A global understanding of genetic interaction networks, and how network perturbations affect cellular function, is crucial to preventing and treating human disease. Currently there is a fundamental gap in our understanding of these networks. Most of our knowledge of genetic interactions comes from the systematic analysis of double deletion (or knockdown) mutants, primarily in the yeast Saccharomyces cerevisiae. However, the reality is that loss-of-function mutations are rarely beneficial and account for less than 5% of the known natural genetic variation. Continued existence of this gap is a significant problem because many biomedically-important interactions are likely missed by current methods. The proposed research will identify genetic interactions involving alteration-of-function variants, variants of essential genes, and higher-order interactions using a novel ?Evolve-and-Map? approach, which combines experimental evolution and quantitative-trait locus mapping. The rationale for this approach is that experimental evolution efficiently selects for perturbations to the genetic interaction network that promote rapid growth, and that the genetic variants isolated in this way will be comparable to the natural genetic variants underlying complex traits in other organisms, including humans.
AIM 1 will leverage the power of evolutionary ?replay? experiments to identify a local network of genetic interactions between cell polarity genes and cell cycle genes. These interactions are strongly supported by preliminary laboratory evolution experiments, but are largely absent from the double-deletion genetic interaction network.
AIM 2 will extend this analysis genome-wide, producing the largest data set to date on the genetic interactions between variants that arose in the context of experimental evolution. Thousands of double-barcoded segregants will be generated from crosses between evolved lines and their ancestor or between pairs of evolved lines. Each segregant will be genotyped by low-coverage sequencing and its fitness will be measured using a highly-multiplexed barcode-sequencing-based assay that is capable of measuring the fitness of thousands of segregants en masse. These data will be used to detect additive effects as well as pairwise and three-way genetic interactions. Since these mapping populations contain far fewer variants than is typical in a genome-wide scan, the power of this method to detect genetic interactions is very high.
AIM 3 will determine the extent to which these genetic interactions persist across environments, including different carbon and nitrogen sources, inhibitory concentrations of antifungals, and non-optimal temperatures. This will add an important new dimension to genetic interaction networks. Overall the results obtained from this work will test the ability of the double-deletion genetic interaction network to predict interactions between growth-promoting variants, and will advance our understanding of genetic interaction networks and the evolution of complex traits.
Many human diseases arise through the contributions of multiple genetic variants and environmental factors that collectively perturb genetic-interaction networks. This project will identify interactions between hundreds of growth promoting variants, including alteration-of-function variants and variants of essential genes, which are hidden from methods that survey pairwise interactions between loss-of-function mutations. These insights will advance our understanding of genetic-interaction networks in other organisms, including humans, and will aid in identifying disease-causing variants and potential therapeutic targets.
