Analysis Of Flowering Time, Hybrid Vigor, Yield, And Lodging In Maize

Abstract

Maize (Zea mays L.) is an important crop and an excellent model organism to study genetic systems. It captures remarkable diversity, which can be observed on both the genotypic and phenotypic level. Because of its diversity, maize responded very effectively to artificial selection during domestication and improvement. Maize adapted to very diverse environments. This adaptation has been possible through heritable changes in flowering time, responses to photoperiod and temperature, and plant architecture. Understanding the underlying architecture of these traits will allow us to utilize all the variation offered and increase productivity for a more sustainable agriculture. The following studies focus on analysis of three different traits. First, is a reanalysis of one of the first generation structured association mapping studies of the Dwarf8 locus with flowering time, using new mapping populations and statistical approaches. This trait is highly correlated with population structure, and we found that basic structured association methods overestimate the phenotypic effect in the region, while mixed model approaches perform better. Combined with analysis of the maize NAM population, it is concluded that the QTL effects at the general location of the d8 locus are from extended haplotypes and that d8 is not associated with flowering time. Second, hybrids were developed using the NAM inbred population crossed to a common tester to examine hybrid vigor in terms of plant height and flowering time, as well as yield. A number of QTL were identified for all three traits using joint linkage mapping. Additionally, reasonable prediction accuracies (~0.55) were obtained using ridge regression in the hybrids. This study gives us a better understanding of yield and hybrid vigor. Last, damage caused by lodging is a significant problem in maize production, resulting in 5-20 % annual loss in yield. In this study, more than 1,500 diverse inbred lines crossed to a common tester were evaluated across multiple environments. Due to a large sample size and despite multiple environments with lodging events occurring at different points in time, we were able to utilize joint linkage mapping to identify a number of QTL with small effects for lodging.