Exploring chromosomal variations in garden roses: Insights from high-density SNP array data and a new tool, Qploidy.

Abstract

Roses (Rosa L.) are among the most economically important ornamentals worldwide, with ploidy ranging from diploid (2x) to hendecaploid (11x), though most cultivars are diploid (2x), triploid (3x), or tetraploid (4x). To enable large-scale analyses of ploidy and aneuploidy in roses using high-density single nucleotide polymorphism (SNP) array data, we developed Qploidy, an R package. Qploidy leverages tools for estimating allele dosage, adapts methods from human genetics for copy number estimation, and optimizes the standardization of allele intensity (R) and B allele frequency (BAF) for ploidy levels greater than 2x. With Qploidy, we analyzed a dataset of 1944 samples consisting of 588 samples from a germplasm collection and 1356 samples from 14 biparental mapping populations. The majority of genotypes in the germplasm collection were tetraploid (56%), followed by diploids (20%) and triploids (11%). The percentage of aneuploids was lower in the germplasm collection (2%) compared to biparental populations (16%). Reduced fitness likely explains the higher frequency of aneuploids in mapping populations compared to the germplasm collection, where stronger selective pressures normally act. In tetraploid biparental populations, pentasomy (65%) was significantly more common than trisomy (19%). Also, aneuploid states were predominantly transmitted through the female parent (87%), suggesting greater gametophyte sensitivity to chromosome number variation, particularly the loss of a chromosome. Since aneuploidy disturbs linkage and quantitative trait loci (QTL) analyses, Qploidy may also be used to guide the removal of aneuploid-affected data prior to downstream analysis. Besides roses, Qploidy can be used to study ploidy and aneuploidy in other polyploid species.