Range expansion shifts clonal interference patterns in evolving populations

Abstract

Increasingly, predicting and even controlling evolutionary processes is a sought after goal in fields ranging from agriculture, artificial intelligence, astrobiology, oncology, and infectious diseases. However, our ability to predict evolution and plan such interventions in real populations is limited in part by our understanding of how spatial structure modulates evolutionary dynamics. Among current clinical assays applied to predict drug response in infectious diseases, for instance, many do not explicitly consider spatial structure and its influence on phenotypic heterogeneity, despite it being an inextricable characteristic of real populations. As spatially structured populations are subject to increased interference of beneficial mutants compared to their well-mixed counter-parts, among other effects, this population heterogeneity and structure may non-trivially impact drug response. In spatially-structured populations, the extent of this mutant interference is density dependent and thus varies with relative position within a meta-population in a manner modulated by mutant frequency, selection strength, migration speed, and habitat length, among other factors. In this study, we examine beneficial mutant fixation dynamics along the front of an asexual population expanding its range. We observe that multiple distinct evolutionary regimes of beneficial mutant origin-fixation dynamics are maintained at characteristic length scales along the front of the population expansion. Using an agent-based simulation of range expansion with mutation and selection in one dimension, we measure these length scales across a range of population sizes, selection strengths, and mutation rates. Furthermore, using simple scaling arguments to adapt theory from well-mixed populations, we find that the length scale at the tip of the front within which local mutant fixation occurs in a successive mode decreases with increasing mutation rate, as well as population size in a manner predicted by our derived analytic expression. Finally, we discuss the relevance of our findings to real cellular populations, arguing that this conserved region of successive mutant fixation dynamics at the wave tip can be exploited by emerging evolutionary control strategies.