TY - JOUR T1 - Epistatic genetic architecture of root length in <em>Arabidopsis thaliana</em> JF - bioRxiv DO - 10.1101/008789 SP - 008789 AU - Jennifer Lachowiec AU - Xia Shen AU - Christine Queitsch AU - Örjan Carlborg Y1 - 2014/01/01 UR - http://biorxiv.org/content/early/2014/10/08/008789.abstract N2 - Efforts to identify loci underlying complex traits generally assume that most genetic variance is additive. This is despite the fact that non-additive genetic effects, such as epistatic interactions and developmental noise, are also likely to make important contributions to the phenotypic variability. Analyses beyond additivity require additional care in the design and collection of data, and introduce significant analytical and computational challenges in the statistical analyses. Here, we have conducted a study that, by focusing on a model complex trait that allows precise phenotyping across many replicates and by applying advanced analytical tools capable of capturing epistatic interactions, overcome these challenges. Specifically, we examined the genetic determinants of Arabidopsis thaliana root length, considering both trait mean and variance as phenotypes. Estimation of the narrow- and broad-sense heritabilities of mean root length found that only the non-additive variance was significantly different from zero. Also, no loci were found to contribute to mean root length using a standard additive model based genome wide association analysis (GWAS). We could, however, identify one locus regulating developmental noise (root length variance) and seven loci contributing to root-length mean through epistatic interactions. Four of the epistatic loci were also experimentally confirmed. The locus associated with root length variance contains a candidate gene that, when mutated, appears to decrease developmental noise. This is particularly interesting as most other known noise regulators in multicellular organisms increase noise when mutated. A mutant analysis of candidate genes within the seven epistatic loci identified four genes that affected root development, including three without previously described root phenotypes. In summary, we identify several novel genes affecting root development, demonstrate the benefits of advanced analytical tools to study the genetic determinants of complex traits, and show that epistatic interactions can be a major determinant of complex traits in natural A. thaliana populations.Author summary Complex traits, such as many human diseases or climate adaptation and production traits in crops, arise through the action and interaction of many genes and environmental factors. Classic approaches to identify contributing genes generally assume that these factors act additively. Recent methods such as genome-wide association studies often adhere to this additive genetics paradigm. However, additive models of complex traits do not reflect that genes can interact non-additively. In this study, we use Arabidopsis thaliana to determine the additive and non-additive contributions to the complex trait root length. Surprisingly, much of the observed phenotypic variation in root length across genetically divergent strains was explained by genetic interactions. We mapped eight such non-additive loci and validated four epistatic genes using mutant analysis. For three of these, this is their first implication in root development. Together, our results emphasize the importance of considering both non-additive and additive effects when dissecting complex traits. ER -