read.bedMethyl: Parsing bedMethyl output from modkit pileup.
In hansenlab/bsseq: Analyze, manage and store whole-genome methylation data

read.bedMethyl

R Documentation

Parsing bedMethyl output from modkit pileup.

Description

Parsing bedMethyl output from modkit pileup.

Usage

read.bedMethyl(files,
               loci = NULL,
               colData = NULL,
               rmZeroCov = TRUE,
               strandCollapse = TRUE,
               BPPARAM = bpparam(),
               BACKEND = NULL,
               dir = tempfile("BSseq"),
               replace = FALSE,
               chunkdim = NULL,
               level = NULL,
               nThread = 1L,
               verbose = getOption("verbose"))

Arguments

`files`	The path to the files created by running modkit pileup, one sample per file. See the methods section of [link to preprint] for validated output.
`loci`	`NULL` (default) or a `GenomicRanges` instance containing methylation loci (all with width equal to 1). If `loci = NULL`, then `read.bedMethyl()` will perform a first pass over the bedMethyl files to identify candidate loci. If `loci` is a `GenomicRanges` instance, then these form the candidate loci. The candidate loci will be collapsed if `strandCollapse = TRUE`.
`colData`	An optional `DataFrame` describing the samples. Row names, if present, become the column names of the BSseq object. If `NULL`, then a `DataFrame` will be created with `files` used as the row names.
`rmZeroCov`	A `logical(1)` indicating whether methylation loci that have zero coverage in all samples should be removed. Default setting is `rmZeroCov = TRUE`
`strandCollapse`	A `logical(1)` indicating whether strand-symmetric methylation loci (i.e. CpGs) should be collapsed across strands.
`BPPARAM`	An optional `BiocParallelParam` instance determining the parallel back-end to be used during evaluation.
`BACKEND`	`NULL` or a single string specifying the name of the realization backend. Currently, the backend is not supported for downstream applications.
`dir`	Only applicable if `BACKEND == "HDF5Array"`. The path (as a single string) to the directory where to save the HDF5-based BSseq object.
`replace`	Only applicable if `BACKEND == "HDF5Array"`. If the directory `dir` already exists, should it be replaced with a new one?
`chunkdim`	Only applicable if `BACKEND == "HDF5Array"`. The dimensions of the chunks to use for writing the data to disk.
`level`	The compression level to use for writing the data to disk.
`nThread`	The number of threads used by `fread` when reading the `files`.
`verbose`	A `logical(1)` indicating whether progress messages should be printed (default `TRUE`).

File formats

The format of each file should be similar to the examples in [link to preprint]. Files ending in .gz, .bz2, .xz, or .zip will be automatically decompressed to tempdir().

Supported file formats

Modkit bedMethyl files from modkit pileup. For downstream likelihood functions we recommend running modkit pileup on output from bam files modification/basecalled using a CG context model and not using a reference genome for pileup.

Unsupported file formats

Other types of output.

One-based vs. zero-based genomic co-ordinates

The genomic co-ordinates of bedMethyl files are zero-based. Since Bioconductor packages typically use one-based co-ordinates, the co-ordinates from the bedMethyl files are converted to one-based in the BSseq object.

Author(s)

Søren Blikdal Hansen (soren.blikdal.hansen@sund.ku.dk)

Examples

# Example: Reading bedMethyl files included in the bsseq package
# Paths to example bedMethyl files in the package's extdata directory
infiles <- c(system.file("extdata/HG002_nanopore_test.bedMethyl.gz",
                         package = "bsseq"),
             system.file("extdata/HG002_pacbio_test.bedMethyl.gz",
                         package = "bsseq"))

# Run the function to import data
bsseq <- read.bedMethyl(files = infiles,
                        colData = DataFrame(row.names = c("test_nanopore", 
                                                          "test_pacbio")),
                        rmZeroCov = FALSE,
                        strandCollapse = TRUE,
                        verbose = TRUE)

# View the resulting BSseq object
bsseq

hansenlab/bsseq documentation built on June 12, 2025, 7:42 p.m.