Fitness effects of altering gene expression noise in Saccharomyces cerevisiae
- University of Michigan, United States
- CNRS UMR 7057, Université Paris Diderot, France
- University of Chicago, United States
Figures
Figure 1 with 1 supplement
A collection of TDH3 promoter alleles with incompletely correlated effects on average expression level and expression noise.
(A) Overview of experimental design used to quantify expression. The transcriptional activity of 171 different variants of the TDH3 promoter (PTDH3) inserted upstream of the YFP coding sequence was …
-
Figure 1—source data 1
Parallel measurements of fluorescence levels by flow cytometry and of YFP mRNA levels by pyrosequencing.
Pyrosequencing data were analyzed with the R script provided in Supplementary file 3. Data used to make Figure 1B–C.
- https://doi.org/10.7554/eLife.37272.006
Figure 1—figure supplement 1
Median expression level and expression noise conferred by 171 variants of the TDH3 promoter using four different metrics of noise.
The four measures of expression noise are: (A) Noise strength, the variance divided by median fluorescence as in Figure 1 (B) SD, the standard deviation of fluorescence level among cells sharing the …
-
Figure 1—figure supplement 1—source data 1
Expression data for an initial set of 171 TDH3 promoter alleles.
Fluorescence levels were measured by flow cytometry and analyzed using the R script provided in Supplementary file 3. Data used to make Figure 1—figure supplement 1.
- https://doi.org/10.7554/eLife.37272.005
Figure 2 with 2 supplements
Fitness consequences of variation in TDH3 expression level.
(A) Overview of competition assays used to quantify fitness. The 43 PTDH3 alleles whose activity was described in Figure 1D were introduced upstream of the native TDH3 coding sequence in a genetic …
Figure 2—figure supplement 1
Comparing effects of 20 alleles of the TDH3 promoter on expression of the YFP reporter at HO and of expression of the TDH3-YFP fusion at the native TDH3 locus.
(A) Median expression and (B) expression noise (noise strength) were quantified in six replicates for 20 pairs of strains carrying 20 different TDH3 promoter variants inserted upstream of the YFP …
-
Figure 2—figure supplement 1—source data 1
Activity of 20 PTDH3 alleles driving expression of YFP at the HO locus compared to the activity of the same alleles driving expression of a TDH3-YFP gene fusion at the TDH3 locus.
The median and noise of activity of the 20 PTDH3 alleles were measured by flow cytometry and analyzed using the R script provided in Supplementary file 3. Data used to make Figure 2—figure supplement 1A–B.
- https://doi.org/10.7554/eLife.37272.009
Figure 2—figure supplement 2
No significant impact of the genetic background on the expression of the fluorescent reporter.
(A) Differences in drug resistance marker or mating type that exist between the strains used in the expression assays and the strains used in the fitness assays do not affect median expression. (B) …
Figure 3 with 8 supplements
Effect of TDH3 expression noise on fitness.
(A) Separation of the 43 PTDH3 alleles into two categories based on their effects on median expression level and expression noise (noise strength). The gray curve shows the LOESS regression of noise …
Figure 3—figure supplement 1
Calculation of ΔNoise and ΔFitness using four different metrics of noise.
The four metrics of noise were: (A,E,I) the variance of expression divided by the median expression (noise strength), (B,F,J) the standard deviation of expression among genetically identical cells …
Figure 3—figure supplement 2
Relationship between median expression level and four different metrics of ΔNoise.
ΔNoise was calculated as the residuals of a LOESS regression of expression noise on median expression, using four different metrics of noise: (A) Noise strength, equal to SD2/median (B) SD, the …
Figure 3—figure supplement 3
Relationship between median expression level and Δ Fitness.
ΔFitness was calculated as the residuals of a LOESS regression of fitness on median TDH3 expression. Genotypes were colored based on their level of noise measured as (A) Noise strength, equal to SD2/…
Figure 3—figure supplement 4
Relationship between ΔNoise and ΔFitness using four different metrics of noise.
ΔNoise was calculated as the residuals of a LOESS regression of expression noise on median expression, using four different metrics of noise: (A,E) Noise strength, equal to SD2/median (B,F) SD, the …
Figure 3—figure supplement 5
Robustness of the correlation between ΔNoise and ΔFitness to variation in the smoothing parameter of the LOESS regression used to compute ΔNoise.
ΔNoise was calculated as the residuals of a LOESS regression of expression noise on median expression, using four different metrics of noise: (A,B) Noise strength, equal to SD2/median (C,D) SD, the …
Figure 3—figure supplement 6
Robustness of the correlation between ΔNoise and ΔFitness to variation in the smoothing parameter of the LOESS regression used to compute ΔFitness.
ΔNoise was calculated as the residuals of a LOESS regression of expression noise on median expression, using four different metrics of noise: (A,B) Noise strength, equal to SD2/median (C,D) SD, the …
Figure 3—figure supplement 7
Robustness of the correlation between ΔNoise and ΔFitness to variation in the fitness threshold used to classify genotypes as far or close to optimum.
ΔNoise was calculated as the residuals of a LOESS regression of expression noise on median expression, using four different metrics of noise: (A,B) Noise strength, equal to SD2/median (C,D) SD, the …
Figure 3—figure supplement 8
Fitness, median expression and noise of genotypes with ΔNoise above +1% compared to genotypes with ΔNoise below −1%.
ΔNoise was calculated as the residuals of a LOESS regression of expression noise on median expression using four different metrics of noise: (A–C) Noise strength, equal to SD2/median (D–F) SD, the …
Figure 4 with 1 supplement
Direct competition between genotypes with different levels of noise but similar median levels of TDH3 expression in glucose.
(A–C) Different colors are used to distinguish pairs of genotypes (PTDH3 alleles) with different median expression levels. (A) Median expression level and expression noise (noise strength) for five …
-
Figure 4—source data 1
Fitness measured in direct competition assays between strains with low and high values of expression noise.
Competitive fitness was measured by pyrosequencing and analyzed using the R script provided in Supplementary file 3. Data used to make Figure 4.
- https://doi.org/10.7554/eLife.37272.022
Figure 4—figure supplement 1
Median expression level and expression noise for five pairs of genotypes that were competed directly against each other.
Four metrics of noise were used: (A) the variance of expression divided by the median expression (noise strength), (B) the standard deviation of expression among genetically identical cells (SD), (B)…
Figure 5 with 1 supplement
A simple model linking single cell expression levels to population fitness.
(A) In our model, the expression level of individual cells is randomly drawn from a normal distribution . is lower for a genotype with low expression noise (top, green line) and higher for a …
Figure 5—figure supplement 1
Single-cell division rates estimated using time-lapse microscopy.
(A) The image series shows selected frames from a time-lapse movie that captured a picture of cells every six minutes in which we have zoomed on one initial cell to illustrate how doubling times …
-
Figure 5—figure supplement 1—source data 1
Single-cell measures of doubling time in four strains with different median levels and noise of TDH3 expression.
Doubling time was determined from time-lapse microscopy data. These data were used to make Figure 5—figure supplement 1.
- https://doi.org/10.7554/eLife.37272.025
-
Figure 5—figure supplement 1—source data 2
Summary statistics for comparing the distributions of single-cell doubling time between genotypes with different expression noise levels.
These data were used to make Figure 5—figure supplement 1.
- https://doi.org/10.7554/eLife.37272.026
Figure 6 with 2 supplements
Simulating the effect of expression noise on fitness at different median expression levels.
(A) The linear function relating the expression level of single cells to their doubling time used for the first set of simulations. (B) Relationship between mean expression () and fitness at …
-
Figure 6—source data 1
Fitness data obtained by modeling the growth of cell populations with different levels of mean expression and expression noise.
Data used to make Figure 6B and D and generated with the Matlab code provided in Supplementary file 5.
- https://doi.org/10.7554/eLife.37272.030
Figure 6—figure supplement 1
Simulating the effect of two different metrics of expression noise on fitness at different median expression levels.
Population fitness was simulated for median expression levels ranging from 10% to 100% and for: (A) the standard deviation of expression ranging from 0.05 to 2 and the linear function …
Figure 6—figure supplement 2
Relationship between expression noise and fitness at different values of mean expression in simulations using a Gaussian function relating single cell expression to doubling time.
Three different noise metrics were used: (A) the noise strength , (B) the standard deviation , (C) the coefficient of variation . (A–C) Error bars show 95% confidence intervals of mean fitness …
Additional files
-
Supplementary file 1
Datasets generated using R scripts available in Supplementary file 3 and used to make Source Data files for figures.
- https://doi.org/10.7554/eLife.37272.031
-
Supplementary file 2
List of oligonucleotides used in this study.
- https://doi.org/10.7554/eLife.37272.032
-
Supplementary file 3
R scripts for the analysis of pyrosequencing and flow cytometry data used to determine the median expression, expression noise and fitness associated with different PTDH3 alleles.
- https://doi.org/10.7554/eLife.37272.033
-
Supplementary file 4
Zip folder containing all input files necessary to run the R scripts available in Supplementary file 3.
- https://doi.org/10.7554/eLife.37272.034
-
Supplementary file 5
Matlab code used to model the growth of cell populations with different levels of mean expression and expression noise.
- https://doi.org/10.7554/eLife.37272.035
-
Source data 1
Expression and fitness data for the final set of 43 TDH3 promoter alleles.
Fluorescence levels and competitive fitness were measured by flow cytometry and analyzed using the R script provided in Supplementary file 3. Data used to make Figure 1D, Figure 2C and Figure 3.
- https://doi.org/10.7554/eLife.37272.036
-
Transparent reporting form
- https://doi.org/10.7554/eLife.37272.037
