3D phenotyping and QTL analysis of a complex character: rose bush architecture

Plant shape, and thereby plant architecture, is a major component of the visual quality of ornamental plants. We have been developing a new method for analyzing the entire plant architecture by 3D digitalization that allows an almost exhaustive description of rose bush architecture and generates a large number of variables, many of them inaccessible manually. We carried out a QTL analysis using this original phenotyping method. In order to evaluate a broader allelic variability as well as the effect of the genetic background on QTL detection, we used two connected, segregating, recurrent blooming populations. The number of QTLs per variable varied from three for the number of determined axes (NbDetA) to seven for the branching angle of order 2 long axes (AngLA2), the two populations taken together. Five new QTLs, located on the linkage groups (LGs) 2, 6, and 7, were detected for the branching angle of axes, and the QTL located on LG7 co-localized with RhBRC1, a branching repressor. Branching and stem elongation QTLs also co-located with RhBRC1, suggesting its pleiotropic nature. Year-specific QTLs were also revealed, that explained the genotype × year interactions observed for the number of order 3 short axes (NbSA3) and AngLA2 from a genetic point of view. We also evidenced an effect of the genetic background on QTL detection. This new knowledge should help to better reason the genetic improvement programs for rose bush architecture and, therefore, rose bush shape.