Natural selection of mutants that modify population structure
summary
The gist
Evolution occurs in populations of reproducing individuals It is well known that population structure can affect evolutionary dynamics Here we study how natural selection acts on mutants that have…
In short
The study investigates how natural selection acts on mutants with identical reproductive rates but different population structures, modeled by dispersal graphs. Competing mutants disperse offspring on different graphs, where a more connected graph implies higher motility. The results show that enhanced structured motility generally increases an invader's fixation probability, and for low-dimensional lattices, this effect is comparable to an increased reproductive rate.
Key concepts
- Population Structure Graph
- This is a mathematical representation of how individuals in a population disperse or move. It's shown as a graph where edges represent the dispersal paths of offspring. Mutation can change this structure, affecting how the population evolves.
- Motility
- Motility refers to how easily individuals can move or disperse within the population structure defined by the graph. A more connected graph means higher motility, allowing offspring to spread further and potentially increase their chance of survival in a new environment.
- Fixation Probability
- This is the likelihood that a single invading mutant type will eventually take over an entire population, replacing all the resident types. The study examines how changing the population structure (motility) influences this probability for different mutants.
Terminology used across episodes
This episode discusses
The paper
Natural selection of mutants that modify population structure · Read on arXiv
Department of Mathematics, Harvard University · Department of Applied Mathematics, University of Washington · Institute of Science and Technology Austria
Evolution occurs in populations of reproducing individuals. It is well known that population structure can affect evolutionary dynamics. Traditionally, natural selection is studied between mutants that differ in reproductive rate, but are subject to the same population structure. Here we study how natural selection acts on mutants that have the same reproductive rate, but experience different population structures. In our framework, mutation alters population structure, which is given by a graph that specifies the dispersal of offspring. Reproduction can be either genetic or cultural. Competing mutants disperse their offspring on different graphs. A more connected graph implies higher motility. We show that enhanced motility tends to increase an invader's fixation probability, but there are interesting exceptions. For island models, we show that the magnitude of the effect depends crucially on the exact layout of the additional links. Finally, we show that for low-dimensional lattices, the effect of altered motility is comparable to that of altered fitness: in the limit of large population size, the invader's fixation probability is either constant or exponentially small, depending on whether it is more or less motile than the resident.
Transcript
Introduction to the show: ident: Genomics Radio. Generated commentary on the latest computational biology and genomics papers.
Ines: I'm Ines.
Marcus: And I'm Marcus, here with Yuki.
Ines: We're looking at the paper: "Natural selection of mutants that modify population structure".
Marcus: Here's the gist of the paper. Evolution occurs in populations of reproducing individuals It is well known that population structure can affect evolutionary dynamics Here we study how natural selection acts on mutants that have the same reproductive rate, but experience different population structures.
Ines: Modeling Population Structure The framework models population structure as a graph specifying the dispersal of offspring, where mutation alters this structure.
Marcus: Reproduction can be genetic or cultural, and competing mutants disperse their offspring on different graphs.
Ines: The population structure is represented by two overlaid graphs GA and GB, where mutant offspring migrate along the edges of GA and resident offspring migrate along the edges of GB.
Paper summary: Ines: So we're talking about this paper called Natural selection of mutants that modify population structure. The main idea here is that natural selection doesn't just pick the fastest breeders; it can select for how well a mutant navigates the environment, even if its reproductive rate stays exactly the same.
Marcus: Right, so traditionally we look at changes in reproductive rates and how those changes interact with existing population structures. This paper flips that around, looking at mutants that have identical reproductive rates but experience different dispersal patterns defined by these graphs. The core claim is about how this altered structure affects the invasion success of the mutant type.
Yuki: From a population genetics angle, this sounds interesting because it suggests that spatial structure isn't just a passive backdrop; it actively shapes evolutionary outcomes when motility changes. We're looking at how mutation itself can change that underlying map of dispersal, which is modeled by these graphs.
Ines: Exactly. The framework sets up two overlaid graphs, GA and GB, where the mutant offspring use one structure to disperse and the residents use another structure to disperse, and the mutation changes one of those structures. It's a way to model how reproduction can be genetic or cultural, but here we focus on motility potential as something independent of proliferation.
