Underhill, Anna2020-08-252020-08-252019-04https://hdl.handle.net/11299/215040University of Minnesota M.S. thesis. April 2019. Major: Applied Plant Sciences. Advisor: Matthew Clark. 1 computer file (PDF); xi, 142 pages.High-throughput phenotyping methods have gained popularity in the plant sciences due to their potential to more quickly collect data, reduce human error, and investigate plant characteristics in new ways. In grapes, many economically important traits are quantitative, varying across a spectrum and displaying diverse phenotypes. Though rating scales exist for such traits, their usefulness can be limited by their ability to capture variation across populations; additionally, they require judgement that can vary based on the individual scoring the trait. Automated systems can be used to remedy these issues, eliminating subjectivity and more fully describing phenotypic variation. In these experiments, a semi-automated image analysis system was used to evaluate fruit cluster compactness and berry color in a multispecies hybrid wine grape (Vitis spp.) population. First, color-based image segmentation was used to isolate components of the fruit cluster morphology. Berry color was quantified using several different color spaces, and a MATLAB program was written to measure several morphological components to evaluate cluster compactness. Both color and compactness traits were used to perform quantitative trait loci (QTL) mapping, where associations between the traits and genetic regions were identified. Known QTL for berry color on chromosome 2 were identified, along with several minor QTL associated with color and anthocyanin content. Image-derived traits were associated with known QTL such as the chromosome 9 rachis length QTL, and also identified other regions of interest relating to cluster compactness. Altogether, these projects demonstrate the advantages of high-throughput phenotyping methods and their ability to identify new variation among quantitative traits.enAnthocyaninsCompactnessGrapesPhenotypingQTLVitisThesis or Dissertation