pith. sign in

arxiv: 2607.00628 · v1 · pith:V4MTZMPInew · submitted 2026-07-01 · 🧬 q-bio.PE · q-bio.QM

Effective population sizes for asymmetrically regulated birth-death processes

Yunbei Pan , Tom Chou This is my paper

Pith reviewed 2026-07-02 01:55 UTC · model grok-4.3

classification 🧬 q-bio.PE q-bio.QM
keywords birth-death processeseffective population sizefixation probabilityMoran processstochastic dynamicspopulation regulationneutral evolution
0
0 comments X

The pith

Regulation split between birth and death creates stochastic bias favoring birth-regulated species even when mean growth rates match.

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

In multispecies birth-death processes the split of regulation between birth suppression and death elevation is invisible to deterministic mean-field equations yet shapes the diffusion of frequencies. Projecting the process onto a lower-dimensional Moran model via a tunable partitioning parameter α yields a regulation-dependent diffusion tensor. For exactly neutral two-species cases this produces fixation probabilities and times that systematically favor the more birth-regulated species. The same construction defines an α-dependent effective population size among neutral species and, via perturbation, a diversity-loss timescale for near-neutral populations.

Core claim

A d-species birth-death process projected onto a (d-1)-dimensional Moran process with regulation-partitioning parameter α_i acquires a diffusion tensor that depends on the mechanism of regulation; for neutral two-species populations this tensor produces fixation probabilities favoring the birth-regulated species and permits definition of an α-dependent effective population size N_e(α).

What carries the argument

The regulation partitioning parameter α that apportions total regulation between birth-rate suppression and death-rate elevation inside the projected Moran process.

If this is right

  • Fixation probability depends on α even when deterministic growth rates are identical.
  • Effective population size becomes a function N_e(α) of the regulation mechanism.
  • Spectral gap of near-neutral systems yields a diversity-loss timescale interpretable as an effective population size.
  • The framework applies directly to T-cell clone dynamics regulated by proliferation and apoptosis.

Where Pith is reading between the lines

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

  • The bias implies that ecological models assuming symmetric regulation may misestimate coexistence times when regulation mechanisms differ across species.
  • Systems allowing independent control of birth and death rates could directly test the predicted α-dependent fixation asymmetry.
  • Extending the projection to spatial or metapopulation settings could link local regulation mechanism to global diversity maintenance.

Load-bearing premise

The projection of the full birth-death process onto the Moran process with tunable α preserves the stochastic features required to compute fixation probabilities and effective sizes.

What would settle it

In a controlled two-species experiment with identical mean growth rates but different regulation splits, measure whether the birth-regulated species reaches fixation more often than 50 percent.

Figures

Figures reproduced from arXiv: 2607.00628 by Tom Chou, Yunbei Pan.

Figure 1
Figure 1. Figure 1: (a) shows that the species dominated by death regulation (αi < 1/2) experiences stronger fluctuations and is more likely to suffer extinction (lower fixation probability q(p0) < p0). Despite having no favorable selection, a birth regulation-dominated population (αi > 1/2) has a weaker demographic noise amplitude, giving rise to a fixation bias q(p0) > p0 for all p0 ∈ (0, 1), in￾dependent of K. Due to the n… view at source ↗
read the original abstract

In multispecies birth-death processes, how population regulation -- through suppressed replication, elevated mortality, or both -- affects macroscopic stochastic dynamics has escaped detailed analysis. Here, we show that the distribution of regulation mechanisms can be invisible in deterministic or mean-field dynamics but play a significant role in the diffusive evolution of population frequencies. By introducing a tunable regulation partitioning parameter $\alpha_i$ and projecting a $d$-species birth-death process onto a $(d{-}1)$-dimensional Moran process, we find a regulation-mechanism-dependent diffusion tensor. For the simple two-species case, we derive exact fixation times and probabilities to show how different regulation mechanisms stochastically favors a more birth-regulated species, even under complete deterministic neutrality. Our model also allows us to define an $\alpha$-dependent effective population size $N_{\rm e}(\alpha)$ among neutral species, generalizing its classical interpretation. For near-neutral populations or populations that are heterogeneous in their regulation mechanism, we used perturbation theory to calculate the spectral gap, identifying it with a diversity loss timescale which can also be interpreted as setting an effective population size. Our results are particularly applicable to interacting subpopulations of T cells ("clones") which are near-neutral, are regulated through proliferation and apoptosis, and lose diversity with time.

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

3 major / 2 minor

Summary. The paper claims that in multispecies birth-death processes, the partitioning of population regulation between birth suppression and death elevation (via a tunable parameter α_i) is invisible in deterministic dynamics but produces a regulation-dependent diffusion tensor when the d-species process is projected onto a (d-1)-dimensional Moran process. For the two-species case this yields exact fixation probabilities and times that favor the more birth-regulated species even under deterministic neutrality; the framework also defines an α-dependent effective population size N_e(α) and uses perturbation theory on the spectral gap to obtain a diversity-loss timescale for near-neutral or heterogeneous populations, with application to T-cell clone dynamics.

