M. Makhoul, C. Rambla, K. P. Voss-Fels, L. T. Hickey, R. J. Snowdon & C. Obermeier

Theoretical and Applied Genetics August 2020; (133): 2413–2430

Key message

Conversion of SNP chip assays into locus-specific KASP markers requires adapted strategies in polyploid species with high genome homeology. Procedures are exemplified by QTL-associated SNPs in hexaploid wheat.

Abstract

Kompetitive allele-specific PCR (KASP) markers are commonly used in marker-assisted commercial plant breeding due to their cost-effectiveness and throughput for high sample volumes. However, conversion of trait-linked SNP markers from array-based SNP detection technologies into KASP markers is particularly challenging in polyploid crop species, due to the presence of highly similar homeologous and paralogous genome sequences. We evaluated strategies and identified key requirements for successful conversion of Illumina Infinium assays from the wheat 90 K SNP array into robust locus-specific KASP markers. Numerous examples showed that commonly used software for semiautomated KASP primer design frequently fails to achieve locus-specificity of KASP assays in wheat. Instead, alignment of SNP probes with multiple reference genomes and Sanger sequencing of relevant genotypes, followed by visual KASP primer placement, was critical for locus-specificity. To identify KASP assays resulting in false calling of heterozygous individuals, validation of KASP assays using extended reference genotype sets including heterozygous genotypes is strongly advised for polyploid crop species. Applying this strategy, we developed highly reproducible, stable KASP assays that are predictive for root biomass QTL haplotypes from highly homoeologous wheat chromosome regions. Due to their locus-specificity, these assays predicted root biomass considerably better than the original trait-associated markers from the Illumina array.

See: https://link.springer.com/article/10.1007/s00122-020-03608-x

Figure 2: Comparison of SNP chip clustering plots from software GenomeStudio, SNP prediction/calling and probe alignment specificities (for two example genotypes) for simple SNP and hemi-SNP probes on chromosome 5 of homozygous hexaploid wheat accessions. a Simple SNP where the probe binds specifically to the 5B homeolog. b Biallelic non-homeologous hemi-SNP probes, where all probes bind to 3 homeologs producing two clusters (1 heterozygous and 1 homozygous). c Biallelic homeologous hemi-SNP probes where the probes bind to all 3 homeologs producing two heterozygous clusters. d Triallelic hemi-SNP probe with two heterozygous clusters, where prediction of a polymorphism on 5B homeologue resulted in an incorrect T/G call when relying on customer probe data, whereas Sanger sequencing revealed a 5B homeologue-specific A/G polymorphism