Tuesday, November 14, 2006
162-12
Enhanced Detection of Inflorescence Architecture QTL in Intermated B73 x Mo17 (IBM) RIL Population.
Torbert Rocheford, AW-101 Turner Hall, 1102 S Goodwin Ave, Urbana, IL 61801 and Narasimham Upadyayula, University of Illinois, AW-101 Turner Hall, 1102 S Goodwin Ave, Urbana, IL 61801.
Genetic control of maize inflorescence architecture is of interest to developmental and evolutionary biologists as well as crop scientists. A number of maize inflorescence mutants have been identified and genes cloned, providing useful resources for developmental, evolutionary and breeding related studies in maize and across the cereals. QTL analysis is complementary and is identifying many more loci that influence inflorescence architecture. We summarize a series of QTL studies, four in standard F2:3 or BC1:S1 populations, and one in a population that was random mated for four cycles and recombinant inbred lines derived. We measured several traits on tassel and ear, and also calculate traits such as ratio of branch number to spikelet number. In standard QTL mapping populations we detect a couple to several QTL for each trait, with total numbers approximately 30 to 50. In contrast, in the IBM population, depending on permutation generated LOD threshold used, we detect approximately 300 – 500 total QTL for 13 different tassel infloresence architecture traits. This demonstrates the power of IBM population. In some cases two QTL are detected in a region where a single QTL was detected in a non-random mated population. Due to the highly correlated nature of inflorescence architecture traits, we performed multivariate principal components analysis (PCA) on the phenotypic data, and also used the PC values as traits and performed QTL analysis on them. This resulted in detection of fewer QTL, 125 in IBM, which may be relatively important in quantitative control of overall inflorescence architecture variation. We identified QTL that map to relevant mutant loci such as ts4, lg3, fea2, td1, ra1. However, the vast majority of QTL are not near inflorescence mutants and genes, which provides initial information for new gene discovery approaches.