Meeting Abstract
P1.207 Saturday, Jan. 4 15:30 Coalescent theory reconstructs the known history of experimentally evolving yeast populations CAPPER, RL*; WRIGHT, AM; MIKHEYEV, A; University of Texas at Austin; University of Texas at Austin; Okinawa Institute of Science and Technology roxana.capper@gmail.com
Coalescent models can estimate relationships among individuals to produce tree topologies, rates of evolution and demographics such as population size fluctuations. However, coalescent theory is rooted largely in statistics and simulations, and no direct experimental tests have previously been performed. To test the performance of coalescent theory on a known phylogeny, we propagated asexual, haploid yeast in the lab for approximately 250 generations. In one experiment, we grew cells for four generations, then split the culture into two populations which were maintained at a constant population size by bottlenecking the culture every 12h. One population was grown in the presence of mutagen to artificially increase the mutation rate, while the other branch was grown normally. In a second experiment, we maintained a culture of yeast at a constant population size for 160 generations, then split it into two populations. One population was kept at the same population size while that of the second branch was reduced by an order of magnitude. These populations were grown for an additional 100 generations. Then, we sequenced the genomes of sampled individuals. We used the single nucleotide polymorphism (SNP) data produced to test commonly applied models of coalescence. Though our experimental design is simplistic, recapitulating the known phylogenies and demographics of our populations has been difficult. The topology estimated in the variable mutation rate experiment is not accurate, but the mutational rate difference between clades is indeed detected. The topology of the second experiment is plausible, but the population size decrease is not detected. These issues may be due to the abundance of mutations acquired through treatment with mutagen, working with a single locus, or small sample sizes.
