Fads and fallacies in the name of small-sample microarray classification - A highlight of misunderstanding and erroneous usage in the applications of genomic signal processing

摘要

The purpose of this article is to highlight those topics that are believed to have displayed a record of misunderstanding and erroneous usage in the applications of genomic signal processing (GSP). We focus on the related subjects of feature selection, classifier design, and error estimation, which together form a microarray classification pipeline, discussing some of the most common "fads" and "fallacies" regarding classification methods that are routinely applied in the analysis of small-sample microarray data