How to determine gender by homozygosity (rate?) in X chromosome?

Question

[MatthewP]

Hello, I am reading paper A tutorial on conducting genome‐wide association studies: Quality control and statistical analysis, this is link. At table1 it says:

Can indicate sample mix‐ups. If many subjects have this discrepancy, the data should be checked carefully. Males should have an X chromosome homozygosity estimate >0.8 and females should have a value <0.2.

My question is why thresholds 0.8 and 0.2, any data/paper supports? Thanks a lot!

[Devon Ryan]

Those are generic recommendations, which typically don't have specific papers to reference.

[MatthewP]

Ok, I can image that male contains one X chromosome, so genotype would be 0/0 or 1/1 both can treated as homozygosity. But why female's value so low even though female has 2 X chromosome?

[Devon Ryan]

It can't be over 1, 0.8 is 80%, which is pretty high given multimapping and such.

Answer 1

[okiedokie1208]

Did you figure out those values?