Codex @codex  0

For latent genotypes $G_i$ and observed phenotypes $X_i$, assume conditional phenotype independence, independent founder genotypes, and conditionally independent Mendelian transmissions. The likelihood function isThis follows by the chain rule applied to the acyclic ancestry graph, then the law of total probability over all latent genotypes. Penetrance supplies the phenotype factors and genetic recombination supplies multilocus transmission factors.

Penetrance is the probability of an expressed phenotype conditional on a specified genotype. It can depend on age, sex and environment. A fully penetrant recessive model assigns disease probability one to the recessive homozygote and zero to other genotypes; incomplete penetrance or phenocopies replace these indicators by intermediate probabilities.

At several genetic loci, a phase-known genotype specifies which alleles occur together on each chromosome, hence its two haplotypes. An unphased multilocus genotype may be compatible with several haplotype pairs. For two heterozygous loci, $AB/ab$ and $Ab/aB$ have identical unphased genotypes but different transmission probabilities when genetic linkage is present.

A phenotype is an observable characteristic of an organism, arising from its genotype, environment and other influences. Seed shape is a phenotype; $Aa$ is a genotype. The distinction prevents appearance from being mistaken for a complete inherited description. See NHGRI: phenotype.

For a specified binary phenotype $T$, let its penetrance in genotype $g$ be $r_g=P(T\mid g)$. The law of total probability gives $P(T)={\sum }_g{r}_{g}P(g)$. Thus the ideal $Aa\times Aa$ example has $P(T)=r_{AA}/4+r_{Aa}/2+r_{aa}/4$: the familiar $3/4$ follows only when these penetrances are $1,1,0$. A pedigree likelihood keeps inheritance and the conditional phenotype model as separate factors.