Skip to main content
. 2014 Nov 12;9(11):e106319. doi: 10.1371/journal.pone.0106319

Figure 2. Comparison of the Area Under the Curves (AUCs) of the re-constructed networks from the 231-gene network with a noise value of 0.25 and different sample sizes (20, 50 or 100) for SPACE (a.), GeneNet (b.), and WGCNA (c.).

Figure 2

In these plots, the y-axis shows the performance of the reconstructed network, measured by the AUCs; a horizontal line is drawn to represent the AUC of the non-bootstrapped reconstruction (a single reconstruction using all available samples). The x-axis represents the number of iterations in the bootstrapping process. Points below the horizontal line represent a loss in accuracy of the reconstructed networks, and points above the horizontal line represent a gain of AUC (i.e., an increase in model performance).