I have tried to create a bed file for the hg38 reference genome with 50kb region length and 1kb padding. I used the following script to creat the bed file but I am not sure if the output is correct or not.

region_size <- 50000  # 50kb
padding <- 1000  # 1kb

# Initialize an empty list to store BED entries
bed_entries_list <- list()

# Generate BED entries for each chromosome
for (i in 1:nrow(chromosomes)) {
  chromosome <- chromosomes[i, 1]
  size <- as.numeric(chromosomes[i, 2])
  num_regions <- ceiling(size / region_size)

  # Generate BED entries for each region with padding
  for (j in 1:num_regions) {
    start <- (j - 1) * region_size + 1  # 1-based start position
    end <- min(j * region_size + padding, size)  # 1-based end position with padding, capped at chromosome size
    bed_entry <- c(chromosome, start, end)
    bed_entries_list[[length(bed_entries_list) + 1]] <- bed_entry
  }
}

bed_entries_matrix <- do.call(rbind, bed_entries_list)

The bed file I have created looks like this: But I am not sure if it is right way to do this. Then, I want to use this bed file to create CRAM chunks using samtools

chr1    1   51000
chr1    50001   101000
chr1    100001  151000
chr1    150001  201000
chr1    200001  251000
chr1    250001  301000
chr1    300001  351000

I am dealing with WGS data which computation time increases in multiple steps of GATK even per chromosome. Thus, I want very small regions to parallelize the computation.