pith. sign in

arxiv: 2606.23946 · v1 · pith:UKADT2Y3new · submitted 2026-06-22 · 🧬 q-bio.PE

Mutant Fixation for a Stochastic Evolutionary Model in Fragmented Populations

Yi Fu , Natalia L. Komarova This is my paper

Pith reviewed 2026-06-26 05:38 UTC · model grok-4.3

classification 🧬 q-bio.PE
keywords mutant fixationfragmented populationsstochastic evolutionMarkov chain reductiontimescale separationmigration networksfixation probabilityabsorption time
0
0 comments X

The pith

Rare migration reduces high-dimensional stochastic evolution in fragmented populations to a lower-dimensional Markov chain on fully mutant and wild-type demes.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper develops a general framework for stochastic evolutionary dynamics in populations divided into demes linked by migration. It establishes that when migration is much slower than local birth-death processes, the full process reduces to a Markov chain whose states are the possible patterns of completely mutant or completely wild-type demes. Fixation probabilities and absorption times in the original model are then asymptotically given by the corresponding quantities on this reduced chain. The result applies to heterogeneous deme sizes, arbitrary deme-dependent birth-death rules, and any strongly connected directed migration network. Explicit formulas are obtained for the single-mutant case.

Core claim

In the limit where migration occurs on a much slower timescale than within-deme dynamics, the full stochastic process can be reduced to a lower-dimensional Markov chain whose states correspond to configurations of fully mutant and fully wild-type demes. The reduction theorem establishes that fixation probabilities and absorption times of the original process are asymptotically determined by the corresponding quantities of the reduced chain. The framework accommodates heterogeneous deme sizes, deme-dependent birth and death processes, and migration on arbitrary strongly connected directed networks with asymmetric rates.

What carries the argument

The reduced Markov chain on configurations of fully mutant and fully wild-type demes, obtained by timescale separation from the original high-dimensional process.

If this is right

  • Fixation probabilities for any initial configuration can be read off from the reduced chain rather than the original high-dimensional process.
  • Absorption times likewise follow from the reduced chain.
  • Explicit closed-form expressions become available for single-mutant fixation on arbitrary migration networks.
  • The same reduction applies to directed asymmetric migration graphs and to demes of unequal sizes.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The method supplies a systematic way to compare fixation outcomes across different network topologies without simulating the full process.
  • It opens the possibility of analyzing how changes in network connectivity or deme-size distribution alter long-term evolutionary outcomes.
  • The reduction may serve as a building block for models that include occasional deme extinction or recolonization events.

Load-bearing premise

Migration occurs on a much slower timescale than within-deme birth and death dynamics.

What would settle it

For a small number of demes, compute fixation probabilities both from direct simulation of the full birth-death-migration process at very small migration rate and from the reduced chain; the two should agree in the limit.

Figures

Figures reproduced from arXiv: 2606.23946 by Natalia L. Komarova, Yi Fu.

