4
$\begingroup$

Say both parent and child are genotyped for all SNPs. In this setup we are only looking at variant positions between one parent and child - so neither parent nor child are homozygous reference (no AA/AA) in the millions of positions we are studying.

Assume all variants are biallelic and there is no de novo mutation (no AA/BB or BB/AA)

P    C
AA   AB
AB   AA
AB   AB
AB   BB
BB   AB
BB   BB

What is the fraction of AB/AB among these possibilites?

We are looking at all positions in the genome so the minor allele frequency (MAF) of B allele varies. You can choose whatever B frequency you like.

$\endgroup$
6
  • 2
    $\begingroup$ I'm a little confused: are the genotypes in the rows representing different loci or just different potential, non AA/AA scenarios of one locus? $\endgroup$
    – Atticus29
    Feb 5, 2013 at 7:16
  • $\begingroup$ Potentials. I am going to go with 1/6, regardless of B's MAF. $\endgroup$ Aug 4, 2014 at 13:51
  • $\begingroup$ I don't understand the question. You say so neither parent nor child are homozygous while in your example you have examples where one of the two or even the two are homozygous. Then it is unclear what you mean by no AA/BB or BB/AA. I suppose the / notation separate one parent genotype from the child genotype. I don't know if you're assuming sexual reproduction and talk only about one parent or not. I think you want to assume that all A alleles are at the same frequency. In the same time you seem to say the MAF varies, however you say the "MAF of B" varies which makes little sense. $\endgroup$
    – Remi.b
    Aug 19, 2016 at 14:21
  • $\begingroup$ For the above reasons, I am voting to close as unclear (four years after the question has been posted :) ) $\endgroup$
    – Remi.b
    Aug 19, 2016 at 14:22
  • $\begingroup$ by no AA/BB or BB/AA I mean if the one parent we know about is AA we never BB in the child. $\endgroup$ Aug 19, 2016 at 18:45

0

Browse other questions tagged .