Skip to contents

Build genome windows from region strings

Source: R/utilities.R
create_genome_windows.Rd

Parses a vector of region strings — either "chr:start-end" ranges or bare chromosome names — into a GRanges of whole-chromosome or sub-chromosomal windows. Bare chromosome names expand to the full length of that chromosome in the supplied genome (via default_genome_windows() when genome = "hg38"). Overlapping / adjacent regions are merged via GenomicRanges::reduce() before return.

Usage

create_genome_windows(regions, padding = 0, genome = "hg38", add_chr = TRUE)

Arguments

regions

Character vector of region strings. Each element is either "chr1:1000-2000" / "1:1,000-2,000" (commas tolerated) or a bare chromosome name ("chr1", "X", etc.). The "chr" prefix is optional on input and controlled on output by add_chr.

padding

Integer. Base pairs to extend on each side of every region. Clipped so start >= 1. Default 0.

genome

Reference genome for bare-chromosome expansion. Only "hg38" is supported.

add_chr

Logical. If TRUE (default), output seqnames carry the "chr" prefix.

Value

A sorted, non-overlapping GRanges of windows.

Examples

create_genome_windows(c("chr1", "chr2:1000000-2000000"))
#> GRanges object with 2 ranges and 0 metadata columns:
#>       seqnames          ranges strand
#>          <Rle>       <IRanges>  <Rle>
#>   [1]     chr2 1000000-2000000      *
#>   [2]     chr1     1-248956422      *
#>   -------
#>   seqinfo: 2 sequences from an unspecified genome; no seqlengths
create_genome_windows("chr3", padding = 5e5)
#> GRanges object with 1 range and 0 metadata columns:
#>       seqnames      ranges strand
#>          <Rle>   <IRanges>  <Rle>
#>   [1]     chr3 1-198795559      *
#>   -------
#>   seqinfo: 1 sequence from an unspecified genome; no seqlengths