Various mainstream publications have also highlight the grape work. Please click on the links below to access these articles.
Black R. (2011) New grapes needed to keep wine flowing. BBC News: January 17, 2011.
Knowledge of the genetics of an organism can speed up the breeding process and help produce cultivars that possess desirable traits. To discover genetic variants associated with a particular trait of interest in the grape, researchers have traditionally used linkage mapping where genetic markers are sought that co-segregate with a trait of interest in progeny produced by a cross between two parents. The grapevine (genus Vitis) is a woody perennial and it takes three years to go from seed to fruit. Linkage mapping is therefore a tedious, expensive and time-consuming process in the grapevine. An alternative approach to finding gene-trait associations is to look at a large number of genetic markers across the entire genome and to measure traits of interest in wide variety of cultivars. One then simply tests if any of the markers are correlated with the traits one has measured. This approach, called a genome-wide association study (GWAS), has proved to be very successful in identifying disease-causing genetic variants in humans and it holds great promise for plant species in which linkage mapping is difficult. The potential to discover genetic variants underlying traits of interest using GWAS in the grapevine is particularly promising as it has a relatively small genome (475Mb), linkage disequilibrium (LD) breaks down very quickly (within 300 bp) and there is substantial trait variation within the domesticated grapevine, Vitis vinifera.
The first step towards GWAS in the grapevine requires the discovery of a large set of genetic markers. In collaboration with researchers from the USDA research station in Geneva, NY and Doreen Ware's lab at Cold Spring Harbor, we have undertaken a large-scale discovery of genetic markers, called single nucleotide polymorphisms (SNPs), in the grapevine genome. We have used next-generation Solexa sequencing on reduced-representation libraries of 10 Vitis vinifera cultivars and 7 wild Vitis species to discover 470,000 “good-quality” SNPs and 110,000 “high-quality” SNPs. From the set of 110,000 SNPs, we chose 9000 SNPs to be assayed by a custom genotyping array. We are now genotyping 1000 Vitis vinifera and 1000 wild Vitis species with the 9K Vitis SNP array and our results indicate that the SNP chip has good power to distinguish between cultivated and wild Vitis, among Vitis vinifera and even among wild Vitis species. These data give us a preliminary glance at the genetic structure of the genus Vitis, which is crucial for the design of future GWAS. Moreover, these data will allow us to assess in detail the degree of LD in the grapevine genome, which ultimately determines the number of markers required for GWAS. Our efforts provide only the very first steps towards GWAS in the grapevine. We anticipate that the decreasing costs of sequencing will soon make it possible to sequence the entire genomes of a large number of grapevine cultivars and that the focus will soon turn to high-throughput phenotyping methods. Overall, we have demonstrated that next-generation sequencing is useful for SNP discovery in a high diversity plant species and we are excited about the future prospect of mapping traits of interest in the grapevine using the most modern genomics tools available.
Figure 1: Genetic structure of the grapevine. This figure depicts the genetic relationships among 17 grapevine types using principal components analysis (PCA). The data used included 9000 SNPs discovered by Solexa sequencing. The x-axis clearly separates the cultivated Vitis vinifera from the wild Vitis species. Note that Vitis sylvestris is the wild ancestor of the cultivated Vitis vinifera. The y-axis distinguishes among Vitis vinifera cultivars and shows a suggestive longitudinal gradient as indicated by the West-East arrow on the side.
Figure 2: Genetic structure of the grapevine. The genetic relationships among ~100 grapevine cultivars are depicted in two dimensions using PCA. Our preliminary analyses suggest that we will easily be able to identify hybrids using the 9K Vitis SNP array and that it will provide good resolution to distinguish among Vitis vinifera cultivars.
Figure 3: Genetic structure of wild Vitis species. Vitis amurensis, of Eurasian origin, was excluded from the current figure because it differs greatly from the other North American wild species included in the study. This figure depicts the relationships among wild North American Vitis species. Our preliminary analyses suggest that the 9K Vitis SNP array will be useful to decipher relationships among wild Vitis species. In addition, as can be seen in this figure, our data will help identify grapevines that are currently marked as “unknown” in the USDA grape germplasm collection.
Figure 4: Here is an example of the extreme berry size variation observed in Vitis vinifera!
We routinely use barcode readers to collect phenotype data in a reliable and efficient fashion. We used barcode readers to collect phenotypic information from grape berries from the USDA grape germplasm collection in Davis, California. You can find detailed information about the barcode system used for phenotyping in the Buckler lab by clicking here. Below you will find photos of the phenotyping instrument that we designed for collecting grape berry phenotypes (e.g. colour, width, height).
Figure 5: The set up is rather simple: an ironing board is used to hold the equipment and a space is found under a tree for shade.
Figure 6: Representative grape clusters are collected from the vineyard, labeled and brought to the phenotyping station.
Figure 7: An example of measuring berry width. The barcodes above the grapes are pasted together and, when scanned at the approrpriate place, will provide a measure of the width of 10 berries as shown above.
Figure 8: As in Figure 7, but for berry length.
Our research on the grape is funded by the USDA-ARS.