Abstract
BackgroundWith DNA microarray data, selecting a compact subset of discriminative genes from thousands of genes is a critical step for accurate classification of phenotypes for, e.g., disease diagnosis. Several widely used gene selection methods often select top-ranked genes according to their individual discriminative power in classifying samples into distinct categories, without considering correlations among genes. A limitation of these gene selection methods is that they may result in gene sets with some redundancy and yield an unnecessary large number of candidate genes for classification analyses. Some latest studies show that incorporating gene to gene correlations into gene selection can remove redundant genes and improve classification accuracy.ResultsIn this study, we propose a new method, Based Bayes error Filter (BBF), to select relevant genes and remove redundant genes in classification analyses of microarray data. The effectiveness and accuracy of this method is demonstrated through analyses of five publicly available microarray datasets. The results show that our gene selection method is capable of achieving better accuracies than previous studies, while being able to effectively select relevant genes, remove redundant genes and obtain efficient and small gene sets for sample classification purposes.ConclusionThe proposed method can effectively identify a compact set of genes with high classification accuracy. This study also indicates that application of the Bayes error is a feasible and effective wayfor removing redundant genes in gene selection.
Highlights
In this article we develop a novel gene selection method based on the Bayes error
Considering the promising aspects of the Bayes error, we propose in this study an approach, BBF (Based Bayes error Filter), for gene selection
The experimental results show that our proposed method can 1) reduce the dimension of microarray data by selecting relevant genes and excluding the redundant genes; and 2) improve or be comparable to the classification accuracy compared with other earlier studies
Summary
Pattern Recognition, statistical, structural and neural approaches. With DNA microarray data, selecting a compact subset of discriminative genes from thousands of genes is a critical step for accurate classification of phenotypes for, e.g., disease diagnosis. A limitation of these gene selection methods is that they may result in gene sets with some redundancy and yield an unnecessary large number of candidate genes for classification analyses. In the presence of thousands of genes in microarray experiments, it is common that a large number of genes are not informative for classification because they are either irrelevant or redundant [19]. Strong relevance indicates that the gene is always necessary for an optimal subset and cannot be removed without affecting the classification accuracy. Based on the relevance definitions of genes, in classification analyses, a redundant gene is the one (which may be useful for classification analyses in isolation) does not provide much additional information if another informative gene is already present in the chosen gene subset
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
