Comparative Subsequence Sets Analysis (CoSSA) is a robust approach to identify haplotype specific SNPs; mapping and pedigree analysis of a potato wart disease resistance gene .

Background: Standard strategies to identify genomic regions involved in a specific trait variation are often limited by time and resource consuming genotyping methods. Other limiting pre-requisites are the phenotyping of large segregating populations or of diversity panels and the availability and quality of a closely related reference genome. To overcome these limitations, we designed efficient Comparative Subsequence Sets Analysis (CoSSA) workflows to identify haplotype specific SNPs linked to a trait of interest from Whole Genome Sequencing data.

Results: As a model, we used the resistance to pathotypes 2, 6 and 18 that co-segregated in a tetraploid full sib population. Genomic DNA from both parents, pedigree genotypes, unrelated potato varieties lacking the wart resistance traits and pools of resistant and susceptible siblings were sequenced. Set algebra and depth filtering of subsequences (-mers) were used to delete unlinked and common SNPs and to enrich for SNPs from the haplotype(s) harboring the resistance gene(s). Using CoSSA, we identified a major and a minor effect locus. Upon comparison to the reference genome, it was inferred that the major resistance locus, referred to as , was located on the north arm of chromosome 11 between 1,259,552 and 1,519,485 bp. Furthermore, we could anchor the unanchored superscaffold DMB734 from the potato reference genome to a synthenous interval. CoSSA was also successful in identifying in a reference genome independent way thanks to the de novo assembly of paired end reads matching haplotype specific -mers. The de novo assembly provided more R haplotype specific polymorphisms than the reference genome corresponding region. CoSSA also offers possibilities for pedigree analysis. The origin of was traced back until Ora. Finally, the diagnostic power of the haplotype specific markers was shown using a panel of 56 tetraploid varieties.

Conclusions: CoSSA is an efficient, robust and versatile set of workflows for the genetic analysis of a trait of interest using WGS data. Because the WGS data are used without intermediate reads mapping, CoSSA does not require the use of a reference genome. This approach allowed the identification of and the design of haplotype specific, diagnostic markers.