where 1G = 1 is an indicator function that takes value 1 for heterozygotes and 0 for homozygotes. We refer to this as the general model. The extra parameter, γ, models the deviation from a multiplicative model at the heterozygote. We refer to it as the dominance parameter. Other commonly used models are special cases of this model and can be recovered by setting the dominance parameter to specific values: γ = 0 gives a multiplicative model, γ = β a dominant model, and γ = −β a recessive model (where β>0, which may be assumed without loss of generality by relabeling the alleles). To distinguish between parameters corresponding to different SNPs we label them with a subscript, e.g. βA is the additive parameter for SNP A.
