The Proceedings of the Eighth International Conference on Creationism (2018)

be invariant. Figure 2d, like Figures 1b and 1c, may all reflect non-variant Edenic chromosomes that lacked designed variants. This makes sense, because in all three cases there are very few SNP alleles, and their allele distributions are very strongly skewed toward lower allele frequencies. These distributions only make sense if these chromosomes were very young and were initially invariant. If all the autosomal chromosomes initially lacked designed variants, it seems possible that they might similarly appear very young. 3. Illustrating the Heterozygous Adam and Eve Model Could the observed human allele frequency distributions be derived from a massively heterozygous Adam and Eve? We used numerical simulations to examine this question. A. Simulations involving only designed alleles, all having an initial frequency of 50%. We first performed simulations with only designed alleles (no new mutations arising). Note that most of these experiments show only the minor allele but still show a full frequency range of 0–100%. We initially specified that all alleles start with an allele frequency of 50%. For these simulations we generally specified zero mutations, a short timeframe (200 generations), 989 designed allele pairs, 989 linkage blocks, and nearly-neutral fitness effects (i.e., no selection). The initial allele frequency distribution (first generation) for all such runs is shown in Figure 3a. The allele “distribution” is a single spike at 50%. Given this starting point, and given a constant population size of 1000, after 200 generations the distribution was a narrow bell-shaped curve that was still centered at 50%, indicating that in this time there had been very limited genetic drift (Figure 3b). A second, more biblically-realistic experiment involved the same initial allele frequency and a time span of 200 generations, but started with a population of two, followed by rapid population growth, a severe population bottleneck to just 6 people in the tenth Sanford et al. ◀ Designed genetic diversity in Adam and Eve ▶ 2018 ICC 206 Model Corresponding Figure MSE (x10 -3 ) Evolutionary 2b 5.75 Evolutionary Adam and Eve 2c 0.70 A & E, Designed Alleles 6b 1.16 A & E, Designed Gametes 7 0.84 Table 2. Differences between the normalized observed allele distribution (chromosome 22) and the normalized distributions of several models. The Mean Squared Error (MSE) is simply the average of the error terms ( Y i – Ŷ i ) 2 , where Y i is the value of the test distribution and Ŷ i is the value of the reference distribution (in this case, chromosome 22) at that same frequency. Smaller values indicate a tighter fit to the reference sequence. The MSE is not a significance test, but the results are a useful guide when tweaking parameters within a model to make the model fit an expected distribution. The most divergent model was the evolutionary simulation. The threeAdam and Eve models all show a much tighter fit to chromosome 22. Further experimentation with parameter settings will improve the fit of the Designed Alleles and Designed Gametes models. Figure 3a. A simple illustration of a designed allele “distribution.”In the first generation of a simulation. In this example, all designed allele pairs begin with a 50/50 ratio, so all designed alleles in this experiment start with an initial frequency of 50%. Because for each contrasting allele pair, one allele has one desireable function and one has another alternative desireable function, we arbitrarily designated one allele “beneficial” (green) and the other allele is ar- bitrarily designated “deleterious” (red). These alleles are made nearly neutral (unresponsive to selection). Note that in these cases Mendel has plotted allele frequencies from 1–100%. Simulated and plotted using Mendel version 2.7.2. Figure 3b. This is a simple illustration of a designed allele distribution, similar to Figure 3a, but after 200 generations of drift. It assumes no mutations and a population with a constant population size of 1000. It is obvious that drift has happened, but in this timeframe the drift is modest. Mendel has plotted allele frequencies from 1–100%. Simulated and plotted using Mendel version 2.7.2. Figure 3c. This is a simple illustration of a designed allele distribution, similar to Figure 3b, after 200 generations of drift, with no mutations, but with two severe population constrictions. The starting population size was 2, followed by rapid growth and an extreme single-generation bottleneck in generation 10 (6 individuals), followed by rapid growth up to 1000. As can be seen, bibli- cal population dynamics greatly increase the rate of genetic drift. Mendel has plotted allele frequencies from 1–100%. Simulated and plotted using Mendel version 2.7.2.