Saket Navlakha and Carl Kingsford, The power of protein interaction network for associating genes with diseases, Bioinformatics 26(8), 2010
We assessed the utility of physical protein interactions for determining gene-disease associations by examining the performance of seven recently developed computational methods(plus several of their variants). We found that random-walk approaches individually outperform clustering and neighborhood approaches.
Typically, a disease is associated with a linkage interval on the chromosome if single nucleotide polymorphism (SNPs) in the interval are correlated with an increased susceptibility to the disease. These linkage intervals define a set of candidate disease-causing genes. Genes related to the same disease are also known to have protein products that physically interact. A class of computational approaches have recently been proposed that exploit these two sources of information-physical interaction networks and linkage intervals-to predict associations between genes and diseases.
Although random walk approaches are superior to clustering and neighborhood approaches, we showed that all methods make unique predictions and can be used together to increase performance.