Sequencing based genotyping in Strawberry

Currently the Affymetrix IStraw35 Axiom array is the most used whole-genome genotyping platform in strawberry at Fresh Forward, containing approximately 35,000 usable SNP markers. However, this genotyping platform has the disadvantages that it only contains a limited number of SNPs and that the selection of these SNPs is biased by the composition of the discovery panel. As a result, high-density marker screening is not possible and therefore important QTLs or certain desirable alleles for breeding could remain undetected in association mapping studies. In recent years, developments of molecular genotyping techniques have gained momentum and resulted in several genotyping techniques alternative to SNP arrays that have the potential of high density marker screening. These alternatives are several genotyping by sequencing (GBS) techniques where high coverage whole genome sequencing (WGS) is the golden standard. Other GBS techniques include skim sequencing and targeted sequencing where the genetic information is reduced in comparison with high coverage WGS, making it more suitable for high-throughput analyses. However, apart from high coverage WGS, none of these GBS techniques have successfully been applied in strawberry breeding programs. Although high coverage WGS as a GBS source could be a viable alternative to SNP arrays, it is too expensive to be applied on a large scale in breeding programs. Thus, we currently lack different validated techniques for large scale high-density marker screening in strawberry. Several alternative GBS methods, such as skim sequencing or targeted sequencing reduce the amount of sequencing information making them promising candidates for large scale application. However, these are not suitable for complete allele discovery due to the reduced genetic information. Therefore part I is about performing allele discovery in a high coverage whole genome sequenced coreset. For this allele discovery several imputation and haplotyping techniques that were already developed in an earlier project on genomic tools for polyploids and/or literature will be investigated. When all/most allelic diversity is discovered, this diversity can be followed in downstream breeding material by SNPs. Typically whole genome (re)sequencing results in a lot of SNPs of which many are co-segregating SNPs. This means that several SNPs can be tagging the same allele. Consequently, complete genetic information (by high coverage WGS) is not needed to tag the discovered alleles in downstream breeding material. Therefore, part II is about investigating whether the in part I found alleles can be tagged in downstream breeding material by using more high-throughput GBS methods such as skim sequencing and/or targeted sequencing. In part I and II a pipeline is developed for large scale high density marker screening. This pipeline will be validated in part III where we will investigate several traits (i.e. cold requirement, white streak and skin vulnerability) in QTL mapping studies. Moreover, we aim to use this pipeline also in a genome wide association study. Successful implementation of aforementioned objectives will enable us to detect more QTLs, get a higher resolution and perform more breeding by design.