randomGODB: Random GO Database
The Gene Ontology (GO) Consortium <https://geneontology.org/> organizes genes
  into hierarchical categories based on biological process (BP), molecular function (MF) and
  cellular component (CC, i.e., subcellular localization). Tools such as 'GoMiner' (see Zeeberg, B.R.,
  Feng, W., Wang, G. et al. (2003) <doi:10.1186/gb-2003-4-4-r28>) can leverage GO to perform
  ontological analysis of microarray and proteomics studies, typically generating a list of
  significant functional categories. The significance is traditionally determined by randomizing
  the input gene list to computing the false discovery rate (FDR) of the enrichment p-value for
  each category. We explore here the novel alternative of randomizing the GO database rather than
  the gene list.
Documentation:
Downloads:
Reverse dependencies:
Linking:
Please use the canonical form
https://CRAN.R-project.org/package=randomGODB
to link to this page.