Related Experiment Videos
Abstract:
Most models previously considered for assortative mating were such that no change in gene frequencies occurred, or that one-sided assortative mating was occurring. Two-sided assortative mating is more realistic for human populations, and in this paper a model of two-sided assortative mating is analysed for two autosomal alleles with dominance, an important case for human genetics. The concept of a relative probability of matin between two phenotypes is used, and this variable can take into account factors such as different propensities for assortment in the various phenotypes and so forth. It is shown that the gene frequency may vary from generation to generation and conditions for the establishment of a stable polymorphism are given. For instance, Stanton (1947) has considered a special case of two-sided assortative mating with two autosomal alleles and a constant correlation between mates, where the value expressed by the heterozygote is numerically exactly half-way between the values exhibited by the two homozygotes. Also he has considered a special form of assortative mating, where the probability of occurrence of matings between the same homozygotes has been increased in each generation by the same constant factor as the probability of occurrence of matings between different homozygotes has been decreased. The probability of occurrence of matings involving heterozygotes remains the same as if mating were random. For this particular model the frequency of the genotypes in those who mate in generation t is the same as for the general population in generation t, but the genotypic frequencies change from generation to generation, reaching an equilibrium value. Stark (1976) has also considered a similar model to Stanton's (1947), the major difference occurring in the construction of the mating frequencies, which in this case occurs according to the canonical decomposition of the 2-way table of mating frequencies. Stark shows the correspondence between the results for this model and those given by Malécot (1939, 1948). The results for all these models depend on the particular values assigned to the genotypes. Also these models could be extended as outlined in this article for the case of two autosomal alleles with dominance.