess tumor growth. ns: not significant, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.