Research Paper Volume 11, Issue 22 pp 10316—10337

Multi-omics analysis reveals epithelial-mesenchymal transition-related gene FOXM1 as a novel prognostic biomarker in clear cell renal carcinoma

Figure 3. Analysis of expression dynamic changed genes (EDCGs). (A) Unsupervised hierarchical clustering for patients with ccRCC (n = 603) based on 756 EMT-related genes. The log2 count per million were used. The samples (x-axis) and genes (y-axis) were clustered into four clusters (sC1 to sC4) and seven clusters (gC1 to gC7), respectively. (B) Overall survival analysis for sample clusters. (C) Progression-free survival analysis for sample clusters. (D) Definition of EDCGs and calculation of the degree of expression change (Delta [δ]) for specific gene. (E) Clustering of Delta values of EDCGs for stage, grade, lymph node metastasis, distant metastasis and recurrence. (F) The expression of EMT/MET markers in ccRCC. Only differentially expressed in at least two out of three datasets were considered statistical significance. * P <0.05, ** P <0.01, *** P <0.001.