Skip to main content
Fig. 1 | Journal of Experimental & Clinical Cancer Research

Fig. 1

From: SF3B1 inhibition disrupts malignancy and prolongs survival in glioblastoma patients through BCL2L1 splicing and mTOR/ß-catenin pathways imbalances

Fig. 1

SF3B1 is mutated and markedly overexpressed in human GBM samples compared to non-tumor brain samples. a Somatic mutation rate of SF3B1 as well as of commonly mutated genes in glioma samples (IDH1, TP53, ATRX, PTEN, IDH2) obtained from the CGGA-dataset (n = 284 patients). Type of alterations, Overall Survival (OS), censored samples, glioma grade, and subtype are also indicated. b Percentage and similarity of somatic mutations rate of IDH1, TP53, ATRX, PTEN, IDH2 and SF3B1 genes in gliomas across the three different available datasets [CGGA-dataset (n = 284); TCGA-dataset (n = 746); MSKCC-dataset (n = 923)]. c Kaplan-Meier survival curves for glioma patients with mutated and wildtype SF3B1 obtained from the CGGA-, TCGA- and MSKCC-datasets (SF3B1mut, n = 15; SF3B1wt, n = 1849). d Non-hierarchical heatmap generated comparing the expression levels of SF3B1 in control brain tissues and/or GBM samples using our cohort, Rembrandt, and CGGA cohorts. Receiver-Operating-Characteristic (ROC)-curve analysis of SF3B1 expression using control and GBM samples from our cohort (e) and the external Rembrandt cohort (f). Single-cell characterization of SF3B1 through intra-tumor human cell populations: [g Principal components analysis (PCA) discriminating tumor microenvironment (TME) cells from tumor-like cells from a single-cell dataset. h Distribution of SF3B1 expression in distinctive Uniform Manifold Approximation and Projection (UMAP) cluster (Top panel: Match of UMAP clusters with intra-tumor cell subtypes; Bottom panel: UMAP feature plot showing SF3B1 expression). i SF3B1 expression across different intra-tumor cell subtypes identified. j GBM cells classified by transcriptional programs in two-dimensional representation using Relative meta-module score [log2(|SC1-SC2| + 1)]. Each quadrant corresponds to one cellular state. k SF3B1 expression across different GBM cell transcriptional programs]. Correlation of SF3B1 with different key prognostic biomarkers (l) and relevant spliceosome components (m) in GBM samples from CGGA (upper panel) and Rembrandt (lower panel) cohorts including non-tumor samples for Rembrandt dataset. Asterisks (*P < 0.05; **P < 0.01; ***P < 0.001) indicate statistically significant differences across different conditions. Plus symbol (+) indicates a tendency between conditions (+P > 0.05 < 0.1)

Back to article page