|Publication Type:||Journal Article|
|Year of Publication:||2011|
|Authors:||Zhang, Y, Li, Y, Wang, Y, Peng, B, LIU, CHENG, Liu, Z, Tan, W, WANG, DI, Shi, Y, SUN, BAOCHENG, Song, Y, Wang, T, Li, Y|
|Keywords:||cross-validation analysis, maize, marker-assisted selection, quantitative trait loci|
With 2 figures and 4 tablesAbstract The magnitude of genotypic correlations between family per se and testcross, r^g (per se, TC), is crucial for testing schemes in hybrid breeding as well as improvement of commercial hybrids. The objectives of this study were to identify the r^g (per se, TC), to examine the proportion of common QTL for both types of progenies and to determine the gene action of QTL identified for family per se and their value for the prediction of testcross progenies. Under multi-environments, r^g (per se, TC) were significant and higher than r^p (per se, TC), the phenotypic correlations between family per se and testcross for plant height and ear height in two maize crosses. The final fit for QTL explained 21.6–70.3% phenotypic variance and 24.4–79.4% genotypic variance for family per se. However, in cross-validation analysis (CV), the genotypic variance explained by all QTL was less than 60% for the two traits. Several common QTL (Chr. 1, Chr. 8 and Chr. 10) could be detected for both family per se and TC, which seems feasible to apply marker-assisted selection in maize breeding programmes.
|Short Title:||Plant Breeding|