Fig. 4

Construction of fibroblast differentiation-based classification of MESO patients in TCGA cohort. A The correlation network of 39 FDGs. The colors of the left half of 39 circles which correspond to the 39 FDGs, figured out the differentiation states of the genes, while the colors of the other side described the genetic effects on survival (Red: high expression was harmful for survival; Blue: high expression was favorable for survival). The size of circles was inversely proportional to the p value of Cox regression analysis. The links between circles displayed the relationship between genes where red referred to the positive correlation and blue indicated the negative correlation with remarkable statistical significance (p < 0.0001). B The Consensus matrix of k = 3. C The CDF curve and the delta area plot. D The cluster heatmap of FDGs expression in the patients’ MESO samples, based on bulk RNA-seq data from TCGA database, identifying three clusters. The clinical information of the patients in the cohort, including survival outcome, age, gender, race, TNM stage and metastatic condition were listed as well. The color in the heatmap from blue to red demonstrated the progression from low expression to high expression. E The comparison of survival conditions of the three clusters with the Kaplan–Meier curve. F The Kaplan–Meier survival analysis of the MESO patients in the high PCA score group and the low PCA score group. G The boxplot which displayed the distribution of the PCA scores of patients in the three clusters