I have NGS data (DNAse seq, chip-seq, rna-seq) and I was wondering if there was a quick way to see if there seemed to be evidence of chromosomal bias in the read mapping (im not looking directly for cnv, just evidence for it and possibly mapping bias)

The naive approach would be (assuming all reads are equal length)

[# of mapped reads on chromosome]/[base pairs in chromosome]

However, # of base pairs in chr include the Ns in centromere as well as repetitive unmappable regions (for 36bp reads) so is there a way to find out how much of each chromosome in Ns and unmappable so I can subtract it from the above denominator?