Environment-independent distribution of mutational effects emerges from microscopic epistasis

Abstract

Predicting how new mutations alter phenotypes is difficult because mutational effects vary across genotypes and environments. Recently discovered global epistasis, in which the fitness effects of mutations scale with the fitness of the background genotype, can improve predictions, but how the environment modulates this scaling is unknown. We measured the fitness effects of ~100 insertion mutations in 42 strains of Saccharomyces cerevisiae in six laboratory environments and found that the global epistasis scaling is nearly invariant across environments. Instead, the environment tunes one global parameter, the background fitness at which most mutations switch sign. As a consequence, the distribution of mutational effects is predictable across genotypes and environments. Our results suggest that the effective dimensionality of genotype-to-phenotype maps across environments is surprisingly low.

Figure 1.. Proportions of beneficial and deleterious mutations vary with strain growth rate.

Filled and empty triangles show the proportions of beneficial and deleterious mutations in each background strain as a function of its growth rate, respectively. Lines are the best-fit linear regressions; all are statistically significant ($P<0.05$, t-test) except for beneficial mutations in 37°C pH 3.0 and 37°C pH 5.0.

Figure 2:. Distributions of global-epistasis slopes and intercepts and their correlation.

A, B. Histograms of slopes and intercepts estimated from fitting equation (1) to data (see Figure S4 for statistical tests). C. Correlation between slopes and intercepts. Each point represents a mutation, colored by environment. Lines are the best fit linear regressions ($P<0.01$ for all, t-test).

Figure 3.. Global-epistasis slopes of mutations are nearly invariant across environments.

Panels show regression lines from fitting equation (1) for each mutation, colored by the environment as in previous figures. Data points are shown in Figure S6. Mutations are displayed in the order of increasing mean slope. Insets show the results of all pairwise slope-comparison tests (legend in lower right). Histogram in top left shows the overall distribution of fractions of significant tests per mutation (see (45), Section 1.3.7).

Figure 4.. Generalized global epistasis equation captures changes in the DFE across strains and environments.

A. The global epistasis contribution to the fitness effect of each mutation (term $b_m({\lambda }_{ge}-{\bar{\lambda }}_e)$ in equation (2)) as a function of background growth rate term ${\lambda }_{ge}$. Each line represents a mutation. Vertical line indicates the pivot growth rate. Thick colored lines on the $x$-axis indicate the range of measured background-strain growth rates in each environment. B. Estimated DFEs for strains whose adjusted growth rate is negative (top panel), approximately zero (middle panel) and positive (bottom panel). Gray bars show DFEs pooled across all environments, colored lines show DFEs for individual environments (colors are as in previous figures). Insets show the distributions of adjusted growth rates for background strains, with the focal bin shaded. Large square, rhombus and triangle are shown for reference with panels F,G,H. C, D, E. DFE moments plotted against the background strain growth rate. Error bars show ±1 standard errors (see (45), Section 1.3.9). Solid curves show the theoretical predictions calculated from equation (2) and parameterized without the YPD data (see (45), Section 1.4). DFE moments are calculated with a median of 74 mutations (interquartile interval [61,79]). F, G, H. Same data as in C, D, E, but plotted against the adjusted growth rate. Dashed curves are the best fitting polynomial of the corresponding degree (see (45), Section 1.4).