Integration of gene functional diversity for effective cancer detection

Abstract

DNA microarray technology has demonstrated to be an effective methodology for the diagnosis of diseases and cancers by means of expression data classification. Although much research has been conducted during the recent years to apply machine learning techniques for microarray data classification, there are two important issues that prevent the use of conventional machine learning techniques, namely the limited availability of training samples and the existence of various uncertainties. This article presents an integrative classification system, based on the ensemble of machine learning, to integrate the diverse functions of multiple groups of genes in order to achieve a robust microarray data classification. Ensemble learning combines a set of base classifiers as a committee to make more appropriate decisions when classifying new data instances. In order to enhance the performance of the ensemble learning process, the approach presented includes a procedure to select optimal ensemble members based on their classification behaviour. The proposed approach has been verified by three microarray data sets for cancer detection. Experimental results showed that the performance of cancer detection can be much improved by integrating a different subgroup of genes, which suggests, instead of seeking individual gene makers, that a robust cancer detection system can be developed based on integration of related gene groups with diverse functions.