Figure 3

Results of the genome-wide cross-validation for disease gene prioritization. (A) The conditional Mean-Hitting-Time (M _p (g)) is calculated by Equation (5). Results showed that genes with high modularity levels to the other adjacent nodes with small M _p (g) values will be more likely to be successfully prioritized during the validation. (B) The receiver operating characteristic (ROC) curves of the genome-wide leave-one-out cross-validation. The horizontal coordinates (1-Specificity) refer to values of θ _R , while the vertical coordinates (Sensitivity) refer to the true-positive rate corresponding to θ _R . The red solid line denotes the inference of modular CIPHER-HIT based on the nodes in , i.e. only the adjacent node u of M _p (u) <θ _M = 0.3 are used for inference. The dashed lines both denote the inference based on the nodes in , i.e. only the adjacent node u of M _p (u) ≥ θ _M = 0.3 are used for inference, where the blue dashed line denotes results from the random walk with restarts (RWR) method, and the red dashed line denotes results from the CIPHER-HIT method.