ORIGINAL PAPER Association mapping in an elite maize breeding population Wenxin Liu • Manje Gowda • Jana Steinhoff • Hans Peter Maurer • Tobias Wu ¨ rschum • Carl Friedrich Horst Longin • Fre ´de ´ric Cossic • Jochen Christoph Reif Received: 23 December 2010 / Accepted: 31 May 2011 / Published online: 17 June 2011 Ó Springer-Verlag 2011 Abstract 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 detec- tion power and possibility to correct for population strati- fication. 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 fin- gerprinted with 425 SNP markers. For grain yield, popu- lation 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 popula- tions. 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. Introduction Dissection of quantitative traits with association mapping (AM) is a promising strategy (Yu et al. 2008; Lu et al. 2010). AM in plant genetics is commonly based on several segregating bi-parental populations, which are analyzed together (e.g., Wu et al. 2002; Stich et al. 2008; Kover et al. 2009; Reif et al. 2010). The controlled crosses result in more balanced allele frequencies compared with associa- tion mapping in a diverse panel of lines by enhancing QTL detection power and decreasing confounding effects of population structure (McMullen et al. 2009). The resolution of AM in multiple segregating popula- tions and the required marker density for a genome-wide scan for QTLs are defined by the extent of linkage dis- equilibrium (LD) present in the population of the parental genotypes (Myles et al. 2009). In natural populations LD is a function of mutation, recombination, selection, migra- tion, and mating pattern (Flint-Garcia et al. 2003). There- fore, it has to be examined empirically for the germplasm under study to determine the number of required markers for a genome-wide scan for QTLs (Rafalski 2002). In heterotic groups of European elite maize breeding pro- grams a high extent of LD was observed based on a limited number of (approximately 100) SSR markers (Reif et al. 2005; Stich et al. 2005). This suggests that genome-wide AM in breeding populations is feasible even with a Communicated by J. Yan. Electronic supplementary material The online version of this article (doi:10.1007/s00122-011-1631-7) contains supplementary material, which is available to authorized users. W. Liu  M. Gowda  J. Steinhoff  H. P. Maurer  T. Wu ¨rschum  C. F. H. Longin  J. C. Reif (&) State Plant Breeding Institute, University of Hohenheim, 70593 Stuttgart, Germany e-mail: jochreif@uni-hohenheim.de F. Cossic Syngenta Seeds SAS, 12, chemin de l’Hobit, B.P. 27, 31790 Saint-Sauveur, France W. Liu Crop Genetics and Breeding Department, China Agricultural University, 100193 Beijing, China 123 Theor Appl Genet (2011) 123:847–858 DOI 10.1007/s00122-011-1631-7