Calculate the p-value of occurrence of an overlap between two sets by chance.
Usage
# S4 method for Venn
enrichment_test(venn, set1, set2, univ = "all",
n = 10000, seed = 42)
Arguments
venn
(Required) A Venn object.
set1
(Required) The name or the index of the set of interest.
set2
(Required) The name or the index of the set to be checked whether
enriched in set1.
univ
(Optional) Population size. Default is "all", implying the union
of all the sets in the Venn object will be used. Another set as the
whole population can be assigned as well.
n
(Optional) Number of randomly generated sets. Default is 10,000 and
minimum is 1,000.
seed
(Optional) An integer passed to set.seed function. It is
used to fix a seed for reproducibly random number generation. Default is
42.
Value
Returns a list containing the probability (Significance) of occurrence
of an overlap between two sets by chance and the number of occurrences
(Overlap_Counts) in randomly generated sets.
Details
This type of analysis can also be performed by hypergeometric test or
Fisher's exact test. Here, the approach is similar to that described in
(Austin et
al., 2016). Briefly, the test is based on randomly generation of sets with
equal size to set1 from the background (universal) set. After creating
n (default is 10,000) random sets, the overlap between these and set2
is calculated to make a null distribution. When this distribution is true,
the probability of seeing an overlap at least as extreme as what was observed
(overlap between set1 and set2) will be returned as the
p-value.