.calcMotifEnrichment | R Documentation |
Given motif counts, foreground/background labels and
weights for a set of sequences, calculate the enrichment of each motif
in foreground compared to background. This function is called by
calcBinnedMotifEnrR()
for each bin.
The default type of test is "fisher"
, which is also what
Homer
uses if "-h" is specified for a hypergeometric test.
Alternatively, a binomial test can be used by test = "binomial"
(what Homer
does by default). Using Fisher's exact test has
the advantage that special cases such as zero background counts are
handled without ad-hoc adjustments to the frequencies.
For test = "fisher"
, fisher.test
is used with
alternative = "greater"
, making it a one-sided test for enrichment,
as is the case with the binomial test.
.calcMotifEnrichment(
motifHitMatrix,
df,
test = c("fisher", "binomial"),
verbose = FALSE
)
motifHitMatrix |
matrix with 0 and 1 entries for absence or presence of motif hits in each sequence. |
df |
a |
test |
type of motif enrichment test to perform. |
verbose |
A logical scalar. If |
a data.frame
containing the motifs as rows and the columns:
: the motif name
: the log p-value for enrichment (natural logarithm).
If test="binomial"
(default), this log p-value is identical to
the one returned by Homer.
: the sum of the weights of the foreground sequences that have at least one instance of a specific motif (ZOOPS mode).
: the sum of the weights of the background sequences that have at least one instance of a specific motif (ZOOPS mode).
: the total sum of weights of foreground sequences.
: the total sum of weights of background sequences.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.