Genotype Frequency Questions

[Dan_USAAZ]

I have a couple questions in regard to Genotype probabilities. The first relates to whether I am correctly identifying the MAF value and applying it appropriately to determine the genotype frequency. The second question relates to what frequency constitutes a “rare” genotype.

1) Based on the following algorithm provided by Valentijn in an older thread, have I correctly identified and applied the MAF value to determine the Genotype frequencies?

- Algorithm

Minor Homozygous = MAF x MAF
Heterozygous = (1-MAF) x MAF x 2
Major Homozygous = (1-MAF) x (1-MAF)​

- MAF from dbSNP – (Is the MAF .0036 or .0036 divided by 18?)

- Calculation using MAF = .0036

Minor Homozygous = 0.0036 x 0.0036 = .000012 or 1 in 77,160 people
Heterozygous = (1 - 0.0036) x 0.0036 x 2 = .0072 or 1 in 140 people
Major Homozygous = (1 - 0.0036) x (1 - 0.0036) = .9928​

2) Often times, posts in this subforum asking about the potential for a problem SNP will get responses indicating that the SNP is too common to be pathogenic. It would need to be more rare in order to be a problem. My question is what MAF would meet this “rare” criteria? Would the above MAF be considered somewhat rare? Would the heterozygous frequency be considered rare enough to pursue further research into the mutation?

Thanks in advance for any input/corrections.
Dan

 

[alicec]

The MAF value from the 1000 genomes database is 0.0036 - the number before the slash. In the other databases listed, it is slightly different. The number after the slash refers to the number of times the allele was encountered in the database.

Rare criteria for an an allele is not fixed but 1% or less is usually considered to be rare - ie MAF 0.01 or less. Your example is definitely rare.

Your genotype frequency calculations are correct. I get slightly different population incidence.

 

[Dan_USAAZ]

Hi @alicec,
Thanks for your input. It is very much appreciated.

I was hoping you could help me with another question in regard to autosomal dominant/recessive?

I have found a lot of information describing what these mean, such as “the phenotype expressing with single copy of minor allele” and “just being a carrier when recessive”. What I have not been able to learn is how you determine if a heterozygous genotype is autosomal dominant. I found one website that seemed to indicate that if the minor allele is listed first, then the SNP genotype is autosomal dominant and the phenotype may be expressed.

rs####### (A;G)

I have not found any other explanations and this seems a bit simple. Is there any accuracy to this? If not, is it in dbSNP? Other?
Thanks,
Dan

 

[alicec]

What I have not been able to learn is how you determine if a heterozygous genotype is autosomal dominant.

The term just indicates that only one copy of the variant allele is needed to cause disease (heterozygous variant). For a recessive allele, two copies are needed (ie variant must be homozygous).

As with all SNPs you need to look for the research on the SNP. A good place to start for any SNP is dbSNP. This gives links to research and a link to ClinVar where the known clinical significance of the SNP is described.

This would show if a SNP is autosomal dominant. If the SNP hasn't been studied and this info is not given then there is no way of knowing until someone studies it.

I found one website that seemed to indicate that if the minor allele is listed first, then the SNP genotype is autosomal dominant and the phenotype may be expressed.

I have no idea where that idea came from. 23andme reports genotype in alphabetical order.