Fig. 5

BREGRS serves as a reliable prognosis predictor in BLCA. (A) The t-SNE algorithm was used to assess the effectiveness of removing batch effects in the BLCA cohorts. (B) BREGRS was calculated for each subject in the BLCA cohorts. Based on the median BREGRS level in the training cohort (TCGA-BLCA), all BLCA cases were divided into low- and high-BREGRS subgroups. (C, D) Kaplan-Meier survival (C) and ROC (D) analyses indicated the prognosis value of BREGRS in these cohorts. (E, F) Meta-analyses were conducted to calculate the pooled HRs of BREGRS in these cohorts using the continuous (E) and binary (F) BREGRS. (G) Subgroup analyses of BREGRS were performed to explore its prognostic value in different patient subgroups. (H) Univariate and multivariate Cox regression analyses were conducted to assess the independent prognostic value of BREGRS. t-SNE, t-distributed Stochastic Neighbor Embedding; ROC, receiver operating characteristic; HR, Hazard Ratio