Comparative Analysis of Transcriptional Regulatory Network using Machine Learning Approach for Colon Cancer

Abstract

The main goal of this paper is to formulate the gene regulatory network and focus on its relationship between the identified genes using the data provided by micro array technology in order to calculate and analyze the spread of cancer in human beings. A gene regulatory network is a collection of molecular regulators that interact with each other and this network describes their interactions and identifies the genes responsible for the disease. With cancer being one of the most dreaded disease and responsible for countless deaths it is important to bring about a proposed system that reconstructs a cancer specific gene regulatory network. The said methodology is done using machine learning algorithm. In this study, the information obtained from the gene expression of colon cancer is used for identifying the significant genes by a two phase strategy called the t-testing and fold change. In the following step, Pearson correlation coefficient and linear regression are used to target those significant genes and their hub nodes for further diagnosis. The obtained results have been validated and checked for the accuracy and have proven to be correct and highly beneficial.