AFLAP: Assembly-Free Linkage Analysis Pipeline using k -mers from whole genome sequencing data

Abstract Background Genetic maps are an important resource for validation of genome assemblies, trait discovery, and breeding. Next generation sequencing has enabled production of high-density genetic maps constructed with 10,000s of markers. Most current approaches require a genome assembly to identify markers. Our Assembly Free Linkage Analysis Pipeline (AFLAP) removes this requirement by using uniquely segregating k -mers as markers to rapidly construct a genotype table and perform subsequent linkage analysis. This avoids potential biases including preferential read alignment and variant calling. Results The performance of AFLAP was determined in simulations and contrasted to a conventional workflow. We tested AFLAP using 100 F 2 individuals of Arabidopsis thaliana , sequenced to low coverage. Genetic maps generated using k -mers contained over 130,000 markers that were concordant with the genomic assembly. The utility of AFLAP was then demonstrated by generating an accurate genetic map using genotyping-by-sequencing data of 235 recombinant inbred lines of Lactuca spp. AFLAP was then applied to 83 F 1 individuals of the oomycete Bremia lactucae , sequenced to >5x coverage. The genetic map contained over 90,000 markers ordered in 19 large linkage groups. This genetic map was used to fragment, order, orient, and scaffold the genome, resulting in a much-improved reference assembly. Conclusions AFLAP can be used to generate high density linkage maps and improve genome assemblies of any organism when a mapping population is available using whole genome sequencing or genotyping-by-sequencing data. Genetic maps produced for B. lactucae were accurately aligned to the genome and guided significant improvements of the reference assembly.