Table 1. Main notations used throughout the article.

Notation	Description	Formula
$m$	Malthusian fitness
${\left\{m_i\right\}}_{i \in \left[1,K_t\right]}$	Fitness classes within a population
$p_t\left(m_i\right)$	Frequency of the fitness class $m_i$ at time $t$
$N, N_{\mathrm{e}}$	Population size, effective size
$K_t$	Number of fitness classes at time $t$
${\overline{m}}_t$	Mean fitness at time $t$	${\displaystyle \sum _{i=1}^{K_t}{p}_t\left(m_i\right)m_i}$
$V_t$	Variance in fitness at time $t$	${\displaystyle \sum _{i=1}^{K_t}{p}_t\left(m_i\right)m_i^2-{\overline{m}}_t^2}$
${\rho }_t$	Weight of the class $m=0$	$p_t\left(m_i=0\right)$
$\overline{X}$	Mean value of any variable $X\left(m\right),$ averaged over the current distribution of genotypes within a population	${\displaystyle \sum _{i=1}^{K_t}{p}_t\left(m_i\right)X\left(m_i\right)}$
〈〉	“Ensemble expectation” of any random variable, averaged over replicate (finite) populations
$M_t\left(z\right)$	“Empirical” MGF of $m$ in a given population, at time $t$	${\displaystyle \sum _{i=1}^{K_t}{p}_t\left(m_i\right)e^{m_{i}z}}$
$C_t\left(z\right)$	“Empirical” CGF of $m$ in a given population, at time $t$	$\log M_t\left(z\right)$
${\mathrm{ℳ}}_t\left(z\right)$	Expected MGF under the deterministic approximation	${\mathrm{ℳ}}_t\left(z\right)\approx \langle M_t\left(z\right)\rangle$
${\mathcal{C}}_t\left(z\right)$	Expected CGF under the deterministic approximation	${\mathcal{C}}_t\left(z\right)\approx \langle C_t\left(z\right)\rangle$
DFE	Distribution of fitness effects of mutations
$s$	Selection coefficient of a mutation relative to its parent
$f\left(s|m\right)$	Probability distribution function of $s$ in background $m$
$E\left(Y|m\right)$	Expectation of any variable$Y\left(s\right)$ over the DFE in background $m$	${\displaystyle \underset{\mathrm{ℝ}}{\int }Y\left(s\right)f\left(s|m\right)ds},$
${\mu }_s$	Mean effect of mutations on fitness in the background with fitness $m=0$ (or any background in nonepistatic models)	${\displaystyle \underset{\mathrm{ℝ}}{\int }s\hspace{0.17em}f\left(s|m=0\right)ds}$
$M^S\left(z,m\right)$	MGF of the DFE	${\displaystyle \underset{\mathrm{ℝ}}{\int }f\left(s|m\right)e^{s\hspace{0.17em}z}ds}$
$M_{\ast }\left(z\right)$	MGF of the DFE in the background with fitness $m=0$	$M^S\left(z,0\right)$
$\omega \left(z\right)$	Linear effect of $m$ on the CGF of the DFE	${{\partial }_m\log M^S\left(z,m\right)|}_{m=0}$
$s_H$	Harmonic mean in absolute value of the DFE in the background $m=0$	$1/E\left(1/\left|s\right|\right)$
$U$	Genomic mutation rate
$L$	Mutation load (with an optimal fitness class at $m=0$)	$L=-\langle {\overline{m}}_{\mathrm{\infty }}\rangle$
FGM	Fisher’s geometric model
$n$	Dimension of the phenotypic space
$\mathit{\lambda }$	Mutational variance at each trait