Coerce a column to a GRanges object from a rectangular object
colToRanges(x, ...)
# S4 method for class 'DataFrame'
colToRanges(x, var, seqinfo = NULL, ...)
# S4 method for class 'GRanges'
colToRanges(x, var, ..., keep_metadata = TRUE)
# S4 method for class 'data.frame'
colToRanges(x, var, seqinfo = NULL, ...)A data-frame or GRanges object containing the column to coerce
Used to pass arguments to lower-level functions
The name of the column to coerce
A seqinfo object to be applied to the new GRanges object. Ignored if the column is already a GRanges object
logical(1) If the original object is a GRanges object, retain all metadata from the original ranges in the replaced ranges
A GenomicRanges object
Take a data.frame-like object and coerce one column to a GRanges object,
setting the remainder as the mcols.
A particularly useful application of this is when you have a GRanges object
with one mcol being a secondary GRanges object.
Alternatively, if you have a data.frame with GRanges represented as a character column, this provides a simple method of coercion. In this case, no Seqinfo element will be applied to the GRanges element.
set.seed(73)
x <- GRanges(c("chr1:1-10", "chr1:6-15", "chr1:51-60"))
seqinfo(x) <- Seqinfo("chr1", 60, FALSE, "Example")
df <- data.frame(logFC = rnorm(3), logCPM = rnorm(3,8), p = 10^-rexp(3))
mcols(x) <- df
gr <- mergeByCol(x, col = "logCPM", pval = "p")
colToRanges(gr, "keyval_range")
#> GRanges object with 2 ranges and 7 metadata columns:
#> seqnames ranges strand | n_windows n_up n_down logCPM
#> <Rle> <IRanges> <Rle> | <integer> <integer> <integer> <numeric>
#> [1] chr1 1-10 * | 2 1 1 7.87263
#> [2] chr1 51-60 * | 1 1 0 8.14117
#> logFC p p_fdr
#> <numeric> <numeric> <numeric>
#> [1] -0.1450469 0.296375 0.440965
#> [2] 0.0937974 0.440965 0.440965
#> -------
#> seqinfo: 1 sequence from Example genome