test for randomness of sequencing reads or enrichment

[shawpa]

I have a very large sequencing library and I am looking for a way to test whether the DNA that I sequenced represents the entire genome or whether there is some enrichment of regions. I don't mean enrichment like a capture. This was whole genome shotgun sequencing and I want to test if the whole genome is represented equally. I figure the simplest way to do that is to calculate coverage over defined windows. However, there are a lot of regions in the genome that are not complex (i.e. telomere, centromere, and other repetitive elements). I am wondering if I should filter out reads aligning to repetitive elements first (using repeat mask genome) before I calculate my coverage. I also would like to know if there is some sort of statistical test I can use once I have the coverage to check if there are regions of enrichment. I am not statistically savvy. So far, I have aligned my reads and removed duplicate reads.

[SylvainL]

I don't really know the answer as I am not a statistician neither but I'm interested by the question.

I think I would do a window scanning, count how many reads are in each window, then do a fisher test to see whether a region is enriched or not (comparing reads in this region among the whole chromosome for example, the rest of the genome...). And I would do it with uniquely mapped reads...

I don't know if it would be statistically relevant so I will follow this thread

s.