Fig. 1
From: Determining breast cancer histological grade from RNA-sequencing data
Prediction models comparison. ROC curves of CV on five models: GGI, TG-Gene, TG-Iso, SC-Gene and SC-Iso for a the Clinseq data set and b the TCGA data set. AUC estimates and 95 % CI of ROC curves are listed for each model. c Cross-data set validation of multivariate prediction models (TG-Gene, TG-Iso, SC-Gene and SC-Iso). Models were estimated based on the Clinseq data set, and grade in the TCGA data set was predicted. d Predictions of HG2 tumours by five models for all observations in the Clinseq and TCGA data sets. AUC area under the ROC curve, CI confidence interval, CV cross-validation, GGI genomic grade index, ROC receiver operating characteristic, TCGA The Cancer Genome Atlas