Please note that we have stopped the regular imports of Gene Expression Omnibus (GEO) data into ArrayExpress. This may not be the latest version of this experiment.
E-GEOD-83936 - Massively parallel interrogation of the effects of gene expression levels on cellular fitness
Released on 25 August 2016, last updated on 12 September 2016
Data of gene expression levels across individuals, cell types, and disease states is rapidly expanding, yet we have limited understanding of how expression levels impact cellular and organismal phenotypes. Here, we present a massively parallel system for assaying the effect of gene expression levels on cellular fitness in Saccharomyces cerevisiae by systematically altering the expression level of each of ~100 endogenous genes at ~100 distinct expression levels spanning a 500-fold range at high resolution. Our results show that the relationship between expression levels and growth is gene- and environment-specific, with the specific relationship exhibited by each gene being highly informative on its function, stoichiometry within complexes, and interaction with other genes. Notably, in one of the two environmental conditions that we tested, we find that ~20% of the genes have expression levels where fitness is greater than that at wild-type expression levels, indicating that wild-type expression is not optimal for growth in that condition. We find that genes whose fitness is greatly affected by small changes in expression level tend to exhibit lower cell-to-cell variability in expression, suggesting that noise in gene expression is shaped in part by the relationship between expression and fitness. Overall, our study addresses a fundamental gap in our understanding of the functional significance of gene expression regulation and offers a powerful framework for evaluating the phenotypic effects of expression variation. 130 synthetic promoters were genomically integrated upstream of 96 endogenous yeast genes to span an expression range for each gene. Fitness as a function of the expression level of each gene was computed by a pooled growth competition assay.
Leeat Keren <email@example.com>, Adina Weinberger, Eran Segal, Hadas Alisar, Ilya Vainberg Slutskin, Jean Hausser, Maya Lotan-Pompan, Ron Milo, Sivan Kaminski, Uri Alon