Tessele, A.; Morris, G. P.; Akhunov, E.; Johnson, B. E.; Clinesmith, M.; Fritz, A. K.

Additive gene action is assumed to underly quantitative traits, but the eventual poor performance of elite wheat lines as parents suggests that epistasis could be the underlying genetic architecture. Sign epistasis is characterized by alleles having either a beneficial or detrimental effect depending on the genetic background, which can result in elite lines that fail as parents in certain parental combinations. Hence, the objective of this study were to test the existence of sign epistasis and examine its consequences to wheat breeding. The presence of sign epistasis is expected to distort the allele frequency distribution between two interacting genes compared to neutral sites, creating strong linkage disequilibrium (LD). To test this hypothesis, analysis of interchromosomal LD in breeding families was performed and detected 19 regions in strong disequilibrium, whose allele frequency distribution matched the sign epistasis prediction and falsified the competing hypothesis of additive selection. To validate these candidate interactions while avoiding the biases of a circular analysis and the confounding effects of genetic drift, two independent sets of populations were analyzed. Genetic drift was attributed to creating the sign epistasis patterns observed in eleven interactions, but there was not sufficient evidence to reject the sign epistasis hypothesis in eight interactions. Sign epistasis may explain the poor performance of elite lines as parents, as crossing lines with complementary allelic combination re-establishes epistatic variance in the offspring. Reduction in the effective population size in certain crosses may also occur when unfavorable sign epistatic combinations are deleterious. The potential existence of di-genic and higher order epistatic interactions in elite germplasm can tremendously impact breeding strategies as managing epistasis becomes imperative for success.