Deep Re-Sequencing of Apricot Genome toward Genome-Wide Association Analysis in Stone Fruit Species

Due to a moderate level of polymorphism, the availability of wild germplasm and a well-described history of dissemination from primary centers of origin, apricot provides one of the best model systems among the Prunus species for marker-trait association analysis.  Here, we tested the potential of the Illumina platform (100bp-paired-ends-reads) coupled with CLC software (http://www.clcbio.com) for reconstruction of apricot genomes using 2 different strategies - ‘de novo’ assembly and mapping apricot reads against the P. persica v1.0 assembly (www.rosaceae.org). For this pilot experiment, we selected eight apricot genotypes representing 6 domesticated cultivars (Stark Early Orange, Harlayne, Goldrich, Krasnoshchekii, Real d’Imola, Shalakh) and 2 non-domesticated species P. mume and P. sibirica var davidiana.  Cultivar selection relied on global germplasm analysis by Zhebentyayeva et al. (2012) to address the domestication-related questions - the fruit quality, seed taste, tree architecture, disease resistance and eco-geographical adaptation to environments. In this presentation, we report results on genome-wide mapping sequences from 6 domesticated apricots (pooled reads) that allowed aligning 87.9% out of 761 mln reads against the peach genome. The low proportion of gene-derived sequences among unmapped reads (less than 10%) was in agreement with the known genetic similarity of both genomes from genetic mapping studies. The total genome coverage of ~165 x (or ~28 x per genotype) captured the entire peach exom with expected allele representation and was sufficient for genome-wide SNP/indel detection across the entire apricot genome.  Fixed sets of markers for regions of breeding interest were generated for SNP verification in germplasm and breeding material.