Association mapping (AM) is a powerful approach to dissect the genetic architecture of quantitative traits. The main goal of our study was to empirically compare several statistical methods of AM using data of an elite maize breeding program with respect to QTL detection power and possibility to correct for population stratification. These models were based on the inclusion of cofactors (Model A), cofactors and population effect (Model B), and SNP effects nested within populations (Model C). A total of 930 testcross progenies of an elite maize breeding population were field-evaluated for grain yield and grain moisture in multi-location trials and fingerprinted with 425 SNP markers. For grain yield, population stratification was effectively controlled by Model A. For grain moisture with a high ratio of variance among versus within populations, Model B should be applied in order to avoid potential false positives. Model C revealed large differences among allele substitution effects for trait-associated SNPs across multiple plant breeding populations. This heterogeneous SNP allele substitution effects have a severe impact for genomic selection studies, where SNP effects are often assumed to be independent of the genetic background.

Multiple-line cross quantitative trait locus (QTL) mapping is considered a promising tool to detect QTL with high power and substantial accuracy. The main goal of this study was to investigate the benefits of combined QTL analysis by applying two biometrical models compared to single-population analyses. For the combined QTL analysis we used a biometrical model that assumes alíele substitution effects specific for every biparental popuiation (disconnected model). We also applied a biometricai model that assumes alíele substitution effects specific for every parent (connected model). Six testeross populations of maize (Zea mays L.) derived from a diallel cross of four parents were tested in 10 Italian environments in 2007 for grain yield and grain moisture. The 788 genotypes were fingerprinted with 857 single nucleotide polymorphism (SNP) markers. Qur findings clearly underline the potential to improve the power to detect QTL and the resolution to localize the QTL in the genome by switching from singie population QTL mapping toward joint QTL analysis across populations. The disconnected model outperformed the connected model with regard to the power to detect QTL. Consequently, our results suggest that the disconnected model is the model of choice for multiple-line cross QTL mapping in elite maize germplasm.

Flowering time is a fundamental quantitative trait in maize that has played a key role in the postdomestication process and the adaptation to a wide range of climatic conditions. Flowering time has been intensively studied and recent QTL mapping results based on diverse founders suggest that the genetic architecture underlying this trait is mainly based on numerous small-effect QTL. Here, we used a population of 684 progenies from five connected families to investigate the genetic architecture of flowering time in elite maize. We used a joint analysis and identified nine main effect QTL explaining approximately 50 % of the genotypic variation of the trait. The QTL effects were small compared with the observed phenotypic variation and showed strong differences between families. We detected no epistasis with the genetic background but four digenic epistatic interactions in a full 2-dimensional genome scan. Our results suggest that flowering time in elite maize is mainly controlled by main effect QTL with rather small effects but that epistasis may also contribute to the genetic architecture of the trait.

Transfer of elite lines across maturity zones is important because it facilitates the exploitation of indirect selection gain. The main goal of this study was to investigate strategies to guide the transfer of elite lines from Southern Europe to the U.S. Corn Belt. Testcrosses of progenies of six biparental populations derived from a diallel cross of four Southern European elite lines were evaluated together with adapted commercial checks in 10 Southern European and six U.S. Corn Belt environments in 2007 for grain yield and grain moisture when crossed to adapted testers. Moreover, the 788 genotypes were fingerprinted with 857 single nucleotide polymorphism (SNP) markers and multiple‐line cross quantitative trait locus (QTL) mapping was performed. Some testcross progenies in the U.S. Corn Belt reached grain yield performance comparable to the best check, which suggests that direct use of Southern European lines is promising. The success of using grain yield or grain moisture data determined in Southern Europe to preselect Southern European lines for use in the U.S. Corn Belt is limited. Moreover, we observed a complex genetic architecture of adaptation with absence of major QTL and strong QTL by background interactions. We found evidence that epistasis influences adaptation, additionally hampering the success of marker‐guided transfer of germplasm from Southern Europe to the U.S. Corn Belt.