Species-wide survey of the expressivity and complexity spectrum of traits in yeast

Assessing the complexity and expressivity of traits at the species level is an essential first step to better dissect the genotype-phenotype relationship. As trait complexity behaves dynamically, the classic dichotomy between monogenic and complex traits is too simplistic. However, no systematic assessment of this complexity spectrum has been carried out on a population scale to date. In this context, we generated a large diallel hybrid panel composed of 190 unique hybrids coming from 20 natural isolates representative of the S. cerevisiae genetic diversity. For each of these hybrids, a large progeny of 160 individuals was obtained, leading to a total of 30,400 offspring individuals. Their mitotic growth was evaluated on 38 conditions inducing various cellular stresses. We developed a classification algorithm to analyze the phenotypic distributions of offspring and assess the trait complexity. We clearly found that traits are mainly complex at the population level. On average, we found that 91.2% of cross/trait combinations exhibit high complexity, while monogenic and oligogenic cases accounted for only 4.1% and 4.7%, respectively. However, the complexity spectrum is very dynamic, trait specific and tightly related to genetic backgrounds. Overall, our study provided greater insight into trait complexity as well as the underlying genetic basis of its spectrum in a natural population. Dissecting the genetic origins of natural phenotypic variation is a major goal in biology. In 1865, Gregor Mendel established principles of inheritance that described the transmission of genetic traits. However, we still lack a precise view of the spectrum and continuum of trait complexity in natural population. In this context, we carried out a study of the complexity of traits in a large population of isolates using the yeast Saccharomyces cerevisiae. We analyzed patterns of distribution and inheritance of offspring of a wide diallel panel and in a large number of environments. We found that on average 91.2% of the traits are complex, while only 4.1% and 4.7% are monogenic and oligogenic, respectively. However, it is also clear that the complexity spectrum depends on genetic background and environment. Interestingly, we have highlighted and dissected the genetic basis of cases showing a broad complexity spectrum, such as in the presence of copper sulfate as well as galactose as a carbon source.