The functional enrichment of the input gene list is evaluated using the well-proven cumulative hypergeometric test. For a single gene list, we are testing numerous functional terms at a time. For example, already >16 000 GO biological process terms are considered for a human gene list. To reduce the amount of false positive findings, g:GOSt performs multiple testing correction. By default we use the original g:SCS (Set Counts and Sizes) correction method introduced back in 2007 (15). Unlike the standard methods, g:SCS considers the dependency of multiple tests by taking into account the overlap of functional terms. We repeated the simulations for verifying the g:SCS method and confirmed that the method still holds the same properties. Namely, g:SCS is more conservative than Benjamini-Hochberg False Discovery Rate but not as strict as Bonferroni correction. Instead of the default g:SCS method user can also choose to apply the Bonferroni correction or the Benjamini–Hochberg False Discovery Rate. It is noteworthy, that in g:GOSt we report only the adjusted enrichment P-values.