Figure 1
Figure 1. Figure 1: Markov chain graphs for wild-type and mutant dynamics with two demes. (a) Markov chain graph for Xε whose infinitesimal generator is given in (2.4) with D = 2, K1 = 4 and K2 = 3. We use blue arrows to represent transitions with O(1) rates, and yellow dotted arrows to represent transitions with O(ε) rates. (b) Markov chain graph for the reduced model XˆR whose infinitesimal generator is given in (3.3). Here… view at source ↗
Figure 2
Figure 2. Figure 2: Markov chain graphs of XˆR for three-deme population with different migration patterns. Three types of migration patterns are considered: (left) ¯µ (ℓ,k) w = 1 for all ℓ ̸= k, (middle) ¯µ (1,3) w = ¯µ (3,2) w = ¯µ (2,1) w = 1, (right) ¯µ (1,2) w = ¯µ (2,1) w = ¯µ (2,3) w = ¯µ (3,2) w = 1. The infinitesimal generator for XˆR is given by (5.5)–(5.6), and the non-zero infinitesimal transitions are shown in th… view at source ↗
Figure 3
Figure 3. Figure 3: Probabilities of mutant fixation for three-deme population with a total capacity 100. Here, K1 + K2 + K3 = 100 and we plot for even integers K1, K2, K3 ≥ 10. For ℓ ∈ {1, 2, 3}, ¯r (ℓ) w = 10, ¯r (ℓ) m = 10τ , ¯d (ℓ) w = 1 and ¯d (ℓ) m = τ . For ℓ ̸= k, ¯µ (ℓ,k) w = τµ¯ (ℓ,k) m . Three types of migration patterns are considered: (left) ¯µ (ℓ,k) w = 1 for all ℓ ̸= k, (middle) µ¯ (1,3) w = ¯µ (3,2) w = ¯µ (2,… view at source ↗
Figure 4
Figure 4. Figure 4: Leading coefficients of expected fixation times (when either the wild-type or the mutant fixates) for three-deme population with a total capacity 100. For parameter values, see [PITH_FULL_IMAGE:figures/full_fig_p017_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Leading coefficients of conditional expected times for mutant fixation for three-deme population with a total capacity 100. For parameter values, see [PITH_FULL_IMAGE:figures/full_fig_p018_5.png] view at source ↗
read the original abstract

Population fragmentation is a common feature of many biological systems. Understanding mutant fixation in such systems is challenging because the underlying stochastic dynamics are high-dimensional. In this work, we develop a general mathematical framework for analyzing stochastic evolution in fragmented populations connected by rare migration. The framework is sufficiently general to accommodate heterogeneous deme sizes, deme-dependent birth and death processes, and migration on arbitrary strongly connected directed networks with asymmetric migration rates. We show that, in the limit where migration occurs on a much slower timescale than within-deme dynamics, the full stochastic process can be reduced to a lower-dimensional Markov chain whose states correspond to configurations of fully mutant and fully wild-type demes. The reduction theorem establishes that fixation probabilities and absorption times of the original process are asymptotically determined by the corresponding quantities of a reduced chain. As an application, we derive explicit formulas for mutant fixation probabilities and fixation times in fragmented populations initiated by the introduction of a single mutant. The results provide a general and tractable approach for studying evolutionary dynamics in complex fragmented populations.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript develops a general framework for stochastic evolutionary dynamics in fragmented populations connected by rare migration on arbitrary strongly connected directed networks, allowing heterogeneous deme sizes and deme-dependent birth-death processes. It establishes a reduction theorem showing that in the slow-migration limit (migration timescale much slower than within-deme absorption), the high-dimensional continuous-time Markov chain reduces to a lower-dimensional chain whose states are the 2^K configurations of fully mutant or fully wild-type demes; fixation probabilities and mean absorption times of the original process converge to those of the reduced chain. Explicit formulas are derived for the single-mutant initiation case.

Significance. If the reduction holds with the stated generality, the work supplies a mathematically rigorous and computationally tractable method for computing fixation quantities in complex metapopulations that would otherwise require simulation of high-dimensional processes. The approach leverages standard singular-perturbation techniques for Markov chains on product spaces and yields explicit results without fitted parameters, which is a clear strength for applications in population genetics.

major comments (2)
  1. [Abstract / main reduction theorem] The reduction theorem is the central claim, but the abstract (and presumably the theorem statement) does not specify the precise mode of convergence or error bounds for absorption times; without these, it is difficult to confirm that the asymptotic equivalence holds uniformly for all initial conditions and network structures.
  2. [Reduction theorem statement and proof] The proof of the reduction relies on the migration graph being strongly connected to ensure the reduced chain is irreducible; this hypothesis must be shown to be necessary and sufficient for the occupation measure to concentrate on mono-type states, with a concrete counter-example if the graph is only weakly connected.
minor comments (2)
  1. [Application section] Clarify whether the explicit formulas for single-mutant fixation require solving a linear system on the reduced chain or admit a fully closed form independent of network size.
  2. [Model definition] Notation for the migration rates and birth-death parameters should be introduced with a single consistent table or list to avoid ambiguity when demes are heterogeneous.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their careful reading, positive assessment of the work, and constructive suggestions. We address each major comment below and will incorporate clarifications via minor revisions.

