Calculating frequencies of possible combinations of SNP genotypes in a population

Question

[Volka]

Hi all,

I am looking to calculate the frequencies of all possible combinations of a group of SNP genotypes, in a population. This would be done based on the minor allele frequencies acquired in our population data. I just have a question on whether the following is the right way to do it:

If for example, we have three SNPs, their alleles, and their minor allele frequencies, with A2 being the minor allele:

SNP  A1  A2  MAF
SNPA  A   T  0.1
SNPB  C   G  0.3
SNPC  A   G  0.4

And I now have compiled the possible genotype combinations from the three SNPs:

AA-CC-AA
AA-CC-AG
AA-CG-AA
AA-CG-AG
...

and so on for 27 combinations.

In order to calculate the frequencies of each 'profile', would I simply multiply all the allele frequencies? For example, the first AA-CC-AA frequency would be:

(0.9)(0.9) x (0.7)(0.7) x (0.6)(0.6)

While for the AA-CC-AG combination it would be:

(0.9)(0.9) x (0.7)(0.7) x (0.6)(0.4)

Would this be correct?

Thanks!

Answer 1

[Gennady Khvorykh]

I would formalize the task like this. We need to calculate the probability of three independent events happened together. The event is the choice of the genotype at the locus (SNP). The probability is estimated by the genotype frequency. Assuming the Hardy-Weinberg equilibrium is true, the frequency of genotype is p^2 for homozygous and 2pq for heterozygous genotypes, where p and q are the allele frequencies. Therefore, the frequency of AA-CC-AG profile would be as follow:

(0.9)(0.9) x (0.7)(0.7) x 2(0.6)(0.4)

[Volka]

Oh, I see! I made an error with the heterozygous genotype frequencies. Other than that, I understand what you're saying, thanks very much!