Schematic representation of the ToppCluster pipeline. Multiple genelists are named by the user and submitted through the ToppCluster interface. 17 different categories of annotations are available as of April 2010. The user can choose the categories to be included, the P-value cutoff, a method of correction for false discovery and the type of output. Functional enrichment analysis is done on each gene list and the results are collated into a single matrix. Significance P-values are log transformed (−log10) into scores and the genes per list that intersect each enriched feature are placed into a separate column. The results are delivered in the chosen output format.