%0 Journal Article
%T Microarray-based gene set analysis: a comparison of current methods
%A Sarah Song
%A Michael A Black
%J BMC Bioinformatics
%D 2008
%I BioMed Central
%R 10.1186/1471-2105-9-502
%X Each of six gene set analysis methods was applied to both simulated and publicly available microarray data sets. Overall, the various methodologies were all found to be better at detecting gene sets that moved from non-active (i.e., genes not expressed) to active states (or vice versa), rather than those that simply changed their level of activity. Methods which incorporate correlation structures were found to provide increased ability to detect altered gene sets in some settings.Based on the results obtained through the analysis of simulated data, it is clear that the performance of gene set analysis methods is strongly influenced by the features of the data set in question, and that methods which incorporate correlation structures into the analysis process tend to achieve better performance, relative to methods which rely on univariate test statistics.Gene expression microarrays provide a snapshot of gene transcript abundance on a genomic scale, and are a popular tool for detecting differences in gene activity across biological samples. While many currently used approaches for analyzing microarray data focus on detach-detecting changes in activity on a per-gene basis [1,2], biological processes are generally the result of interactions between multiple genes (i.e., a gene pathway or a network), and are thus not easily interrogated via single gene methods. To facilitate such analyses, statistical methods have been developed which focus on detecting changes in groups of functionally related genes, thus allowing additional biological information to be incorporated into the analysis process [3-9]. Despite commonality of purpose, these methods take quite different approaches to achieving their goal, and can thus produce differing results when applied to the same data set. Recently Goeman [10] discussed the assumptions underlying these differing methodologies. Here we attempt to examine the performance differences between a number of methods through the analysis of real
%U http://www.biomedcentral.com/1471-2105/9/502