Global analysis of the yeast knock-out phenome

Abstract

Genome-wide phenotypic screens in the budding yeast Saccharomyces cerevisiae have produced the largest, richest and most systematic phenotypic description of any organism. Such an achievement was enabled by the development of highly scalable phenotypic assays and construction of the yeast knock-out (YKO) collection, comprising ~5,000 isogenic strains each deleted for exactly one open reading frame. Systematic screening of the YKO collection led to ~500 publications describing ~14,500 phenotypes capturing nearly every aspect of yeast biology. Yet, integrative analyses of this rich data source have been virtually impossible due to the lack of a central repository and consistent meta-data annotations. Here, we describe the aggregation, harmonization and analysis of all published phenotypic screens of the YKO collection, which we refer to as the Yeast Phenome (www.yeastphenome.org). To demonstrate the power of data integration and illustrate how much it facilitates the generation of testable hypotheses, we present three discoveries uniquely enabled by Yeast Phenome. First, we use the variation in the number of phenotypes per gene to identify tryptophan homeostasis as a central point of vulnerability to a wide range of chemical compounds, including FDA-approved drugs. Second, using phenotypic profiles as a tool for predicting gene function, we identify and validate the role of YHR045W as a novel regulator of ergosterol biosynthesis and DNA damage response, and YGL117W as a new member of the aromatic amino acid biosynthesis pathway. Finally, we describe a surprising exponential relationship between phenotypic similarity and intergenic distance in both yeast and human genomes. This relationship, which stretches as far as 380 kb in yeast and 100 Mb in humans, suggests that gene positions are optimized for function to a much greater extent than appreciated previously. Overall, we show that Yeast Phenome enables systematic enquiries into the nature of gene-gene and gene-phenotype relationships and is an important new resource for systems biology.

Systematic Profiling of Ale Yeast Protein Dynamics across Fermentation and Repitching

“…Fig. 1E, adjusted R 2 =- 0.04194, p≤0.5392). There was also no correlation at any of the later time points (Sup.…”

“…1D, Sup. Table 3, adjusted R 2 =0.00008246, p≤0.2703), nor by comparing the average read depth per gene to the average abundance of the corresponding protein product for all genes on each chromosome (Sup. Fig.…”

“…D) Comparison between read depth (x-axis) at each gene and level of the corresponding protein abundance (y-axis). Adjusted R 2 =0.00008246, p≤0.2703. E) Comparison between average read depth/chromosome (x-axis) and average LFQ abundance of corresponding protein by chromosome (y-axis), adjusted R 2 =-0.04194, p≤0.5392.…”

“…Adjusted R 2 =0.00008246, p≤0.2703. E) Comparison between average read depth/chromosome (x-axis) and average LFQ abundance of corresponding protein by chromosome (y-axis), adjusted R 2 =-0.04194, p≤0.5392.…”

“…The ease of culture and genetic manipulation of S. cerevisiae has also made it one of the most profiled organisms in academic research, including being the first eukaryote to have its genome sequenced 1 . Brewing yeast has served as a model organism to understand fundamental cellular and molecular processes and continues to provide valuable insights into human health and disease [2][3][4][5][6][7] . Despite its prevalence in laboratory settings, the systematic profiling of yeast in commercial contexts is less common.…”