A DECISION TREE-BASED CLASSIFIER COMPARES THREE DATA ANALYSIS METHODS FOR THE IDENTIFICATION OF MIRNAS ASSOCIATED WITH EARLY-STAGE LUNG CANCER

Abstract

Lung cancer is a major global health problem, with high mortality rates and increasing incidence in developing countries and among women. MicroRNAs (miRNAs) have emerged as potential biomarkers for lung cancer diagnosis and treatment due to their stability, abundance, and easy detection in tumor tissues and body fluids. However, there is no consensus on the most suitable statistical method for identifying differentially expressed miRNAs (DE-miRNAs) in large datasets. In this study, we compared the performance of EdgeR, DESeq2, and the Wilcoxon-Mann-Whitney test in identifying deregulated miRNAs in lung adenocarcinoma using a large miRNA sequencing dataset from The Cancer Genome Atlas. We focused on miRNAs associated with early-stage disease to identify potential biomarkers for disease detection in high-risk patients. Our results demonstrate the usefulness of applied computational mathematics/statistics in improving miRNA analysis using large cancer datasets and contribute to the identification of clinically applicable biomarkers for lung cancer diagnosis and treatment.