Significance. If the projection step is rigorously justified, the work supplies exact two-species results and a perturbation treatment of the spectral gap, both of which are concrete strengths. The α-dependent generalization of effective population size offers a new interpretive tool for neutral systems whose regulation mechanisms differ, and the link to T-cell diversity loss supplies a timely biological application. These elements would be of interest to stochastic population genetics and mathematical immunology.

major comments (3)
  1. [projection step (following introduction of α_i)] The projection of the continuous-time multi-type birth-death process onto the (d-1)-dimensional Moran process with α-partitioned regulation is load-bearing for every central claim (fixation bias, N_e(α), and the spectral-gap perturbation). The manuscript must demonstrate that this reduction exactly preserves fixation probabilities and times (or quantify the error) rather than merely matching the diffusion limit to first order; otherwise the reported stochastic favoritism could be an artifact of omitted jump correlations or total-population fluctuations.
  2. [two-species fixation derivation] For the two-species exact results, the derivation of the regulation-dependent diffusion tensor must be shown to be free of higher-order terms that survive in the original birth-death process; if the mapping is only first-order accurate, the claimed bias under deterministic neutrality requires an explicit error bound.
  3. [perturbation theory section] The perturbation theory for the spectral gap (used to identify the diversity-loss timescale and an effective population size) assumes the unperturbed neutral process is well approximated by the projected Moran dynamics; the validity of this assumption for heterogeneous α_i distributions should be checked against the full multi-type generator.
minor comments (2)
  1. Notation for the regulation partitioning parameter α_i should be introduced with an explicit statement of its range and normalization before the projection is defined.
  2. The biological motivation for T-cell clones would benefit from a short paragraph clarifying how proliferation versus apoptosis rates map onto the α parameter in the model.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the detailed and constructive report. The comments correctly identify that the projection step is central to our claims, and we address each point below with plans for revision.

read point-by-point responses

  1. Referee: The projection of the continuous-time multi-type birth-death process onto the (d-1)-dimensional Moran process with α-partitioned regulation is load-bearing for every central claim. The manuscript must demonstrate that this reduction exactly preserves fixation probabilities and times (or quantify the error) rather than merely matching the diffusion limit to first order.

    Authors: We agree that an explicit justification beyond first-order diffusion matching is required. In the revision we will add a new subsection deriving that the projected process exactly matches the infinitesimal generator of the frequency process up to o(1/N) terms for fixed total population size N, and we will include Monte Carlo comparisons of fixation probabilities between the original multi-type birth-death process and the projected Moran process for N up to several hundred, thereby providing quantitative error bounds. revision: yes

  2. Referee: For the two-species exact results, the derivation of the regulation-dependent diffusion tensor must be shown to be free of higher-order terms that survive in the original birth-death process; if the mapping is only first-order accurate, the claimed bias under deterministic neutrality requires an explicit error bound.

    Authors: The two-species fixation probabilities and times are obtained from the exact solution of the one-dimensional diffusion with the derived drift and diffusion coefficients. We will augment the derivation with a remainder estimate showing that the neglected jump correlations contribute O(1/N^2) corrections to the fixation probability under the neutral scaling; this bound will be stated explicitly and verified numerically for representative α values. revision: yes

  3. Referee: The perturbation theory for the spectral gap assumes the unperturbed neutral process is well approximated by the projected Moran dynamics; the validity of this assumption for heterogeneous α_i distributions should be checked against the full multi-type generator.

    Authors: We will add a short numerical validation subsection comparing the spectral gap obtained from the projected perturbation formula against direct diagonalization of the full multi-type generator for small d and moderate N with heterogeneous α_i. Where discrepancies appear we will report them and discuss the regime of validity; this constitutes a partial revision because exhaustive analytic comparison for arbitrary d remains intractable. revision: partial

Circularity Check

0 steps flagged

No significant circularity; derivations follow from projection and exact solutions

full rationale

The paper introduces α_i as a modeling parameter to partition regulation mechanisms and projects the multi-species BD process onto a reduced Moran process. From the resulting diffusion tensor it derives exact fixation probabilities/times (two-species case) and defines Ne(α) via the standard relation to fixation time or variance effective size. These quantities are obtained by direct solution of the projected master equation or diffusion approximation rather than by fitting to data or by self-referential definition. No self-citation chain, ansatz smuggling, or renaming of known results is used to establish the central claims. The projection step is an explicit modeling reduction whose validity can be checked against the original process; the subsequent results are not forced by construction to equal the inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract alone supplies insufficient detail to enumerate specific free parameters, axioms, or invented entities; the tunable α_i appears introduced as a modeling choice but its status cannot be audited without the full text.

pith-pipeline@v0.9.1-grok · 5751 in / 1143 out tokens · 33405 ms · 2026-07-02T01:55:09.468203+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

29 extracted references · 29 canonical work pages

  1. [1]

    In di- verse ecosystems ( d ≥ 3), the multidimensional structure of the noise splits the macroscopic relaxation timescales

