TY - JOUR T1 - Natural selection can favor the evolution of ratchet robustness over evolution of mutational robustness JF - bioRxiv DO - 10.1101/121087 SP - 121087 AU - Yinghong Lan AU - Aaron Trout AU - Daniel M Weinreich AU - C Scott Wylie Y1 - 2017/01/01 UR - http://biorxiv.org/content/early/2017/03/27/121087.abstract N2 - The vast majority of mutations are deleterious. How natural populations evolve to cope is a question of fundamental interest. Previous studies have reported the evolution of mutational robustness, that is, natural selection favoring mutations that reduce effects of deleterious mutations elsewhere in the genome. Here, we demonstrate that finite, asexual populations’ ability to purge recurrent deleterious mutations declines with increased mutational robustness. Consequently, higher mutational robustness leads to higher risk of extinction by Muller’s ratchet. We therefore hypothesize that in the long run, natural populations may instead evolve robustness against Muller’s ratchet by increasing sensitivity to deleterious mutations, despite the short-term fitness consequences. We call this phenomenon “ratchet robustness”. Using individual-based simulations, we first confirm that ratchet robustness is inversely correlated with mutational robustness on fitness landscapes without epistasis. Next, we demonstrate that negative epistasis increases ratchet robustness, precisely because on fitness landscapes with negative epistasis sensitivity to deleterious mutations increases as mutations accumulate. We also show that on a fitness landscape with adjacent fitness peaks exhibiting exclusively positive and negative epistasis, the populations will converge on the latter. On the other hand, introducing even a vanishingly small region of positive epistasis on a single-peaked fitness landscape that otherwise exhibits negative epistasis is enough to entirely ablate ratchet robustness, exposing the population to Muller’s ratchet. We conclude that while regions of the fitness landscape with higher mutational robustness could be favored in the short term, purely because of temporary fitness advantage, in the long run mutational robustness may render a population vulnerable to extinction. Some empirical support exists for this prediction. ER -