STATISTICAL MODEL TO ANALYSE THE POTENTIAL BINDING BETWEEN MIRNA AND TARGET MRNA

Abstract

Gene expression regulation is a vital process in the body to ensure that cells produce the correct amount of proteins when they need them. Any disruption to this regulation can lead to serious consequences, including cancer). miRNAs are micro molecules that control gene expression by targeting a mRNA and binding to specific sites within the 3’UTR or the 5’UTR and increase or decrease gene expression. Hence, it’s crucial to predict gene regulatory response in order to be able to control it. Two of the most widely used statistical methods for analyzing categorical outcome variables are linear discriminant analysis and logistic regression. While both are appropriate for the development of linear classification models, the two methods differ in their basic idea. We consider the problem of choosing between the two techniques and set some guidelines for proper choice threw carry out a comparison study between Logistic Regression and Linear Discriminant Analysis to check the performance of both methods in the prediction of gene regulatory response. The dependent variable was the effect of the potential binding between miRNA and target mRNA on gene expression regulation. In order to predict the genetic regulatory response, six features were studied. These features which represent the independent variables were the number of nucleotides in the seed sequence, the G U wobble, the number of G:U binding in the 3'UTR, the number of spaces between complementary bases, the Length of the 3"UTR, the G+C% in the 3"UTR and the hybridization energy.