Skip to main content
Fig. 4 | BMC Bioinformatics

Fig. 4

From: Differential correlation analysis of glioblastoma reveals immune ceRNA interactions predictive of patient survival

Fig. 4

Prognostic significance of ceRNA interaction in GBM. a Kaplan-Meier curves of ceRNA regulatory strength, as well as expressional abundance, of three immune response genes, CCL22, IL2RB, and IRF4, of the TCGA dataset. b Illustration of the association between patients’ survival and the strength of ceRNA interaction among the three genes. In samples with lowly-expressed hsa-miR-34a, ceRNA interactions among the three genes were intensified and prognosis was improved. c-d Kaplan-Meier curves of two validation datasets. In each dataset, a machine learning procedure was performed to identify a group of samples with high average correlation among the three ceRNAs. The procedure was repeated for 1,000 iterations to ensure convergence. Left panels, Kaplan-Meier curves between the optimized group and others in a representative iteration (see Additional file 9: Figure S5 for summaries of all iterations). Right panels, Kaplan-Meier curves of individual ceRNAs

Back to article page