Genetic architecture of phenotypic indices for simultaneous improvement of protein content and grain yield in triticale (×triticosecale)

Abstract

Our study aimed to assess the potential of indices, including grain protein deviation and protein yield, to simultaneously improve grain yield and protein content in triticale. Our analysis of 1.218 triticale genotypes revealed significant genotypic variation and high heritabilities for all traits. We analysed the suitability of the indices for phenotypic selection for both high grain yield and increased protein content. We show that some genotypes were selected irrespective of the applied index, but also observed considerable differences between the indices. To uncover the genetic architecture of grain yield, protein content and derived indices, we performed genome-wide association mapping and biparental QTL mapping. This revealed many pleiotropic QTL with opposite effects on grain yield and protein content and a complex inheritance of all traits, suggesting that the genetic architecture limits the potential of marker-assisted selection. Nevertheless, the use of these indices is recommended for breeding of special-purpose cultivars, like high-protein cultivars for on-farm feed production.