Population Genetics Edge Effects - Genetic Diversity in Establishing Plant Populations: Founder Number and Geometry - page 113

Further Analyses Will Use Loci with Numerous Low-Frequency
Unique Alleles
As shown in Fig. 8.14, the rate of heterozygosity loss across generations,
if absolute values are disregarded, is similar between trials in which
the allelic diversity of the source populations differs (uppercase versus
lowercase letters) but the distance of founders from the edge of the preserve
is the same. However, in terms of absolute values, the comparisons differ
markedly between two trials with identical input conditions except that
in one trial there are many unique alleles and in the other there are only
two different alleles of high frequency per locus. Initial heterozygosity
approaches 1 in the former trials (uppercase letters) and 0.5 in the latter
trials (lowercase letters).
The examples above demonstrate that as trial populations develop
under some sets of conditions, the changing level of heterozygosity can
be informative as to diversity loss whether loci with low or high allelic
diversity are used, although loci with similar levels of allelic diversity must
be compared. The latter consideration raises the issue of the validity of
comparing genetic diversity statistics across populations or species that have
different loci-allele array characteristics. Further, unique allele retention as a
measure of effects of different factors on genetic diversity is a more sensitive
indicator when loci have numerous different alleles of similar or equal low
frequency compared to the simpler two-allele analyses ( Fig. 8.17).
Thus, from this point on, we will conduct practically all of the following
trial comparisons using source populations with 30 loci, each with 100
different alleles of frequency = 0.01. The latter types of loci, or at least alleles,
are of great interest to the conservationist. Loci with many unique alleles
can exhibit an increased propensity to change under many of the processes
that drive the establishment and maintenance of small populations. As is
demonstrated above, it is more diffi cult to lose a unique allele to drift when
that allele is common in a population. Complete loss of rare unique alleles
via drift will occur with much greater frequency. It is often noted that very
common alleles may be under stronger selection and may not contribute
much in terms of phenotypic diversity compared to some multiallelic loci.
Populations can be more homogenized with respect to a particular feature
when only one or two alleles for that feature are available. As will be seen
in some of the later trial comparisons, when using loci with high diversity,
change in the number of unique alleles can be a more sensitive indicator,
compared to shifts in population heterozygosity, of major changes in genetic
diversity. Restoration genetic conservation models designed according to
analyses involving numerous rare alleles will be the most forgiving: if it