Biology Reference
In-Depth Information
tive mating on the frequency of deafness would be statistically trivial. The
discovery that one form of recessive deafness is so much more common than
all others, raises the possibility that this assumption may not be correct. In
each generation after the appearance of assortative mating, the deaf chil-
dren of deaf parents entered the new deaf-by-deaf mating pool, along with
a substantially constant frequency of deaf offspring with genetic deafness
who were born to hearing parents. However, the relative frequency of genes
for different forms of recessive deafness should not be the same in the two
groups. Among the former, many of the deaf offspring are the products of
noncomplementary matings. Since the frequency of non-complementary
matings for each type of recessive deafness is proportional to the fourth
power of the respective gene frequencies, there is a strong bias towards the
transmission of the most common form(s) of recessive deafness to the deaf
offspring from these matings. The net effect of this process is the preferen-
tial transmission of
Cx26
deafness to the deaf-by-deaf mating pool of the
succeeding generation. This in turn will progressively increase the frequency
of
Cx26
deafness, the proportion of non-complementary matings, and the
overall incidence of genetic deafness. Nonrandom mating by itself can only
alter the genotype frequencies and not the underlying gene frequencies. But,
when it is accompanied by relaxed selection, it can greatly accelerate the
changes in gene frequency that can accompany attainment of a new muta-
tional equilibrium. The magnitude of these effects will depend on many
factors, including the number of recessive forms of deafness and their rela-
tive frequency, the overall proportion of deafness that is genetic, and the
relative frequency and fertility of marriages among the deaf.
For any genetic trait, it is to be expected that gene and genotype fre-
quencies vary in different populations; but in the case of the genes that
cause deafness, the mating structure of the population is another important
potential source of variation. Available data suggest that the incidence
of
Cx26
deafness in India (Green, personal communication), Mongolia
(Pandya, personal communication), China (Liu, personal communication)
and Japan (Fuse et al. 1999) is substantially lower than in populations where
there has been a long tradition of intermarriages among the deaf. In the
past, marriages among the deaf were virtually unheard of in India. In Mon-
golia, not a single one of 380 probands studied at the School for the Deaf
in Ulan Bator was the offspring of a DxD mating. In China, Liu observed
only two DxD matings among 184 marriages of the deaf (Liu et al. 1994).
On the other hand, if there are populations in which the total proportion
of
Cx26
deafness is 50%, for example, then 25% of all marriages among
the deaf should have all deaf offspring because of non-complementation.
In the limiting case, if all deafness in a population is caused by a recessive
mutation at a single locus, 100% of marriages among the deaf should be
non-complementary. This finding appears to characterize the genetic
epidemiology of
DFNB3
in the Balinese population reported by Friedman
et al. (1995).
