Skip to main content
Fig. 17 | BMC Medicine

Fig. 17

From: Statistical analysis of high-dimensional biomedical data: a gentle introduction to analytical goals, common approaches and challenges

Fig. 17

Receiver operating characteristic (ROC) curve that illustrates the predictive performance of a gene signature including 227 genes for the prediction of chemotherapy response in serous ovarian cancer, obtained using the TCGA (The Cancer Genome Atlas) data set. The arrow indicates the sensitivity and specificity values obtained for a selected cutoff value that can serve as a threshold for patient stratification. In this example, the AUC is evaluated on the same data used to train the classifier, so it is likely to be overoptimistic. Source: [185]

Back to article page