The Carrot Genome PDF Download

Carrot (Daucus carota subsp. sativus) is a nutritionally significant vegetable crop. An important target of selection in carrot breeding programs is suite of morphological root traits which together define market class-i.e., the market into which a specific variety is intended to be sold (e.g., juicing, dicing, storage, fresh market, baby carrot production). The size and shape the taproot, which can range from long and tapered to short and blunt, have been used for at least several centuries to classify cultivars in this way according to human preference and production methods. Mechanization in the cultivation, harvesting and post-harvest handling of the crop has made these traits increasingly relevant for both farmers and breeders. However, these quantitative phenotypes have historically been challenging to objectively evaluate, and thus subjective visual assessment of market class remains the primary method by which selection for these traits is performed. This has hindered not only the establishment of metric-based standards for market classes, but also the investigation the genetic basis of such quantitative phenotypes. In order to dissect the genetic control of the shape features that define market class in carrot, a tool is required that quantifies the specific shape features used by humans in distinguishing between classes. Advancements in digital image analysis have recently made possible this high-throughput quantification of size and shape attributes, and Chapter 2 of this dissertation describes the functioning and performance of a phenotyping pipeline which implements such methods. This is the first such platform to include a series of a preprocessing algorithms whereby RGB images are converted to binary masks, which are then standardized to remove curvature and residual root hairs. Phenotyping is then performed, which includes the quantification of traits that could be measured by hand, such as length and width, as well as measurement of higher-dimensional traits, through the implementation of principal components analysis of the root contour and its curvature. Of particular importance is the idnetification of a previously undescribed phenotype - root fill - as the most significant source of variation across carrot germplasm. This platform's high-throughput performance and accuracy was validated in two experimental panels: a diverse, global collection of germplasm was used to assess its capacity to identify market classes through clustering analysis, and diallel mating design between inbred breeding lines of differing market classes was used to estimate the heritability of the key phenotypes that define market class. Together with the recent development of a high-quality reference genome for carrot, it is now feasible to utilize modern methods of genetic analysis in the investigation of the genetic control of root morphology. To this end, in Chapter 3 of this dissertation, the digital phenotypes of the diversity panel described in Chapter 2 are combined with a set of dense molecular markers developed using high-throughput sequencing. The use of both genome wide association analysis and genomic predictions based on genomic-estimated breeding values is described. Novel QTL were identified for four of the traits underlying market class; of particular interest is an extremely well-defined peak of chromosome 2 for the novel, and previously uncharacterized "root fill" trait. This comparative analysis provides the first convincing evidence that the traits underlying market class are highly polygenic in nature, under the influence of many small effect quantitative trait loci (QTL), but that relatively large proportions of additive genetic variance for many of the component phenotypes support high predictive ability of genomic-estimated breeding values. This study thereby represents a novel advance in our understanding of the genetic control of market class in carrot root. In addition, concrete guidelines are presented outlining the practical potential of using genomic predictions for quantitative traits in horticultural crops.