Plink sex check - why would female calls have more than 0 nonmissing Y genotypes?

Question

[Volka]

Hi all, I am currently learning how to perform GWAS quality control, and I have a question on the sex check step. As I understand it, the sex check using plink with the 'ycount' option would report female calls as ambiguous if they have more than 0 non missing Y genotypes.

This is definitely evident from my plink files, where I have female samples showing about half the number of non missing Y genotype as the male samples, and they are being flagged as ambiguous. My question is, why would the female samples have any Y genotypes in the first place, when females only have chromosome X? Where do these Y genotypes come from? Thanks!

[WouterDeCoster]

Not sure if it fully explains what you see, but chrX and chrY have pseudo autosomal regions.

Answer 1

[lakhujanivijay]

I don't know much about GWAS but I can surely give a analogy from microarray experiments. Sometimes, you may see aberrations in the Y chromosome in case of CGH arrays despite the sample being female.

There is a concept of unique and non-unique probes. The unique probes have the same chromosome number, start and stop coordinates while the non unique probes have different, however, the sequence between the start and the stop is identical and hence the peak. In brief, short identical stretches between chromosome X and Y could be the reason.

However, please wait for expert comments.