Biometrics and Bioinformatics

Abstract

Gene selection is an important issue in microarray data processing. In this work, propose a capable method for selecting relevant genes. This work aim at finding the smallest set of genes that can ensure highly accurate classification of cancers from microarray data by using supervised machine learning algorithms. Initially utilized spectral biclustering to achieve the best two eigenvectors for class partition. Then gene combinations are chosen based on the similarity among the genes and the best eigenvectors. Proposed simple yet very effective method involves two steps. In the first step, choose some important genes using a feature importance ranking scheme. In the second step, test the classification capability of all simple combinations of those important genes by using a good classifier. This work demonstrates semi-unsupervised and T-Score gene selection method using two microarray cancer data sets, i.e., the lymphoma and leukemia data sets. Experimental result shows proposed method is able to identify a single gene which leads to predictions with very high accuracy.

Keywords

Full Text