Figure 2

Shared neighborhood scoring algorithm. SN score can be calculated by summing up "Shared Nodes Count" and "Shared Nodes Weight". First, “Shared Nodes Count” is defined as the number of shared nodes to consider the effect of direct connectors. Similary, “Shared Nodes Weight” is defined as the product of each weight of links bridging two end nodes to trace the effect of indirect neighbors (Bottom right). Here weight is a measure for connecting probability of each pair. For all possible pairs of the network, firstly weight 1 is assigned to each connected pairs directly linked between two nodes. Weight for unconnected pairs is assigned the connection probability, the fraction of directly connected pairs among the total number of pairs having the given “Shared Nodes Count”. For example, (1) the weight product of two indirect links (2), (3) weight of the upper (direct link / indirect link) multiplied by weight of the lower one (indirect link / direct link). The sum of (1), (2), and (3) is "Shared Weight" of the unconnected pair (protein, drug) (bottom left).