Breast cancer prediction and categorization in the molecular era of histologic grade
Publication Name
Multimedia Tools and Applications
Abstract
Breast cancer is the second utmost common cancer among women. In this research study, two feature selection methods, namely Consistency first Search-Best First search (CFS-BFS) and Consistency-Best First Search (Consistency-BFS) are used to find out the significant biomarkers for breast cancer. The feature selection methods pick a small count of important and relevant genes providing a higher degree of accuracy. The count of biomarkers selected in Consistency-BFS is less in comparison to CFS-BFS method. Three common identified biomarkers’ genes are recognised to help in diagnosis and prognosis of breast cancer, namely SLC39A6, ESR1, and CDC20 are selected on the basis of histologic-grade and molecular subtype of breast cancer. These three genes identified are even found in PAM50 and judged for survival variances using Kaplan-Meier Survival Model. The result suggested that overexpression of SLC39A6, ESR1, and CDC20 proves to be an influencing factor for the poor prognosis of patients suffering from breast cancer. The proposed method is successful in creating a prognostic gene signature in forecasting the survival likelihood of patients.
Open Access Status
This publication is not available as open access