Hi I have a count data (basically histone modification data) from a sample. The counts represents number of reads falling into each genomic region. I need to divide my data into low|medium|high based on the number of counts. What method should I use in order to get the most statistically sound result.