    Once species-specific regulation mechanisms are introduced, the isotropy of the diffusion is broken. In di- verse ecosystems ( d ≥ 3), the multidimensional structure of the noise splits the macroscopic relaxation timescales. The loss of diversity is therefore governed by a spectrum of decay rates that are determined by the heterogeneous regulation mechanism...

    work page

  2. [2]

    R. A. Fisher, Proceedings of the royal society of Edin- burgh 42, 321 (1923)

    work page 1923

  3. [3]

    T. D. Tran, J. Hofrichter, and J. Jost, Theory in Bio- sciences 132, 73 (2013)

    work page 2013

  4. [4]

    R. A. Blythe and A. J. McKane, Journal of Statisti- cal Mechanics: Theory and Experiment 2007, P07018 (2007)

    work page 2007

  5. [5]

    W. J. Ewens and W. J. Ewens, Mathematical popula- tion genetics: theoretical introduction , Vol. 27 (Springer, 2004)

    work page 2004

  6. [6]

    P. A. P. Moran, in Mathematical Proceedings of the Cam- bridge Philosophical Society , Vol. 54 (Cambridge Univer- sity Press, 1958) pp. 60–71

    work page 1958

  7. [7]

    Lieberman, C

    E. Lieberman, C. Hauert, and M. A. Nowak, Nature 433, 312 (2005)

    work page 2005

  8. [8]

    Zukewich, V

    J. Zukewich, V. Kurella, M. Doebeli, and C. Hauert, PLoS One 8, e54639 (2013)

    work page 2013

  9. [9]

    Kaveh, N

    K. Kaveh, N. L. Komarova, and M. Kohandel, Royal Society open science 2, 140465 (2015)

    work page 2015

  10. [10]

    Kaveh, A

    K. Kaveh, A. McAvoy, and M. A. Nowak, Royal Society open science 6, 181661 (2019)

    work page 2019

  11. [11]

    Svoboda, S

    J. Svoboda, S. Joshi, J. Tkadlec, and K. Chatterjee, PLoS Computational Biology 20, e1012008 (2024)

    work page 2024

  12. [12]

    D. A. Brewster, Y. Huang, M. Mitzenmacher, and M. A. Nowak, arXiv preprint arXiv:2511.18252 (2025)

    work page arXiv 2025

  13. [13]

    Kimura, Journal of Applied Probability 1, 177 (1964)

    M. Kimura, Journal of Applied Probability 1, 177 (1964)

    work page 1964

  14. [14]

    Traulsen, J

    A. Traulsen, J. C. Claussen, and C. Hauert, Physical Review Letters 95, 238701 (2005)

    work page 2005

  15. [15]

    F. A. C. C. Chalub and M. O. Souza, Journal of Mathe- matical Biology 68, 1089 (2014)

    work page 2014

  16. [16]

    A. J. Black and A. J. McKane, Trends in Ecology & Evolution 27, 337 (2012)

    work page 2012

  17. [17]

    K. S. Korolev, M. Avlund, O. Hallatschek, and D. R. Nelson, Reviews of Modern Physics 82, 1691 (2010)

    work page 2010

  18. [18]

    Chotibut and D

    T. Chotibut and D. R. Nelson, Journal of Statistical Physics 167, 777 (2017)

    work page 2017

  19. [19]

    C. D. Surh and J. Sprent, Immunity 29, 848 (2008)

    work page 2008

  20. [20]

    Min, Frontiers in immunology 9, 547 (2018)

    B. Min, Frontiers in immunology 9, 547 (2018)

    work page 2018

  21. [21]

    Hataye, J

    J. Hataye, J. J. Moon, A. Khoruts, C. Reilly, and M. K. Jenkins, Science 312, 114 (2006)

    work page 2006

  22. [22]

    K. S. Schluns and L. Lefran¸ cois, Nature Reviews Im- munology 3, 269 (2003)

    work page 2003

  23. [23]

    S. I. Mannering, J. Zhong, and C. Cheers, Immunology 106, 87 (2002)

    work page 2002

  24. [24]

    Mayer, Y

    A. Mayer, Y. Zhang, A. S. Perelson, and N. S. Wingreen, Proceedings of the National Academy of Sciences 116, 5914 (2019)

    work page 2019

  25. [25]

    Chotibut and D

    T. Chotibut and D. R. Nelson, Physical Review E 92, 022718 (2015)

    work page 2015

  26. [26]

    G. W. A. Constable and A. J. McKane, Physical Review Letters 114, 038101 (2015)

    work page 2015

  27. [27]

    G. W. A. Constable and A. J. McKane, Physical Review E 96, 022416 (2017)

    work page 2017

  28. [28]

    M. Xia, S. Shao, and T. Chou, SIAM Journal on Scien- tific Computing 43, A3244 (2021)

    work page 2021

  29. [29]

    T. Chou, S. Shao, and M. Xia, Applied Numerical Math- ematics 183, 201 (2023). 1 Supplementary Material Appendix A: Master equation and mean-field ODEs We consider ad-dimensional continuous-time Markovian birth-death process. Ge nerally, regulation in a birth-death process manifests itself in population-dependent b irth and death rates of each species i. A...

    work page 2023