Overlap score explanation. Only nodes that are connected by an edge in the functional network and the pattern network were used for the overlap score calculation, discounting the initial set used for the pattern discovery. The left panel shows a sample real network, with three edges forming an initial seed of nodes (A,B,C). For one of the patterns discovered using this seed, it also found genes D, E, F, and G, many of which share edges with the functional network. The overlap score in this case is 0.86. In contrast, the rightpanel shows a sample random network, which was chosen to have the same number of nodes (four) as the proposed pattern network. In this case however, only one edge is shared, and the overlap score is 0.44.
