A Bayesian variable selection to classify prognostic cancer biomarkers in time-to-event data using Accelerated Failure Time (AFT) model with the frailty effect is represented in this communication. Before classification, efficient and significant markers from a high-dimensional gene expression dataset needs to be identified. A three-step feature selection method is adopted to select the most relevant markers from high-dimensional data. A threshold value for each selected marker has been obtained through classification using the minimum deviance criterion incorporating AFT model along with the frailty effect. The frailty effect is considered to deal with unobserved heterogeneity in the expression value of the subject for investigating the risk of cancer-related outcomes. A simulation study is performed to validate our proposed methodology. Application of the proposed classification technique is given through the high-dimensional TCGA dataset of breast cancer patients. Our results provide evidence about the effect of classified gene expression values on the survival risk of breast cancer patients.

Gajendra K. Vishwakarma has completed his PhD from Vikram University Ujjain, India. He is working as an Associate Professor of statistics in the Department of Mathematics & Computing, Indian Institute of Technology Dhanbad, India with several years of academic as well as industrial research experience in the field of statistics. His research experience covers both applied as well as theoretical provinces. He has over 150 publications that have been cited over 904 times, and his publication h-index is 25. He is editorial board member of some reputed journals. He received international travel awards from DST Govt of India, World Bank Trust Fund USA. International Biometric Society USA to attend the scientific events. He received Professor P.V. Sukhatme  and Smt Suraj Kali Jain Award of Indian Society for Medical Statistics, India.