A triangle based edge scoring of protein interaction network

Abstract

Motivation: Uncovering the mystery of evolutionary mechanism of protein interaction networks has been actively conducted in order to understand interactions of proteins that induce biological processes in organisms. There have been many attempts to solve the mystery by proposing evolutionary models of protein interaction networks. Topological properties of protein interaction networks are mentioned several times and given an important role in these attempts since a validation of suggested models is made through topological properties of known protein interaction networks. While one group of researchers have made efforts to generate current protein interaction networks from some hypothetical infant state of protein interaction networks through suggested evolutionary models, another group of researchers have made efforts to estimate the phylogenetic age of proteins from evolutionary relationships. Recently, these efforts gave rise to the database of phylogenetic age of proteins and this allows many researchers to estimate ages of proteins in their interest easily. Recent studies on Mendelian diseases and cancer suggested that proteins associated with specific diseases populate certain category of the phylogenetic age of proteins. The fact that the topological properties of the protein interaction network have played important roles in the evolution of protein interaction networks tells us that topological properties of protein interaction network and properties of proteins, which is related to the evolution of the protein interaction network, is closely related in some level. As one can see from closeness in terms, the evolutionary model of protein interaction networks and phylogenetic age of proteins are closely related and thus topological properties of protein interactions, which is important in studies of the evolutionary models, can be used to estimate the phylogenetic age of proteins. Besides, the research results on the relationship between diseases and phylogenetic age of proteins motivate us to predict proteins associated to diseases by utilizing topological properties of protein interaction networks. Results: We construct a weighted human protein interaction network from a human protein interaction network which is provided via BioGRID database. The weight of an edge is defined as the number of triangles which contains this edge in the protein interaction network and thus we call this weight as the triangle score. We make comparison between the edge scores of a human protein interaction network given by STRING database and the triangle score. In an attempt to find relationship between the triangle score and properties of proteins that is related to the evolution of protein interaction networks, we make comparison between the triangle score and bit score, which is a measurement of protein sequence similarity. Moreover, we attempt to sieve out self-interacting proteins from the whole human proteins based on the triangle score. In an effort to predict the phylogenetic age of proteins based on the triangle score, firstly, we extract proteins that are incident on an edge that has a high triangle score from the weighted protein interaction network which we constructed with the triangle score. After the extraction, we make inquiries to the ProteinHistorian database to get phylogenetic ages of extracted proteins. Finally, we show that there is a relationship between triangle score and phylogenetic age by comparing the ratio of proteins with each phylogenetic age to whole human proteins and the ratio of extracted proteins with each phylogenetic age to whole extracted proteins. Based on the triangle score, we also attempt to predict disease associated proteins for several diseases. The fact that the topological properties of the protein interaction network have played important roles in the evolution of protein interaction networks tells us that topological properties of protein interaction network and properties of proteins, which is related to the evolution of the protein interaction network, is closely related in some level. As one can see from closeness in terms, the evolutionary model of protein interaction networks and phylogenetic age of proteins are closely related and topological properties of protein interactions, which is important in studies of the evolutionary models, can be used to estimate the phylogenetic age of proteins. Besides, the research results on the relationship between diseases and phylogenetic age of proteins motivate us to predict proteins associated to diseases by utilizing topological properties of protein interaction networks. Results: We construct a weighted human protein interaction network from a human protein interaction network which is provided via BioGrid database. The weight of an edge is defined as the number of triangles which contains this edge in the protein interaction network and thus we call this weight as the triangle score. We make comparison between the edge scores of a human protein interaction network given by STRING database and the triangle score. In an attempt to find relationship between the triangle score and properties of proteins that is related to the evolution of protein interaction networks, we make comparison between the triangle score and bit score, which is a measurement of protein sequence similarity. Moreover, we attempt to sieve out self-interacting proteins from the whole human proteins based on the triangle score. In an effort to predict the phylogenetic age of proteins based on the triangle score, firstly, we extract proteins that are incident on an edge that has a high triangle score from the weighted protein interaction network which we constructed with the triangle score. After the extraction, we make inquiries to the ProteinHistorian database to get phylogenetic ages of extracted proteins. Finally, we show that there is a relationship between triangle score and phylogenetic age by comparing the ratio of proteins with each phylogenetic age to whole human proteins and the ratio of extracted proteins with each phylogenetic age to whole extracted proteins. Based on the triangle score, we also attempt to predict disease associated proteins for several diseases.