Founder Number and Allelic Diversity - Genetic Diversity in Establishing Plant Populations: Founder Number and Geometry - page 71

eggs, tissue for cloning), since that will cover the full range of potential
source population mating systems.
Lawrence et al. outline a means to calculate the probabilities of capturing
all unique source alleles, given that a specifi ed number of founders are
selected, for loci with multiple unique alleles. These calculations quickly
become quite complex: for the four-allele situation, one must compute over
40 terms (their table 5). They argue that, for a multiallelic locus, even as
the number of the unique alleles at frequency 0.05 increases, 172 founders
should still be suffi cient since increasing the number of such alleles causes
only a “very small” reduction in the probability of conserving all alleles.
They then generalize further to multiple loci: the probability of capturing
all unique alleles across a specifi ed number of unlinked loci is equal to
the product of the probabilities calculated for each locus. For example,
consider one locus in a 100% outcrossing source population with four
unique alleles of frequencies 0.85, 0.05, 0.05, and 0.05. When choosing 172
founders, the probability of capturing all unique alleles is 0.999999934 for
that one locus (Lawrence et al. 1995: table 5). The probability of capturing
all alleles across 1000 loci of this type = 0.999999934 1000 = 0.999934002.
Note, however, that if we conduct the same estimation for a 100% selfi ng
population, the probability of capturing all unique alleles across 1000 such
loci = 0.999557878 1000 = 0.642608397. While the probability of losing unique
alleles thus increases, exactly how many alleles will be lost on average, or
the range across multiple trials, is not obvious. Probabilities for total unique
allele capture and estimates of the loss of unique alleles become even less
clear for cases where there are numerous loci differing in the number of
unique alleles and allelic frequencies. It is only fair to note that the estimate
that 172 seeds are needed from the source population to preserve most of
the source population genetic diversity has been found to be problematic
on several grounds (e.g., see Brown and Hardner 2000).
Beyond preservation of unique alleles in colonizing populations,
when new populations are established, some estimation of the level of
average heterozygosity of the founders, as well as ways in which F is
changing as the population develops, can be important to assessment of
possible inbreeding or drift effects under different population development
scenarios. An estimate of the average heterozygosity and F for the founding
generation, used in conjunction with subsequent targets and monitoring
for heterozygosity and F levels in a developing population, can provide
information on possible management options. For example, a trend of
declining levels of estimated heterozygosity for a developing population
might indicate that more artifi cially supported population mixing or
cross-breeding, or increased supplementation of the population with new
individuals, is called for.