Image analysis for QTL mapping of flower colour and leaf characteristics in pot azalea (Rhododendron simsii hybrids)

Abstract

In azalea breeding, flower colour is the most essential selection criterion, but leaf morphology also influences attractiveness of the plant. Despite extensive study of the inheritance of flower colour, no explanation has yet been found for the pink phenotype. We have used image analysis to quantify flower colour and leaf morphology. Flower colour was quantified in a whole population. Pink flower colour has been confirmed to be the result of a gene-dosage effect; two major QTLs were found for flower colour as well as some minor QTLs that seem to be related to pink coloration. Leaf morphology (colour and shape) was scored in four unrelated populations by means of image analysis. The image analysis generated continuous, highly informative data for QTL mapping. Both classical parameters and symmetrical elliptic Fourier descriptors successfully described leaf morphology. Image analysis resulted in large data sets that had to be combined in principal components. Only a limited number of QTLs were found for leaf colour, but we could discern some major QTLs for leaf shape and size that were consistent over the different mapping populations. These QTLs are the most interesting candidates for future analysis of the selected traits. The multi-population approach certainly proved to be valuable for QTL mapping of complex traits. For leaf morphology, however, more research is needed to identify the most valuable QTLs for future use in MAS.