I have a toy dataframe where the first 3 columns signify chromosome location, and the fourth column shows whether consecutive rows overlap.

df <- data.frame(
  chrom = c("chr1", "chr1", "chr1", "chr1", "chr1", "chr1"),
  start = c(10, 20, 25, 30, 90,  100),
  end = c(20, 30, 38, 40, 120,  200),
  count = c("no_overlap", "overlap", "overlap", "overlap", "overlap", "overlap")
)

Now I’d like to create a new dataframe in which the region breaks if there is an overlapping region. Such as, from 2nd read (20 30) becomes (20 24 1), where 1 is the count (end is the subtracting 1 from the next start, 24=25-1). However, the 3rd read (chr1 25 38 2) location would be unchanged but count will become 2 as it overlaps with the previous read (25-30) and also with subsequent read (31-38). Similarly for other reads as well.

This is our anticipated output.

chrom   start end count 
 <chr>   <dbl> <dbl> <int>     
  chr1       10    20 1
  chr1       20    24 1 
  chr1       25    38 2
  chr1       39    40 1
  chr1       90    99 1
  chr1      100   120 2
  chr1      121   200 1   

Hopefully, I’m able to convey my message properly.