Skip to main content
Fig. 5 | Genome Medicine

Fig. 5

From: Combined exome and transcriptome sequencing of non-muscle-invasive bladder cancer: associations between genomic changes, expression subtypes, and clinical outcomes

Fig. 5

Transcriptome-wide characteristics of the expression subtypes identified. To further investigate the characteristics of the expression subtypes identified in the RNA sequencing cohort, immune deconvolution, regulon analysis, and circular RNA prediction were undertaken. A Regulon analysis was performed using the RTN BioConductor package, on the 23 regulons previously reported in UBC. Of the 23, seven regulons were detected with significant differential activity across expression subtypes. The regulon activity score as estimated by RTN is represented as a heatmap. B As the RNA sequencing was performed using total RNA, the potential impact of regulatory RNA was assessed using circular RNA (circRNA) prediction. The circular-to-linear ratio was used to perform the differential expression analysis across the expression subtypes, and 31 circRNAs were found to be statistically significant. These belong to 28 unique genes. The gene involved and the chromosomal boundaries of the back-splicing junction are given on the right-hand side. C Immune deconvolution was performed (using ConsensusTME) on the bulk RNA sequencing data, and the proportion of the immune cell classes estimated is denoted as a heatmap. The last row in the heatmap is for the combined “Immune_Score” as determined by ConsensusTME. D The contrast in the immune profile of the three expression subtypes (classes) is further highlighted in the boxplot where immune scores (as estimated by ConsensusTME) per patient are compared. Non-parametric Kruskal-Wallis test was applied to determine the statistical significance

Back to article page