This column shows the neighborhood of protein interactions surrounding the selected gene. The neighborhood is computed from a genome-wide protein-protein interaction network. The network connects genes if the proteins they encode have been detected to physically interact in high-throughput experiments.
The network distance listed in the column shows the number of interactions between a gene and the selected gene. For example, genes whose protein products have been detected to directly interact are listed with a network distance of one. Genes that share a direct interaction with a protein that also directly interacts with the selected gene are listed with a distance of 2. For example, if the protein of gene X interacts with the protein of gene Y (but not the protein of the selected gene), and Y's protein interacts with the selected gene, then gene X will have a distance of 2.
Because the protein interaction network is highly connected (i.e. most genes are connected to one another by some distance), we limited the size of the neighborhood to contain only those genes within a pre-specified distance (2 by default).
The interaction data were created by the Marc Vidal laboratory, Center for Cancer Systems Biology and Department of Cancer Biology, Dana-Farber Cancer Institute and Department of Genetics, Harvard Medical School.
Rual, J-F. et al. Towards a proteome-scale map of the human protein-protein interaction network. Nature. 2005 Oct 20;437(7062):1173-78.