Single Nucleotide Polymorphism Identification, Genotyping, and Validation

A DNA library was prepared using gDNA from parents and 140 F₁ individuals, and was sequenced using an Illumina HiSeq2500 instrument. Alignment of these reads with a reference genome (Ren et al., 2019) was then conducted with the Burrows-Wheeler Aligner (BWA, mem -t 4 -k 32 -M -R) tool, as it can facilitate the alignment of low-divergence sequences with large reference genomes (Li and Durbin, 2010). Alignment files were converted into BAM files with the SAM tools program (Li et al., 2009), and SNPs were identified using GATK (4.1.3) based on the following criteria: parents should not have <11 base support numbers, and offspring should not have < five base support numbers, and the quality of the variation should not be smaller than 60. After hard filtering, a custom Perl script was used to filter SNPs (only exhibiting segregation X² < 0.00001 are included), and SNP markers with more than 10% of missing data or duplicated markers (markers with the same genotype for all individuals) were discarded. The variant effects of identified high-quality SNPs were predicted using an ANNOVAR approach (Wang et al., 2010).

KASP marker design was based on SNPs within 50 bp upstream and downstream regions as determined using the Cereals DB website¹ (Wilkinson et al., 2016). SNP accuracy was validated using KASP assays designed based on corresponding read sequences, harboring SNPs of interest mapped to QTL regions. For these assays, a 1.6 μl PCR reaction system, containing 0.8 μl of KASP Master mix (LGC, Biosearch Technologies), 0.05 μl of each primer, and 0.8 μl of DNA (5–10 ng/μl), was analyzed via PCR based on provided directions with an IntelliQube instrument (LGC, Biosearch Technologies). KASP-SNPs were compared with corresponding SNPs to establish numbers of mismatches, and converted KASP-SNPs were validated on a population of 50 F₁ individuals.