Skip to main content
ARS Home » Southeast Area » Raleigh, North Carolina » Plant Science Research » Research » Publications at this Location » Publication #222182

Title: Maize Phenomics: Massively Parallel Phenotyping of the Nested Association Mapping Population

item Holland, Jim - Jim

Submitted to: Maize Genetics Conference Abstracts
Publication Type: Abstract Only
Publication Acceptance Date: 1/10/2008
Publication Date: N/A
Citation: N/A

Interpretive Summary: N/A

Technical Abstract: How does the tremendous genetic variation within maize cause the phenotypic diversity displayed in this species? A goal of the Maize Diversity Project is the development of the genetic resources and methods to answer this question. We created a Nested Association Mapping (NAM) population consisting of 25 new recombinant inbred line (RIL) subpopulations, each derived from a cross between B73 (the reference parent) and one of 25 inbred lines selected to capture much of the diversity available globally among inbred lines. We created 200 RILs from each of the 25 crosses, and we are integrating the IBM population to form a 26th subpopulation. Each line has been genotyped at 1106 SNP markers. The combined NAM population of 5000 lines plus 200 IBM lines and an additional 281 diverse inbreds representing the maize association mapping platform has been phenotyped for up to 20 traits in up to 11 environments. Phenotyping nearly 5500 unique lines in presents many challenges, including the limitation that only one replication can be grown within each environment. To control error variation within environments, we employed a blocking design in which about 10% of the plots were planted to replicated check inbreds. Plots were labeled with bar-coded tags and trait data were collected with hand-held computers or scanning devices. Data analysis employs information on the repeated checks to adjust unreplicated RIL phenotypes for extraneous micro-environmental effects in the field. To obtain RIL phenotype values adjusted to remove extraneous error effects, we obtained Best Linear Unbiased Predictors (BLUPs) for each line. The BLUPs are the optimal phenotypic values to use for QTL mapping. The same analysis also permits estimation of important genetic architecture parameters such as trait heritabilities and the distribution of genetic variance among and within populations.