Peter M. Bourke, Virginia W. Gitonga, Roeland E. Voorrips, Richard G. F. Visser, Frans A. Krens, Chris Maliepaard

 Theoretical and Applied Genetics; October 2018, Volume 131, Issue 10, pp 2055–2069

 Key message

Rose morphological traits such as prickles or petal number are influenced by a few key QTL which were detected across different growing environments—necessary for genomics-assisted selection in non-target environments.

Abstract

Rose, one of the world’s most-loved and commercially important ornamental plants, is predominantly tetraploid, possessing four rather than two copies of each chromosome. This condition complicates genetic analysis, and so the majority of previous genetic studies in rose have been performed at the diploid level. However, there may be advantages to performing genetic analyses at the tetraploid level, not least because this is the ploidy level of most breeding germplasm. Here, we apply recently developed methods for quantitative trait loci (QTL) detection in a segregating tetraploid rose population (F₁ = 151) to unravel the genetic control of a number of key morphological traits. These traits were measured both in the Netherlands and Kenya. Since ornamental plant breeding and selection are increasingly being performed at locations other than the production sites, environment-neutral QTL are required to maximise the effectiveness of breeding programmes. We detected a number of robust, multi-environment QTL for such traits as stem and petiole prickles, petal number and stem length that were localised on the recently developed high-density SNP linkage map for rose. Our work explores the complex genetic architecture of these important morphological traits at the tetraploid level, while helping to advance the methods for marker–trait exploration in polyploid species.

See: https://link.springer.com/article/10.1007/s00122-018-3132-4

Figure 2: Comparison of the QTL results using a one-stage linear mixed model approach versus a two-stage analysis using BLUEs in an ordinary linear model. Significance thresholds, as determined by permutation tests (N = 1000, α = 0.05), are shown as dashed red lines (data were re-scaled so that these overlap). The one-stage results (− log₁₀(p) values) are shown in blue, with the two-stage results (LOD scores) shown in red (color figure online)