Skip to main content
Figure 1 | BMC Bioinformatics

Figure 1

From: Graph based fusion of miRNA and mRNA expression data improves clinical outcome prediction in prostate cancer

Figure 1

Comparison of Prediction Error Curves. The figure shows the prediction error curves of CoxBoost models trained on the mRNA and miRNA data. The prediction model was trained with and without the bipartite graph describing the relations between the features. The incorporation of the graph resulted in a reduction of the prediction error. The .632 estimation of the prediction error was used in this plot, averaging over the 500 bootstrap samples. As a reference the prediction error of the Kaplan-Meier estimator is shown.

Back to article page