Genes and their products typically act in multiple pathways [2], and each role is potentially important to a disease or treatment mechanism. As a result, analyses can expect to have some degree of pathway overlap. However, high pathway overlap can obscure the true source of an association signal. While this problem can exist with any pathway analysis, Gene Ontology (GO) annotations are particularly susceptible due to the database’s large, hierarchical structure [27]. Some studies have restricted analysis of GO terms to certain levels in the hierarchy [13, 28], while a new Bayesian method incorporates the structure of the hierarchy as prior information into its pathway association metric [29]. However, users of these approaches should be aware that the information content at particular GO levels is unpredictable [30]. Pathway overlap can also be addressed during post-analysis to prioritize related pathways for further exploration. Extant strategies include hierarchical clustering in a study of breast cancer [4], overlap-based network creation in the visualization tool Enrichment Map [31], and the listing of overlapping pathways alongside results in the analytical software PARIS [32].
