Journal Article

A Comparison of Logistic Regression and Linear Discriminant Analysis in the Understanding of Gene Regulatory Response

Abd El Alim, Mamdouh, Aal, Medhat Adel, Etman, Nihal Aly

2021

Overview

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 LDA and logistic regression. While both are appropriate for the development of linear classification models, i.e. models associated with linear boundaries between the groups. Nevertheless, the two methods differ in their basic idea. LDA makes more assumptions about the underlying data. It is therefore reasonable to expect LDA to give better results in the case when the normality assumptions are fulfilled, but in all other situations LR should be more appropriate. However, in practice, the assumptions are nearly always violated; therefore, we try to check the performance of both methods with simulations. Previously (In our last paper) we have studied gene regulatory mechanisms using Logistic Regression. In this paper, we present a simulation study between Logistic Regression and LDA in the prediction of gene regulatory response.

• Publisher: جامعة عين شمس - كلية التجارة
• Subject: الاستجابة التنظيمية للجينات / الانحدار اللوجستي / التعبير الجينى / النماذج الإحصائية / تحليل التمييز الخطي / تنظيم التعبير الجينى