read point-by-point responses

  1. Referee: [Abstract / main reduction theorem] The reduction theorem is the central claim, but the abstract (and presumably the theorem statement) does not specify the precise mode of convergence or error bounds for absorption times; without these, it is difficult to confirm that the asymptotic equivalence holds uniformly for all initial conditions and network structures.

    Authors: The reduction theorem establishes that, as the migration rate ε → 0, the fixation probabilities converge to those of the reduced chain and the scaled mean absorption times (multiplied by ε) converge to the corresponding reduced-chain quantities. This follows from standard singular-perturbation analysis on the product-space Markov chain and holds uniformly over initial conditions that share the same mono-type deme configuration. To address the request for explicitness, we will revise the abstract and Theorem 3.1 to state the mode of convergence (convergence of probabilities and of scaled expectations) and add a remark on the O(ε) error bounds that are uniform under the maintained strong-connectivity hypothesis. revision: yes

  2. Referee: [Reduction theorem statement and proof] The proof of the reduction relies on the migration graph being strongly connected to ensure the reduced chain is irreducible; this hypothesis must be shown to be necessary and sufficient for the occupation measure to concentrate on mono-type states, with a concrete counter-example if the graph is only weakly connected.

    Authors: Strong connectivity is used to guarantee that the reduced chain on the 2^K mono-type configurations is irreducible on its transient states, which is required for the occupation measure to concentrate on the two global absorbing states. The assumption is both necessary and sufficient: without it the network decomposes into strongly connected components whose fixation events are independent. We will add a short discussion of necessity together with the following concrete counter-example: two demes with a single directed migration A → B but no return path. The reduced chain then admits additional absorbing states in which A is mutant and B is wild-type (or vice versa), so the occupation measure fails to concentrate on global mono-type configurations. revision: yes

Circularity Check

0 steps flagged

No significant circularity

full rationale

The paper derives a reduction theorem for a continuous-time Markov chain on a product space of demes by invoking a standard singular-perturbation argument under an explicit separation-of-timescales hypothesis (migration rate ε→0 while intra-deme absorption rates remain O(1)). The resulting convergence of fixation probabilities and mean absorption times to those of the embedded jump chain on mono-type configurations follows directly from the model assumptions, the strong connectivity of the migration graph, and finite deme sizes; no parameter is fitted to data and then re-used as a prediction, no self-citation supplies a load-bearing uniqueness theorem, and the central claim is not equivalent to its inputs by definition. The derivation is therefore self-contained against external mathematical benchmarks for averaging principles on Markov processes.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the slow-migration limit and standard Markov chain theory; no free parameters or invented entities are introduced in the abstract.

axioms (1)
  • domain assumption Migration occurs on a much slower timescale than within-deme dynamics
    This limit is invoked to justify the reduction to the lower-dimensional chain on fully mutant/wild-type demes.

pith-pipeline@v0.9.1-grok · 5704 in / 1218 out tokens · 31974 ms · 2026-06-26T05:38:22.977224+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

25 extracted references

  1. [1]

    and Fu, Yi and Del Vecchio, Domitilla and Williams,Ruth J

    Bruno, Simone and Campos, Felipe A. and Fu, Yi and Del Vecchio, Domitilla and Williams,Ruth J. , title=. SIAM Journal on Applied Dynamical Systems , year=

  2. [2]

    and Filar, Jerzy A

    Avrachenkov, Konstantin E. and Filar, Jerzy A. and Howlett, Phil G. , year =. Analytic Perturbation Theory and Its Applications , ISBN =

