Figure 5From: Microrna expression signatures predict patient progression and disease outcome in pediatric embryonal central nervous system neoplasmsROC analysis of miRNA data. ROC curves of the differentially miRNAs in each tumor type using the microarray expression data. Embryonal tumors could be separated by miR-34a (A) and miR-891a (B) following meta-analysis. MBs could be separated by miR-192 (C), miR-3617 (D) following initial analysis, whereas miR-3617 (E), miR-3912 (F), miR-4313 (G), miR-548j (H) and miR-548x (I) following meta-analysis. AT/RT samples could be separated by miR-34a following initial analysis (J), miR-3617 following meta-analysis (K), miR-3912 following initial analysis (L), miR-3912 following meta-analysis (M), miR-4313 following initial-analysis (N), miR-4313 following meta-analysis (O), miR-548j following initial analysis (P), miR-548j following meta-analysis (Q), miR-548x following initial analysis (R), miR-548x following meta-analysis (S), miR-607 following initial analysis (T), miR- following initial analysis (U), miR-651 following initial analysis (V) and miR-891a following initial analysis (W). MiRNAs with a p < 0.05 and an AUC > 0.8 were selected as successful distinguishing markers between tumor and normal tissues. Where miRLink; Initial Analysis, All Data; Meta-Analysis.Back to article page