ORIGINAL PAPER Application of identified QTL-marker associations in rice quality improvement through a design-breeding approach Jiankang Wang Æ Xiangyuan Wan Æ Huihui Li Æ Wolfgang H. Pfeiffer Æ Jonathan Crouch Æ Jianmin Wan Received: 28 May 2006 / Accepted: 10 April 2007 / Published online: 4 May 2007 Ó Springer-Verlag 2007 Abstract A permanent mapping population of rice con- sisting of 65 non-idealized chromosome segment substitu- tion lines (denoted as CSSL1 to CSSL65) and 82 donor parent chromosome segments (denoted as M1 to M82) was used to identify QTL with additive effects for two rice quality traits, area of chalky endosperm (ACE) and amylose content (AC), by a likelihood ratio test based on stepwise regression. Subsequently, the genetics and breeding simu- lation tool QuLine was employed to demonstrate the appli- cation of the identified QTL in rice quality improvement. When a LOD threshold of 2.0 was used, a total of 16 chro- mosome segments were associated with QTL for ACE, and a total of 15 segments with QTL for AC in at least one envi- ronment. Four target genotypes denoted as DG1 to DG4 were designed based on the identified QTL, and according to low ACE and high AC breeding objectives. Target geno- types DG1 and DG2 can be achieved via a topcross (TC) among the three lines CSSL4, CSSL28, and CSSL49. Re- sults revealed that TC2: (CSSL4 · CSSL49) · CSSL28 and TC3: (CSSL28 · CSSL49) · CSSL4 resulted in higher DG1 frequency in their doubled haploid populations, whereas TC1: (CSSL4 · CSSL28) · CSSL49 resulted in the highest DG2 frequency. Target genotypes DG3 and DG4 can be developed by a double cross among the four lines CSSL4, CSSL28, CSSL49, and CSSL52. In a double cross, the order of parents affects the frequency of target genotype to be selected. Results suggested that the double cross between the two single crosses (CSSL4 · CSSL28) and (CSSL49 · CSSL52) resulted in the highest fre- quency for DG3 and DG4 genotypes in its derived doubled haploid derivatives. Using an enhancement selection meth- odology, alternative ways were investigated to increase the target genotype frequency without significantly increasing the total cost of breeding operations. Introduction The rapid progress in the development of polymorphic molecular markers has led to the intensive use of QTL mapping in genetic study for quantitative traits (Lander and Botstein 1989; Zeng 1994; Piepho 2000; Dekkers and Hospital 2002; van Eeuwijk et al. 2002; Wan et al. 2004, 2005; Wang et al. 2006). The mapping populations re- quired for QTL detection, such as F 2 , backcross, recom- bination inbred lines (RIL), and doubled haploids (DH), can be classified into two categories, temporary popula- tions and permanent populations. In a permanent popula- tion such as RIL and DH, each individual in the population is genetically homozygous at all loci, and the genetic composition will not change during self-pollination. Thus, permanent populations allow precisely determining the phenotype of complex quantitative traits through replicated Communicated by F. van Eeuwijk. J. Wang  X. Wan  H. Li  J. Wan (&) Institute of Crop Science and The National Key Facility for Crop Gene Resources and Genetic Improvement, Chinese Academy of Agricultural Sciences, No. 12 Zhongguancun South Street, Beijing 100081, China e-mail: wanjm@caas.net.cn J. Wang  H. Li  J. Crouch Crop Research Informatics Laboratory (CRIL) and Genetic Resources Enhancement Unit (GREU), International Maize and Wheat Improvement Center (CIMMYT), Apdo. Postal 6-641, 06600 Mexico DF, Mexico W. H. Pfeiffer HarvetPlus and International Center for Tropical Agriculture (CIAT-HarvestPlus), A.A. 6713, Cali, Colombia 123 Theor Appl Genet (2007) 115:87–100 DOI 10.1007/s00122-007-0545-x