Why is Chromosome 19 unreliable?

[ENT Fellow]

Hi there,

I'm doing low-coverage CNV-seq on illumine HiSeq 2500. I've been told that chromosome 19 is not trustworthy but am really struggling to find out why? Can anyone explain or throw me a reference I can read on this?

Many thanks

[GenoMax]

What kind of "not trustworthiness" are you referring to? It does appear that Chr19 has the highest number of alternate reference loci: http://www.ncbi.nlm.nih.gov/projects...urrent-regions

Issues with chromosome 19: http://www.ncbi.nlm.nih.gov/projects...filters=chr:19

[ENT Fellow]

Chromosome 19

Many thanks for your reply.

Well I was told that when analysing my data not to trust gains/losses that are in centromeric/telomeric regions (CNAs are overrepresented here) and to ignore Chr 19 as apparently these are not reliable. Is that a reflection of this high incidence of alternate reference loci you mentioned? Also why is this the case - is it a particularly repetitive chromosome?

[mastal]

I have just been reading a comparative genomics paper, and apparently human chr 19 is particularly 'gene rich'.

[saraki]

centromeric/telomeric regions tend to be highly repetitive regions which is why you might run into overrepresentation of CNAs. 19 does have a very high gene:bp length ratio ("gene rich") but I don't know why that would make it unreliable for CNAs?

[henry.wood]

I think the high number of alternate reference loci is a fairly plausible explanation. Looking at all the alternatives, for chr19, in total, they add up to about a quarter of the length of the chromosome. By comparison, for chr18 is is 0.7% and for chr20 it is 2%. It's not quite as simple as that, but it does suggest that there are a lot of different ways of arranging chr19. If you are doing copy number and using a pooled control sample - i.e. not the same person as the normal, there could be quite a difference in how well your test and control samples align to the main chr19 reference, which will make the apparent copy number of chr19 quite variable from person to person.