Identification of Critical Genes in Autism Disorder Using Centrality Measures

Abstract

Learning of the protein and pathway interactions for the implicated genes is required for a enhanced understanding of the basic pathogenic mechanisms of autism. In Protein-protein interaction network, proteins are the vertices and their edges as interaction among the proteins. Mutations in a protein may change its functionality. Thus it may affect the interactions with its neighbor which results malfunction. Therefore, it is of interest to use various graph centrality measures integrated with the genes associated with the Autism human network for discovery of potential drug targets. The data set that we used is the data source of Jensenlab (Novo Nordisk Foundation Center for Protein Research, Denmark) for the analysis of Autism disorder network. We have extracted 1135 genes involved in Autism disease progression using text mining, 19 genes from Experimental evidence Jensenlab disease database and 345 genes from New drug targets database. Finally we have constructed Protien-Protien Interaction (PPI) network with 54 proteins and 74 interactions after eliminating parallel edges, self-loops. Thus we have identified the genes that are importantly associated Autism Disorder using network centrality measures. In this paper, we also worked out clustering coefficient, which is usually used to study social engineering networks and protein-protein interaction networks. Thus we listed the most influential genes belonging to Autism Disorder which are potential drug targets.