UC Irvine

Population differentiation is both one of the central processes underlying the diversity that we observe in the natural world, and a mechanism that can be used to differentiate between evolutionary forces both at the level of the polymorphism, and at the level of the entire genome. Here, I use simulated evolution to analyze the statistical power to detect signals of selection in artificially selected laboratory populations, and use genomic data from wild populations of the clam shrimp Eulimnadia texana to identify genomic signals of selection in wild populations. Several loci in the wild populations appear to be under selection, and I analyze the types of genes that appear to contribute to differentiation of these populations. Additionally, I describe an analysis of genome assembly techniques that allowed for the creation of a highly contiguous genome assembly in the clam shrimp. I find that a pipeline that uses custom software to combine the results of several different genome assemblers is capable of producing genomes using long-read genomic sequencing data that are orders of magnitude more contiguous that pre-long-read methods. Simulations of experimental evolution indicated that extremely high levels of replication were necessary in order to achieve high power to detect signals of selection in experimental evolution. To this end, I describe a set of replicate experimentally evolved populations of E. texana that can be used to identify regions under selection with much higher power than could be accomplished with earlier experimental evolution schemes.