| Description |
Making sense of natural variation has been of longstanding interest for biologists, and understanding the degree to which traits are genetically controlled and how genetic variants translate into phenotypic variation are fundamental to many fields in the biological sciences. Nevertheless, progress in linking genotypic variation to phenotypic variation has been slow for many reasons, including the difficulty in identifying genetic variants in populations, and genotyping the variants across many individuals for which phenotypes have been collected. Recently, improvements in genotyping made possible by high-throughput sequencing have alleviated important bottlenecks for genetic mapping studies. Here, we present the genome sequence of two-spotted spider mite,<italic>Tetranychus urticae</italic>, and report the development of approaches for genetic mapping in mites tailored to their life history. <italic>T. urticae</italic> is an agriculturally relevant herbivore, and rapidly develops resistance to pesticides. As a test case, we present our findings on the identification of a locus conferring cross-resistance to the commercially used acaricides (pesticides) etoxazole, hexythiazox and clofentezine. A bulk-segregant analysis mapping approach that we developed identified a single nonsynoymous mutation in <italic>chitin synthase 1</italic> that underlies resistance to all three pesticides. The identification of this locus has important consequences for programs to control mites in agriculture, and the methods we used are of relevance broadly for genetic studies in nonmodel organisms. Additionally, I present my work to understand the genetic and environmental control of gene expression using variance components analysis and association mapping to identify loci influencing expression levels (expression quantitative trait loci, or eQTL) in two independent populations of the plant <italic>Arabidopsis thaliana</italic>. This work is important because it sheds light on the pathways that translate genetic variation into end organismal phenotypes. We find that <italic>cis</italic> eQTL are highly reproducible across environments and studies. In contrast, <italic>trans</italic> effects are less so, but are pervasive in a highly structured natural collection of <italic>Arabidopsis</italic> lines for which global kinships capture the collective action of trans effects and explain much of the expression variation observed. Collectively, this dissertation highlights the utility of leveraging high throughput technologies to improve our identification and understanding of the genetic basis of diverse phenotypes. |