Marcus: The paper sets up this model where the population structure is given by a graph specifying offspring dispersal, and mutation alters this graph. We see competing mutants dispersing their offspring on different graphs, and a more connected graph means higher motility. This leads to the central question about how enhanced structured motility affects an invader's fixation probability.
Yuki: I wonder what this means for evolution in real species, especially when we think about how different regions or habitats can create distinct selective pressures based on dispersal capability. The paper suggests that the way you move between patches can be as important as the fitness you have locally.
Ines: Well, they show that enhanced structured motility generally leads to an increase in the fixation probability of the invading mutant. This is counterintuitive because random motility usually works against invasion, but here, with structured movement—like moving through a specific lattice—it can actually help the invader spread.
Paper summary: Marcus: They get more specific about this effect when looking at low-dimensional lattices. For these graphs, they find that the effect of altered motility is comparable to the effect of altering the reproductive rate in a well-mixed population, especially in the limit of a large population size N.
Yuki: That comparison to reproductive rate changes is key because it connects this structural effect back to something we already study, but now we're seeing how structure modifies that relationship. It suggests that for these simple spatial arrangements, the mechanism driving fixation probability is similar whether you change the breeding speed or the movement potential.
Ines: They look at specific graph structures and find some interesting things. For small graphs, they mention that for N equals two a single type A individual has a fixation probability of one half <ref:2111.10890#pg1>. And for N equals six they found that the Complete graph K6 is locally optimal for fixation probability.
Marcus: The paper also looks at dense regular graphs and finds something unexpected about the Isothermal Theorem. They show that two regular graphs with the same total number of edges can have asymptotically different fixation probabilities as N gets bigger. Specifically, for a Complete Bipartite graph BN, both rho(BN) and rho?(BN) are within a constant factor of one over N for large N.
Yuki: That violation of the Isothermal Theorem is significant because it suggests that simple structural similarities between population graphs don't guarantee similar evolutionary dynamics, which really supports the idea that structure matters deeply here. How does this translate to real-world populations?
Ines: It translates to saying you can't just look at the average connectivity or edge density and assume everything is fair in terms of invasion success; the specific layout of those links matters a lot. And for one-dimensional lattices, they find that fixation probability decays exponentially quickly when the mutant graph has lower connectivity <ref:2111.10890#pg1>.
Marcus: They also make a comparison to something else here: this decay behavior in low-dimensional lattices is reminiscent of how the fixation probability looks for a single mutant with a relative reproductive rate r in a well-mixed population. This suggests that for neutral mutants, living on a more densely connected lattice gives them an effective fitness comparable to having some relative reproductive advantage.
Yuki: So, if you look at the history of species and how they occupy space, this paper hints that the connectivity of the habitat network might be a hidden selective factor influencing who survives and spreads across those regions. It's about how movement potential dictates success in a structured environment.
Paper summary: Ines: They also explored game theory aspects by asking what the best mutant response is to a resident graph. For example, if the residents are on a Star graph S6, the paper finds that invading through a graph M6 that misses one edge is better than invading through the Complete graph K6.
Marcus: And they conclude something about distinct graphs GA and GB with the same edge density: the fixation probability rho(GA, GB) typically drops below one over N. This means even if the structure looks similar in terms of total edges, it can still be bad for invasion success.
Yuki: I think this pushes us to consider that when modeling species spread, we can't just use a single metric for connectivity; we need to account for the specific geometry of dispersal pathways. The sample size constraints might limit how much this fine-grained structural detail can actually support in real ecological data.
Ines: That’s a fair point about the data supporting it. They do flag that there are open problems, specifically whether there exist two regular graphs where both rho(GN, G0N) is greater than one over N and rho(G0N, GN) is greater than one over N.
Marcus: And they also noted that for one-D lattices, the fixation probability rho(two one) is bounded away from zero as N goes to infinity <ref:2111.10890#pg1>. This means there's a certain level of success possible even in simple structures over long time scales.
Yuki: So what this changes for those of us looking at population genetics? It suggests that when we analyze structured populations, we should be paying very close attention to the specific connectivity topology rather than just broad metrics like average degree.
Ines: Exactly. The single most important point from this paper is that enhanced structured motility generally leads to an increase in the fixation probability of the invading mutant. It shows how movement potential, not just reproductive rate, can be a powerful driver of evolution when space is structured in a specific way.
