Main content area

A genomic approach to selecting robust and versatile SNP sets from next-generation sequencing data for genome-wide association study in citrus cultivars

Shimizu, T., Kaminuma, E., Nonaka, K., Yoshioka, T., Goto, S., Matsumoto, T., Katayose, Y., Mochizuki, T., Tanizawa, Y., Toyoda, A., Fujiyama, A., Nakamura, Y.
Acta horticulturae 2016 no.1135 pp. 23-32
Citrus, crossing, cultivars, genetic analysis, genetic variation, genome, genome-wide association study, genomics, genotype, genotyping, high-throughput nucleotide sequencing, marker-assisted selection, nucleotide sequences, single nucleotide polymorphism
The large-scale genotyping assay is a prerequisite for modern genetic analysis, and single-nucleotide polymorphism (SNP) markers that enable high-throughput genotyping are widely used for genome-wide association studies (GWAS) and genomic selection (GS). However, SNP markers randomly selected from limited genome data often fail in genotyping certain citrus crosses because of high genetic diversity and frequent polymorphism among accessions. Ambiguous markers that may yield false genotyping results add undesired bias to GWAS and GS. In this study, we aimed to select an SNP marker set that minimise genotyping failure among a wide range of citrus varieties. We initially selected 15 ancestral varieties for an SNP discovery panel. Possible SNP regions among the cultivars in the panel were extracted by next-generation sequencing (NGS). The selected SNPs were evaluated for their scores and also their distances to adjacent SNPs. We consequently selected the 756 most discriminative SNPs according to their position in the reference genome sequence and designed a GoldenGate SNP array. Genotyping of 120 citrus cultivars passed 674 SNP markers according to their scores. The robustness and versatility of the passed SNP markers were evaluated with nine validated trios, and consistent SNP markers were selected. Using these consistent markers, we confirmed cross combinations of 71 trios. Equivocal SNP markers that showed combination-dependent aberrant segregation were eliminated, leaving 598 SNP markers. The genotypes obtained with the selected SNP markers were consistent in the 80 trios. These results demonstrate SNP discovery in diverse citrus cultivars using NGS technology and SNP marker design according to a reference genome, and show that validating genotypes with multiple trios permitted the design of robust and versatile SNP markers. They will contribute to accurate GWAS and GS analysis in citrus cultivars.