Fig. 3
From: Predicting links between tumor samples and genes using 2-Layered graph based diffusion approach
2 layered graph is constructed using 2 different data source (i) COSMIC methylation data for tumor samples and genes (ii) STRING or BIOGRRID database for genetic interaction. The input is 2-layered graph with missing links between tumor samples and genes. The heat diffusion is applied to the 2-layered graph containing missing links. The output is the final link prediction from the heat diffusion graph. Red node is a Tumor sample node and the green node is a Gene node. The Links between Tumor sample and gene node is represented by directed line whereas links between gene nodes is represented by undirected line. The red line is missing link and dotted black line a predicted link by heat diffusion algorithm