How To Use GOstats Testing Gene Lists for GO Term Association

TheGOstats package has extensive facilities for testing the association of Gene Ontology (GO) The Gene Ontology Consortium (2000) terms to genes in a gene list. You can test for both over and under representation of GO terms using either the standard Hypergeometric test or a conditional Hypergeometric test that uses the relationships among the GO terms for conditioning (similar to that presented in Alexa et al. (2006)). In this vignette we describe the preprocessing required to construct inputs for the main testing function, hyperGTest, the algorithms used, and the structure of the return value. We use a microarray data set (Chiaretti et al., 2004) from a clinical trial in acute lymphoblastic leukemia (ALL) to work an example analysis. In the ALL data, we focus on the patients with B-cell derived ALL, and in particular on comparing the group with ALL1/AF4 to those with no observed cytogenetic abnormalities. To get started, load the packages needed for this analysis: