Yeast Phenomics: An Experimental Approach for Modeling Gene Interaction Networks that Buffer Disease.

John Hartman,Sean Santos,Chandler Stisher,Jingyu Guo,Richard White,Darryl Outlaw,John Rodgers,Dehua Tian,Najaf Shah

doi:10.3390/genes6010024

Abstract

The genome project increased appreciation of genetic complexity underlying disease phenotypes: many genes contribute each phenotype and each gene contributes multiple phenotypes. The aspiration of predicting common disease in individuals has evolved from seeking primary loci to marginal risk assignments based on many genes. Genetic interaction, defined as contributions to a phenotype that are dependent upon particular digenic allele combinations, could improve prediction of phenotype from complex genotype, but it is difficult to study in human populations. High throughput, systematic analysis of S. cerevisiae gene knockouts or knockdowns in the context of disease-relevant phenotypic perturbations provides a tractable experimental approach to derive gene interaction networks, in order to deduce by cross-species gene homology how phenotype is buffered against disease-risk genotypes. Yeast gene interaction network analysis to date has revealed biology more complex than previously imagined. This has motivated the development of more powerful yeast cell array phenotyping methods to globally model the role of gene interaction networks in modulating phenotypes (which we call yeast phenomic analysis). The article illustrates yeast phenomic technology, which is applied here to quantify gene X media interaction at higher resolution and supports use of a human-like media for future applications of yeast phenomics for modeling human disease.

Highlights

Yeast gene interaction networks can provide insight to buffering and variable expression of disease when yeast phenomic experiments are designed within a cellular context analogous to human biology [10]
Given this genetic complexity of phenotypic expression, yeast offers key advantages for mapping gene interaction as comprehensively and quantitatively as possible with respect to both environment and other genes: (1) Much of the overall fitness is encapsulated by the phenotype of cell proliferation, which lends comprehensiveness; and (2) Cell proliferation is a continuous trait that’s straightforward to quantify, where analysis with a logistic growth function resolves fitness into three components, providing additional resolution for phenomic analysis and gene interaction network construction (Figure 1)
Though phenotypes are more complex in humans than yeast, it is possible to extrapolate across species between potentially any phenotype, based upon gene interaction networks that function across evolution [6,7,8,9]

Summary

Introduction

Yeast Phenomic Analysis Reveals Gene Interaction Networks. We consider genetic buffering to underlie phenotypic stability/variability within a population and to derive from interaction between sets of gene variants and environmental factors, and that sets of buffering genes represent functional networks that distinctly modulate each phenotype [3]. In yeast one can experimentally define functional gene networks in terms of their capacity to buffer, or stabilize phenotypes. Buffering networks mask functional genetic variants subject to natural selection, and comprise a reservoir within populations for the complex expression of phenotypes [3]. Yeast gene interaction networks can provide insight to buffering and variable expression of disease when yeast phenomic experiments are designed within a cellular context analogous to human biology [10]

Methods

Results

Discussion

Conclusion