Mapping density response in maize: A direct approach for testing genotype and treatment interactions

Maize yield improvement has been strongly linked to improvements in stress tolerance, particularly to increased interplant competition. As a result, modern hybrids are able to produce kernels at high plant population densities. Identification of the genetic factors responsible for density response in maize requires direct testing of interactions between genetic effects and density and evaluation of that response in multiple traits. In this article we take a broad view of the problem and use a general approach based upon mixed models to analyze data from eight segmental inbred lines in a B73 background and their crosses to the unrelated parent Mo17 (hybrids). We directly test for the interaction between treatment effects and genetic effects instead of the commonly used overlaying of results on a common map. Additionally, we demonstrate one way to handle heteroscedasticity of variances common in stress responses. We find that some SILs are consistently different from the recurrent parent regardless of the density, while others differ from the recurrent parent in one density level but not in the other. Thus, we find positive evidence for both main effects and interaction between genetic loci and density in cases where the approach of overlapping results fails to find significant results. Furthermore, our study clearly identifies segments that respond differently to density depending upon the inbreeding level (inbred/hybrid).