Regular neutral networks outperform robust ones by reaching their top growth rate more quickly

TL;DR

This study reveals that regular neutral networks, despite lower growth rates, reach their maximum growth faster and lead to larger populations, outperforming robust networks in evolutionary dynamics.

Contribution

It introduces the concepts of regularity and robustness of neutral networks and demonstrates that regularity, not growth rate, determines population size in evolution.

Findings

Regular networks reach their top growth rate more quickly.

More regular ribozymes achieve larger populations.

Robustness and regularity are uncorrelated properties.

Abstract

We study the relative importance of "top-speed" (long-term growth rate) and "acceleration" (how quickly the long-term growth rate can be reached) in the evolutionary race to increase population size. We observe that fitness alone does not capture growth rate: robustness, a property of neutral network shape, combines with fitness to include the effect of deleterious mutations, giving growth rate. Similarly, we show that growth rate alone does not capture population size: regularity, a different property of neutral network shape, combines with growth rate to include the effect of higher depletion rates early on, giving size. Whereas robustness is a function of the principal eigenvalue of the neutral network adjacency matrix, regularity is a function of the principal eigenvector. We show that robustness is not correlated with regularity, and observe in silico the selection for regularity by evolving RNA ribozymes. Despite having smaller growth rates, the more regular ribozymes have the biggest populations.