Calculating non-integer copy number variations

[Dingersrun]

I have 2 cell lines that have unusual karyotypes. One is a mixture of diploid and tetraploid cells (in average it is a triploid) and the other has chromosome number of ranging from 85 to 102.

I was wondering the following:

1. what kind of damage will I have if I use non-standard diploid samples as the reference?
2. is there a way that I can calculate the non-integer copy number variation? In my case, The average copy number of a cell line can very well be 3.8 as it is a mixture.

Another question about the sex of the sample:
I got the following output from a male sample using a male reference sample. and I specified --male-reference for the sample.
Relative log2 coverage of chrX=-1.23, chrY=-4.56 (maleness=14.1 x 0.0678 = 0.954) --> assuming female

Could you please kindly explain the output above?

Thank you very much,
Dingwen

[etal]

Sex inference has since been overhauled; 2023 output no longer applies. Closing as resolved — please reopen with a reproduction on a current release if you still encounter it.