Why is there so much LoH in the mouse data tutorial?

[thomasveith]

In Numbat's walkthrough for mouse data, there is a lot of CNLoH in the results, I was wondering why is that the case? Is that to be expected for all hybrid mice, for the hybrid mouse used in the walkthrough in particular, or is there some bias in Numbat's algorithm that leads to this output with hybrid mice?

[teng-gao]

Hi,

That's a good question - I don't think this is to be expected in general or specific to hybrid mice. It's likely specific to this tumor's biology - for example, HR-deficient tumors often have widespread LoH across the genome.

https://ascopubs.org/doi/full/10.1200/PO.16.00031

[thomasveith]

Thanks!

I was also wondering if you had any advice on how to choose a value for the genetic distance parameter (cM) used in the vcf_pu call from the walk-through? In my case, the mice I'm working with were bred from C557BL/6 and SKH1.

[teng-gao]

0 is fine there is effectively no phase switch produced by homologous recombination in either of the parents during meiosis (parents are both homozygous in all SNPs).