  3. [3]

    and Helton, J

    Hughes, Jared A. and Helton, J. William and Schlosser, Peter , keywords =. The discrete Laplace asymptotic method and its application to the 3XOR satisfiability problem , publisher =. 2025 , copyright =

    2025

  4. [4]

    1997 , publisher=

    Principles of population genetics , author=. 1997 , publisher=

    1997

  5. [5]

    2004 , publisher=

    Mathematical population genetics: theoretical introduction , author=. 2004 , publisher=

    2004

  6. [6]

    and Wodarz, Dominik , title=

    Komarova, Natalia L. and Wodarz, Dominik , title=. Nature Communications , year=

  7. [7]

    , title=

    Lieberman, Erez and Hauert, Christoph and Nowak, Martin A. , title=. Nature , year=

  8. [8]

    Reviews of nonlinear dynamics and complexity , volume=

    Stochastic evolutionary game dynamics , author=. Reviews of nonlinear dynamics and complexity , volume=. 2009 , publisher=

    2009

  9. [9]

    Nature , volume=

    Evolutionary dynamics on any population structure , author=. Nature , volume=. 2017 , publisher=

    2017

  10. [10]

    Genetics Research , volume=

    On the fixation probability of mutant genes in a subdivided population , author=. Genetics Research , volume=. 1970 , publisher=

    1970

  11. [11]

    Evolution , pages=

    Fixation probabilities and fixation times in a subdivided population , author=. Evolution , pages=. 1981 , publisher=

    1981

  12. [12]

    Journal of The Royal Society Interface , volume=

    Counterintuitive properties of the fixation time in network-structured populations , author=. Journal of The Royal Society Interface , volume=. 2014 , publisher=

    2014

  13. [13]

    Scientific Reports , volume=

    Fixation probabilities in network structured meta-populations , author=. Scientific Reports , volume=. 2021 , publisher=

    2021

  14. [14]

    and Quince, Christopher , title=

    Parsons, Todd L. and Quince, Christopher , title=. Theoretical Population Biology , year =

  15. [15]

    and Quince, Christopher , title=

    Parsons, Todd L. and Quince, Christopher , title=. Theoretical Population Biology 72 , year=

  16. [16]

    and Quince, Christopher and Plotkin, Joshua B

    Parsons, Todd L. and Quince, Christopher and Plotkin, Joshua B. , title=. Genetics , year=

  17. [17]

    Toward a universal model for spatially structured populations , journal=

    Marrec, Lo\". Toward a universal model for spatially structured populations , journal=. 2021 , volume=

    2021

  18. [18]

    Ecology and Evolution , doi=

    Marrec, Loïc and Bank, Claudia and Bertrand, Thibault , title=. Ecology and Evolution , doi=. 2023 , volume=

    2023

  19. [19]

    and Kurtz, Thomas G

    Anderson, David F. and Kurtz, Thomas G. , TITLE =. 2015 , PAGES =

    2015

  20. [20]

    2006 , PAGES =

    Wilkinson, Darren James , TITLE =. 2006 , PAGES =

    2006

  21. [21]

    Mathematical proceedings of the Cambridge philosophical society , volume=

    A mathematical theory of natural and artificial selection, part V: selection and mutation , author=. Mathematical proceedings of the Cambridge philosophical society , volume=. 1927 , organization=

    1927

  22. [22]

    Mathematical proceedings of the cambridge philosophical society , volume=

    Random processes in genetics , author=. Mathematical proceedings of the cambridge philosophical society , volume=. 1958 , organization=

    1958

  23. [23]

    Genetics , volume=

    On the probability of fixation of mutant genes in a population , author=. Genetics , volume=

  24. [24]

    Genetics , volume=

    The stepping stone model of population structure and the decrease of genetic correlation with distance , author=. Genetics , volume=

  25. [25]

    2008 , publisher=

    Probability models for DNA sequence evolution , author=. 2008 , publisher=

    2008