Figure 4

Comparison of the distributions of pair-wise network distances between pairs of paralogous protein pairs vs. that between non-paralogous protein pairs in the protein-protein interaction networks of the yeast S. cerevisiae (A) and human (B). Same number of non-paralogous protein pairs as paralogous pairs were randomly picked from the networks. The horizontal axis is the network distance, calculated as the length of shortest path between two proteins in the protein-protein interaction network (in terms of number of proteins in the path; 2 indicates directly connected nodes). The vertical axis is the percentage of gene pairs with corresponding network distance. The p-value of the two distributions in A is 1.886e-7 (Pearson’s χ² test). When the data in B were subdivided, based on the shape of the distributions, into two groups using a boundary of 3.5, the two distributions have a p-value of 0.003 (Pearson’s χ² test).