期刊名称：International Journal of Computer Science and Network Security
印刷版ISSN：1738-7906
出版年度：2019
卷号：19
期号：9
页码：65-75
出版社：International Journal of Computer Science and Network Security
摘要：The study of any complex system in the form of a network structure has always been an efficient approach, being the underlying aspect of graph theory. For the topological and structural network analysis, the concept of identifying and understanding the influential nodes is a beneficial method based on the connectivity of its structure. For this purpose, centrality measures are computed and the elements of the network are ranked through the obtained centrality scores. This method has been widely used for social networks, however, it gained emerging importance in biological networks and different areas of application. In this study, we have computed and compared degree centrality, closeness centrality, betweenness centrality, Katz centrality and PageRank algorithm on a biological network of Saccharomyces Cerevisae (eukaryotic organism) protein interaction. These measures predicted hundreds of important nodes interpreting the essential proteins. The biological significance of our result was sought through established literature. Out of top 30 proteins (i.e. 5 for each measure) we predicted, 29 were found to be highly significant which depicted the fact that absence of these proteins may result in lethality or destruction. Through these findings we concluded that for structural analysis of a complex biological network, centrality measures are proved to be helpful based on the strong prediction of relevant information regarding underlying biological mechanisms. The integration of centrality metrics with the biological knowledge developed an improved index for identification of network essentiality.
