dea takes as input the recently built SummarizedExperiment object and calls
DESeq2 to test for each line, i.e. each region, the difference in expression
between the two compared biological conditions. The user can choose to control
all the resulting p-values or only a subset of interest. In the first case
a Benjamini–Hochberg procedure is used to control the False Discovery Rate
(FDR). In the second case, the user specifies selection criteria for the
regions, e.g. a threshold on the absolute Log2-FC per-regions. Then, a
post-hoc procedure is used to control the joint Family wise error Rate
(jFWER) on the selected subset of regions.
Usage
dea(
SExp,
design = ~condition,
sizeFactors = NA,
significanceLevel = 0.05,
predicate = NULL,
postHoc_significanceLevel = 0.05,
postHoc_tdpLowerBound = 0.95,
verbose = TRUE
)Arguments
- SExp
The
SummarizedExperimentobject returned bycounting().- design
A
formulaobject or aMatrixobject. Passed on to DESeq2::DESeqDataSet.- sizeFactors
A vector of
Double. Sample-specific size factors.- significanceLevel
A
Double. The significance cutoff on q-values.- predicate
A predicate function that returns a single TRUE or FALSE if the region on which it is applied meets the conditions defined in the predicate. The criteria used in the predicate have to be defined for each regions in mcols(SExp).
- postHoc_significanceLevel
A
Double. The significance level of the test procedure (See sanssouci::posthocBySimes).- postHoc_tdpLowerBound
A
Double. The minimum true positive proportion on the returned set of rejected regions.- verbose
A
Logical. Should all the operations performed be displayed ?