Question about A+D model

[JinRenelf]

Hi,

Thank you developing this fantastic R package, which is very useful to simulate genetic data.
I have a question about the A+D mode, does this mean a single SNP can has both additive and dominant effect on a continuous trait?

[Foredawnbio]

Thank you for using SIMER software. Your understanding is correct. In A+D mode, a single SNP has both additive effect and dominant effect.

[JinRenelf]

Thank you for answering my question.

For the dominant effect at a locus.
Does this represent the deviation of the heterozygote phenotype from the midpoint of the two homozygote phenotypes.
For example:
Homozygous reference (AA): Effect = 0.
Heterozygous (Aa): Effect = 𝑑.
Homozygous alternate (aa): Effect = 0.

[Foredawnbio]

Hi，

The situation you described is a correct way to assign dominant effects. However, if the additive-dominant interaction effect is taken into account, this method cannot guarantee that the additive effect and the dominant effect are orthogonal. Therefore, SIMER adopts another coding method. For example:
For additive effect,
Homozygous reference (AA): Effect = 1a.
Heterozygous (Aa): Effect = 0a.
Homozygous alternate (aa): Effect = -1a.
For dominant effect,
Homozygous reference (AA): Effect = -0.5d.
Heterozygous (Aa): Effect = 0.5d.
Homozygous alternate (aa): Effect = -0.5d.

(1a)(-0.5d) + (0a)(0.5d) + (-1a)(-0.5*d) = 0 --> the additive effect and the dominant effect are orthogonal.

reference: Kao C and Zeng Z (2002) doi:10.1093/genetics/160.3.1243.

[JinRenelf]

Thank you for answering my question. It is very helpful.

[Foredawnbio]

Glad to help you. Thank you again for using SIMER software.