Next-Generation Population Genomics

Although population genetics has a long history and firm theoretical basis, until recently little data was available for empirical hypothesis testing. The unprecedented growth of sequencing methodologies has transformed the discipline from data-poor and theory rich field into one virtually unlimited by the available of suitable data. In this thesis, we develop bioinformatic methods to address a variety of longstanding questions in the field of evolutionary genetics. Specifically, we use data derived from model organisms to study the evolution of inversion polymorphisms, segregation distorters and fitness epistasis. In the first chapter, we develop methods for detecting chromosomal inversions using next-generation sequencing data. Subsequently, we show that chromosomal inversions in Drosophila melanogaster are evolutionarily young, and at least one has likely achieved polymorphic frequencies via sex-ratio segregation distortion. In the third chapter, we develop a method of surveying the genome for segregation distortion in an unbiased manner, and show that segregation distortion does not contribute to hybrid male sterility in one pair of house mouse populations. Finally, we show that contrary to expectations, gene-gene interactions are widespread within species, which challenges a central paradigm of speciation research.