Essential Gene Identification for a Microarray Data of Yersinia Pestis

This paper is motivated by a DNA micro array data obtained from a genome-wide mutation library for the bacterium Yersinia Pestis. The purpose of this study is to identify essential genes for the bacterium Yersinia Pestis. The data set contains more than four thousands genes and each gene has different number of observations with unequal number of probe observations. We propose a feature selection method for the representing three probes and a new gene level adjusted multiple statistical test to handle the problem of unequal number of observations. The proposed method is compared with two other methods based on Behrens-Fisher method and Hotelling t-square method. Our results show that our proposed method is more suitable among the three for identifying essential genes using the DNA micro array data.