Table 2 Several datasets were used to verity the results, corresponding to different experimental setups and conditions. In each case, only the 3000 genes with highest variance were selected for further analysis. Note that this implies that the set of selected genes is (slightly) different for each dataset. For details on the preprocessing and normalization of the data see Appendix. In the following, all shown results will refer to the dataset of Spellman et al.[24]

From: Validation and functional annotation of expression-based clusters based on gene ontology