Skip to main content
Fig. 2 | Acta Neuropathologica Communications

Fig. 2

From: Differentiated glioblastoma cells accelerate tumor progression by shaping the tumor microenvironment via CCN1-mediated macrophage infiltration

Fig. 2

Validation of DGC-specific transcriptomic signatures in larger tumor cohorts and their correlation with the mesenchymal subtype and a poor patient prognosis. a Hierarchical clustering of human TCGA GBM samples (HG-UG133A, n = 528) into DGC-high (n = 177), DGC-medium (n = 163), and DGC-low (n = 188) groups using DGC signature genes. The corresponding DGC signature groups for each sample determined via hierarchical clustering are labeled (top). Heat map showing Z-scores of DGC and GSC signatures determined by single sample gene set enrichment analysis (ssGSEA) of individual GBM samples (second and third from top). The corresponding GBM subtype for each sample is labeled (fourth from top). b Heat map shows expression the DGC and GSC signature ssGSEA scores (top and second from top) and genes of each signature (rows) for each RNA sample in the anatomic structure study dataset (122 RNA sample data from 10 patients) from the IVY GAP database (columns). The corresponding GBM subtype and histology for each sample is are (bottom and second from bottom). c ssGSEA scores of DGC signature genes of DGC-high (n = 177), DGC-medium (n = 163), and DGC-low (n = 188) patients in TCGA GBM dataset (HG-UG133A, n = 528). Violin plots represent the median (thick dotted line) and quartiles (dotted line). ****P < 0.0001, one-way ANOVA with Tukey’s multiple comparisons test. d ssGSEA scores of the GSC signature genes of three DGC groups in TCGA GBM dataset (HG-UG133A, n = 528). Violin plots represent the median (thick dotted line) and quartiles (dotted line). **P < 0.01, ****P < 0.0001, one-way ANOVA with Tukey’s multiple comparisons test. e Correlation analysis between DGC and GSC ssGSEA scores in TCGA GBM dataset (HG-UG133A, n = 528) and IVY GAP dataset (122 RNA sample data from 10 patients). Pearson’s correlation test. f ssGSEA scores of the DGC signature genes in multiple regions of the IVY GAP data (122 RNA sample data from 10 patients). Violin plots represent the median (thick dotted line) and quartiles (dotted line). **P < 0.01, ****P < 0.0001, one-way ANOVA with Tukey’s multiple comparisons test. g Molecular subtype distribution among DGC signature groups in TCGA GBM dataset (HG-UG133A, n = 528). ****P < 0.0001, chi-squared test. h Kaplan–Meier analyses between patients in DGC-high and DGC-low groups of TCGA GBM dataset (HG-UG133A). Log-rank P value analyses

Back to article page