Microarrays might be used for future diagnostic and prognostic purposes. High-density oligonucleotide arrays are promising in this respect. The microarray data consist of intensity files, which are transformed into expression matrices by the application of several mathematical modifications. However, pitfalls regarding data analysis seem to be a critical factor for the impact of this new technology. This article focuses on the data analysis, from raw data file to marker gene lists used to retrieve knowledge about underlying biological processes.