An efficient framework for mining biological network

The research community is flooded with large set of data such as the genome sequences of various organisms, microarray data and so on, of biological origin. This data-volume is rapidly increasing day by day, and the process of understanding this data is lagging behind the process of obtaining it. There is a need to draw a systematic approach to understanding this using computational method. The rapid progress of biotechnology and bio-data analysis methods has led to the emergence and fast growth of a promising new field: bioinformatics. It is a field having incredible amount of bio-data which needs exhaustive analysis. Biodata is available as, Nucleotide sequences (DNA and RNA sequences), Protein sequences, Genomes and structures in the form of Biological networks (metabolic pathways, gene regulatory network and protein interaction network). In this paper I am presenting a framework to discover frequent patterns and modules from biological networks. From the study of different Biological networks it can be concluded that the best way to analyze and extract the information (frequent functional module) from biological network is through graph mining, since these networks can be modeled into different types of graphs according to the information needs to be extracted. But these graph based mining approach often leads to computationally hard problem due to their relation with sub graph isomorphism. Here I am using graph simplification technique, suitable to biological networks, which make it possible to help the graph mining problem computationally scalable and tractable to very large numbers of networks. So detection of frequently occurring patterns and modules will be a computationally simpler task since the reduction in the effective graph size significantly.