Efficient use of the large data sets generated by gene expression microarray experiments requires computerized data analysis approaches. In this chapter we briefly describe and illustrate two broad families of commonly used data analysis methods: class discovery and class prediction methods. A wide range of alternative approaches for clustering and classification of gene expression data are available. While differences in efficiency do exist, none of the well established approaches is uniformly superior to others. Choosing an approach requires consideration of the goals of the analysis, the background knowledge, and the specific experimental constraints. The quality of an algorithm is important, but is not in itself a guarantee of the quality of a specific data analysis. Uncertainty, sensitivity analysis and, in the case of classifiers, external validation or cross-validation should be used to support the legitimacy of results of microarray data analyses.
Garrett-Mayer, Elizabeth and Parmigiani, Giovanni, "Clustering and Classification Methods for Gene Expression Data Analysis" (December 2004). Johns Hopkins University, Dept. of Biostatistics Working Papers. Working Paper 70.