Open Access Highly Accessed Open Badges Research article

Genetic architecture of complex agronomic traits examined in two testcross populations of rye (Secale cereale L.)

Thomas Miedaner1*, Marlen Hübner1, Viktor Korzun2, Brigitta Schmiedchen2, Eva Bauer3, Grit Haseneyer3, Peer Wilde2 and Jochen C Reif1

Author affiliations

1 State Plant Breeding Institute, Universität Hohenheim, Stuttgart, 70593, Germany

2 KWS LOCHOW GmbH, Bergen, 29303, Germany

3 Plant Breeding, Technische Universität München, Freising, 85354, Germany

For all author emails, please log on.

Citation and License

BMC Genomics 2012, 13:706  doi:10.1186/1471-2164-13-706

Published: 17 December 2012



Rye is an important European crop used for food, feed, and bioenergy. Several quality and yield-related traits are of agronomic relevance for rye breeding programs. Profound knowledge of the genetic architecture of these traits is needed to successfully implement marker-assisted selection programs. Nevertheless, little is known on quantitative loci underlying important agronomic traits in rye.


We used 440 F3:4 inbred lines from two biparental populations (Pop-A, Pop-B) fingerprinted with about 800 to 900 SNP, SSR and/or DArT markers and outcrossed them to a tester for phenotyping. The resulting hybrids and their parents were evaluated for grain yield, single-ear weight, test weight, plant height, thousand-kernel weight, falling number, protein, starch, soluble and total pentosan contents in up to ten environments in Central Europe. The quality of the phenotypic data was high reflected by moderate to high heritability estimates. QTL analyses revealed a total of 31 QTL for Pop-A and 52 for Pop-B. QTL x environment interactions were significant (P < 0.01) in most cases but variance of QTL main effect was more prominent.


QTL mapping was successfully applied based on two segregating rye populations. QTL underlying grain yield and several quality traits had small effects. In contrast, thousand-kernel weight, test weight, falling number and starch content were affected by several major QTL with a high frequency of occurrence in cross validation. These QTL explaining a large proportion of the genotypic variance can be exploited in marker-assisted selection programs and are candidates for further genetic dissection.