Marcus: So for anyone looking at modeling evolutionary dynamics in spatially structured populations, this paper gives you tools to test how dispersal graphs change the expected outcome of natural selection on mutants that aren't changing their own breeding speed.
Conclusion: Ines: So we're wrapping up this look at "Natural selection of mutants that modify population structure." This paper essentially shows how selective pressure can act not just on how fast a mutant reproduces, but on how well it moves through its environment.
Marcus: Yeah, the authors are looking at these mutants who have the same reproductive rate but experience different dispersal patterns defined by graphs. It's about that structural difference being what drives the evolution here.
Yuki: From a population genetics view, this suggests that spatial structure isn't just a passive backdrop; it actively shapes evolutionary outcomes when motility changes, which is interesting for how we model species spread across landscapes.
Ines: Exactly. The main takeaway is that enhanced structured motility generally leads to an increase in the fixation probability of the invading mutant. It’s counterintuitive because random movement usually works against invasion, but here it actually helps the invader spread more effectively when things are structured.
Marcus: They get specific about this effect when looking at low-dimensional lattices, finding that altered motility has a comparable effect to altering the reproductive rate in a well-mixed population for large populations. That’s a solid comparison to make because it links movement potential directly back to fitness advantage in a way we already study.
Yuki: That connection is important because it suggests that for these simple spatial arrangements, the mechanism driving fixation probability is similar whether you change the breeding speed or the movement potential. It makes sense if you think about how different habitats create distinct selective pressures based on dispersal capability.
Ines: They also looked at specific graph structures, finding that for small graphs, the Complete graph K6 is locally optimal for fixation probability when N equals six. And they found that for dense regular graphs, two graphs with the same edge density can actually have different asymptotic fixation probabilities as N grows.
Marcus: That violation of the Isothermal Theorem is significant because it suggests that simple structural similarities between population graphs don't guarantee similar evolutionary dynamics, which really supports the idea that structure matters deeply here. The specific layout of those links isn't just noise; it’s a huge factor in invasion success.
Yuki: So what this means for the wider history of species is that you can't just look at the average connectivity or edge density and assume everything is fair in terms of invasion success; the specific geometry of dispersal pathways matters a lot for how a population establishes itself.
Ines: The paper also addressed game theory aspects, asking what the best mutant response is to a resident graph, finding that invading through one specific type of graph structure is better than another. It shows how structure influences the competitive landscape between different types of invaders.
Marcus: And they concluded that for distinct graphs with the same edge density, the fixation probability typically drops below one over N. This means even if the structure looks similar in terms of total edges, it can still be bad for invasion success depending on exactly how those edges are connected.
Yuki: I think this pushes us to consider that when modeling species spread, we can't just use a single metric for connectivity; we need to account for the specific geometry of dispersal pathways when we analyze structured populations.
Ines: The single most important point from this paper is that enhanced structured motility generally leads to an increase in the fixation probability of the invading mutant. It shows how movement potential, not just reproductive rate, can be a powerful driver of evolution when space is structured in a specific way.
Marcus: So for anyone looking at modeling evolutionary dynamics in spatially structured populations, this paper gives you tools to test how dispersal graphs change the expected outcome of natural selection on mutants that aren't changing their own breeding speed.
More episodes
- 2610.33881-Sensorimotor behavior from an untrained brain-and-nerve-cord connectome
- 2610.40344-Morphological Features of Murine Pulmonary Arterial Networks for Diagnosing Vascular Disease
- Daily Summary for 2026-11-19
- 2609.10193-Are You Learning Biological Signal or Shortcuts? Auditing and Mitigating Bias in Protein-Protein Interaction Datasets
- 2610.38236-Generative discovery of chemical reaction networks
- 2507.11176-An Interpretable AI framework Quantifying Traditional Chinese Medicine Principles Towards Enhancing and Integrating with Modern Biomedicine
- The day in conversation for 2026-11-19
- Daily Summary for 2026-11-19
- 2504.03590-Towards a Unified Framework for Determining Conformational Ensembles of Disordered Proteins
- 2603.15217-A multiscale discrete-to-continuum framework for structured population models