Construction and Analysis of Tissue-Specific Protein-Protein Interaction Networks in Humans

Abstract

We have studied the changes in protein-protein interaction network of 38 different tissues of the human body. 123 gene expression samples from these tissues were used to construct human protein-protein interaction network. This network is then pruned using the gene expression samples of each tissue to construct different protein-protein interaction networks corresponding to different studied tissues of the body. This study is helpful for understanding how human protein interactions change in different tissues. In this way, similar tissues of the body and special functions of each tissue, corresponding to their individualized subnetworks, can be identified. We have calculated graph parameters for studying these protein-protein interaction networks and hubs and non-hubs of the studied protein-protein interaction networks are identified. We found a common subnetwork among protein-protein interaction networks of the studied tissues and a tree of tissue similarities has been constructed. We have also found that average correlation coefficient of hubs in human protein-protein interaction networks obeys a normal-like distribution though it is not possible to separate party and date hubs.