FunNet is an original integrative tool for exploring transcriptional interactions in microarray gene expression datasets. The analytical approach implemented in FunNet relies on knowledge extracted from public annotation databases to improve the biological relevance of the modular interaction patterns identified in co-expression networks. FunNet’s algorithmic core is implemented as an R package freely downloadable from one of the CRAN mirrors. This web site provides a user-friendly graphical interface designed to promote FunNet’s open use by the community. Details about the format of the data files and various parameters and analytical options are provided in FunNet’s tutorial.