Takes a BED file with overlapping regions and produces a BedGraph file.  This
can optionally normalize the counts by a given factor.

 See: http://genome.ucsc.edu/goldenPath/help/bedgraph.html
      http://genome.ucsc.edu/goldenPath/help/bigWig.html

Usage: bedutils tobedgraph [-plus | -minus] {-norm N} bamfile

Options:
    -plus             only count reads on the plus strand
                      (default: count all reads)
    -minus            only count reads on the minus strand

    -norm VAL         the count at every position is calculated as:
                      floor(count * VAL).