Quantitative analysis of fitness and genetic interactions in yeast on a genome scale

Abstract

Global quantitative analysis of genetic interactions is a powerful approach for deciphering the roles of genes and mapping functional relationships among pathways. Using colony size as a proxy for fitness, we developed a method for measuring fitness-based genetic interactions from high-density arrays of yeast double mutants generated by synthetic genetic array (SGA) analysis. We identified several experimental sources of systematic variation and developed normalization strategies to obtain accurate single- and double-mutant fitness measurements, which rival the accuracy of other high-resolution studies. We applied the SGA score to examine the relationship between physical and genetic interaction networks, and we found that positive genetic interactions connect across functionally distinct protein complexes revealing a network of genetic suppression among loss-of-function alleles.

Figure 1

The SGA score for measuring quantitative genetic interactions. (a) An SGA experiment crossing a strain carrying a query mutation to an input array of single mutants, each of which carries a wild-type copy of the query gene and a unique array strain mutation. A final output array of double mutants is generated after several SGA selection steps, photographed and processed using software that measures colony areas in terms of pixels. Relative colony size, determined by measuring deviation of individual colonies from the median size for the same colony across 1,712 different experiments, is shown. (b) Schematic depiction of the five factors that contribute to experimental variance of colony size. (c) Relative colony size after normalization. Single-mutant fitness (W_A, W_B) and double-mutant fitness (W_AB) derived from normalized colony size measurements were used to identify and measure genetic interactions (SGA score; ε).

Figure 2

Evaluation of single-mutant fitness measures. (a) Comparison of single-mutant fitness or relative growth measurements for nonessential gene deletions derived from eight independent approaches: colony size measurements (this study), competitive growth analyzed by barcode hybridization, flow cytometry or gene expression profiles, gene-expression microarrays, liquid growth profiling^, and a spot assay on solid growth medium (phenotypic array). (b) Distribution of single-mutant fitness measures reported by the studies described in a reporting fitness or relative growth rate. (c) Correlation of double-mutant fitness measures obtained from two independent replicates of a representative genome-wide SGA screen. Red line, y = x. Inset, distribution of correlations between double-mutant fitness measures obtained from 211 genome-wide SGA screens conducted in duplicate.

Figure 3

Evaluation of quantitative genetic interactions. (a) Scatter plots of genetic interactions derived from 211 genome-wide SGA screens conducted in duplicate. Pearson correlation coefficients were computed after applying a lenient (P < 0.05) or intermediate (SGA score absolute value, |ε| > 0.08; P < 0.05) confidence threshold on the SGA score. (b) Scatter plots of genetic interaction measures between reciprocally tested gene pairs. Pearson correlation coefficients were computed after applying a lenient (P < 0.05) or intermediate (|ε| > 0.08, P < 0.05) confidence threshold on the SGA score. (c) A scatter plot illustrating the overlap between genetic interaction scores for 239 unique gene pairs extracted from a large-scale SGA dataset and a small-scale, high-resolution liquid growth profiling study. Data in c were adapted from reference .

Figure 4

Evaluation of functional information derived from genetic interactions. (a) Plots of precision versus recall (number of true positives (TP)) for negative and positive genetic interactions, as determined by the SGA score or the S score. An SGA score without normalization methods applied is also plotted. True positive interactions were defined as those involving both genes annotated to the same GO gold standard set of terms. Precision and recall were calculated as described previously. FP, false positive. (b) Plots of precision versus recall (number of TP) for genetic interaction profile similarities computed using the SGA score or the S score. An SGA score without normalization methods applied is also plotted. Pearson correlation was used to compute profile similarity for every pair of array mutant strains across profiles consisting of interactions with the 1,712 query mutant strains. True positive pairs were those for which both genes were annotated to the same GO gold standard set of terms (GO annotation) or pairs encoding physically interacting proteins (physical interaction standard). Precision and recall were calculated as described previously.

Figure 5

Analysis of genetic interactions within and between protein complexes. (a) For 92 complexes enriched for negative and/or positive genetic interactions that we assembled, nonessential genes are represented as circles and essential genes as diamonds. Complexes connected by purely positive or negative genetic interactions are indicated by yellow and blue, respectively; gray denotes complexes connected by a mixture of positive and negative genetic interactions. (b) Degree analysis of the complex-complex genetic interaction network in which node color reflects the prevalence of positive (yellow) or negative (blue) genetic interactions in a complex. Gray nodes denote complexes for which too few gene pairs were screened to assess within-complex interactions. Node size indicates the number of proteins associated with the complex. Positive and negative degree are the number of positive and negative genetic interactions, respectively, for a given complex. Inset, number of between-complex interactions was measured for complexes connected by purely negative and purely positive genetic interactions. Error bars, s.e.m. (n = 37, positive and n = 25, negative; *P < 0.002 by a rank-sum test).

Figure 6

Cross-complex genetic suppression network revealed by quantitative genetic interaction analysis. (a) A network illustrating suppression interactions between protein complexes (nodes). Edges indicate positive SGA interactions classified as suppression; arrows point to the complex whose fitness defect was suppressed. (b) Colony size–derived single- and double-mutant fitness plotted for the indicated strains (top). Error bars for single mutants, s.e.m. derived from bootstrapping (n = 800); error bars for double mutant, s.d. (n = 4). Liquid growth profiling of the same strains (bottom). (c) Schematic showing activation of the Rim101 pathway in response to alkaline stress. (d) Growth of the indicated strains on glucose and galactose, with plated strains indicated (left). WT, wild type. (e,f) Serial dilution growth assays of the indicated yeast strains at the indicated temperatures (e) and after exposure to UV light.