pith. sign in

arxiv: 2606.30325 · v1 · pith:JQR7XK2Rnew · submitted 2026-06-29 · 🧮 math.PR · q-bio.MN

Thermodynamic Limits of Stochastic Chemical Reaction Networks with Phosphorylation

Lucie Laurence , Philippe Robert This is my paper

Pith reviewed 2026-06-30 05:10 UTC · model grok-4.3

classification 🧮 math.PR q-bio.MN
keywords phosphorylationchemical reaction networksstochastic CRNmass action kineticsaveraging principleequilibrium stabilityMarkov processcatalytic constants
0
0 comments X

The pith

In a stochastic model of sequential phosphorylation, certain choices of catalytic constants produce three equilibrium points of which two are stable in the large-molecule limit.

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

The paper studies a chemical reaction network modeling phosphorylation where a substrate is transformed into two phosphorylated forms by enzymes. With the substrate fixed at N molecules and enzymes scaling with N, the authors analyze the stochastic dynamics under mass action kinetics. They identify stable subsets of species that remain O(1) in number for large N and establish an averaging principle. The analysis reveals a regime of the twelve rate constants where the limiting dynamical system has three equilibria, two of which are stable. This provides insight into how multistability can arise in cellular signaling processes from the underlying reaction rates.

Core claim

The central discovery is that for specific values of the catalytic constants, the averaged deterministic system in four dimensions has three equilibrium points, two of which are asymptotically stable, leading to the possibility of the stochastic process converging to different stable regimes depending on initial conditions as N becomes large.

What carries the argument

Stable subsets of chemical species, combined with an averaging principle for the Markov process and stability analysis of the associated ODE system in R_+^4.

If this is right

  • The concentrations of species converge to one of the stable equilibria in the large N limit.
  • Different initial conditions can lead to different long-term behaviors in the system.
  • The twelve catalytic constants determine which regimes exhibit multistability.
  • An averaging principle holds for the Markov process in several regimes of the parameters.

Where Pith is reading between the lines

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

  • This multistability may correspond to biological switch mechanisms in cells.
  • Similar analysis could apply to other sequential modification networks.
  • Experimental measurement of rate constants could predict the number of stable states.

Load-bearing premise

The total mass of enzymes scales proportionally to N and a convenient initial state is chosen so that the number of copies of species in a stable subset remains O(1) on finite time intervals.

What would settle it

Simulating or observing the system for large N with rate constants in the identified regime and checking whether the process approaches one of two distinct stable concentration levels rather than a single one or diverging.

Figures

Figures reproduced from arXiv: 2606.30325 by Lucie Laurence, Philippe Robert.

Figure 1
Figure 1. Figure 1: M/M/∞ queue with input rate a and service rate b The state of an M/M/∞ queue corresponds to the state of the basic CRN ∅ a−↽⇀− b S We now state some useful technical results on the M/M/∞ queue. We begin with the following elementary lemma. Lemma 2.4. If (L(t)) is the process of an M/M/∞ queue with input rate a≥0 and service rate b>0, if L(0)=N, then there exists some η>0 and t0>0 such that (17) lim N→+∞ P … view at source ↗
Figure 2
Figure 2. Figure 2: The M/M/∞ Network [PITH_FULL_IMAGE:figures/full_fig_p014_2.png] view at source ↗
read the original abstract

In this paper we investigate the stability properties of a fundamental mechanism of biological cells called phosphorylation. The system is a chemical reaction network (CRN) for which a chemical species, {\em the substrate}, can be sequentially transformed into two phosphorylated forms, by the activity of two types of enzymes, one type for phosphorylation, the other for dephosphorylation. We investigate a stochastic representation of this model, under the mass action kinetics. The total mass of the substrate is fixed at $N$, while the total mass of enzymes scales proportionally to $N$. The asymptotic behavior, when $N$ is large, of the concentrations of all chemical species is studied. We investigate the possible {\em stable} subsets of chemical species for the kinetics of the law of mass action. A stable subset is such that, with a convenient initial state, the number of copies of the species of this subset remains $O(1)$ on any finite time interval as $N$ gets large. The role of the twelve reaction rate constants, {\em the catalytic constants} of the CRN, is investigated from this point of view. An averaging principle of the corresponding Markov process is established for several regimes of the CRN. It is shown in particular that there exists a regime with three equilibrium points, with two of them stable. The proofs of the results rely on stochastic calculus with Poisson processes, convenient couplings of subsets of coordinates of the Markov process, technical results on $M/M/\infty$ queues, and a stability analysis of a dynamical system in $\mathbb{R}_+^4$.

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

1 major / 2 minor

Summary. The manuscript investigates the large-N asymptotic behavior of a stochastic CRN modeling sequential phosphorylation of a substrate (total mass fixed at N) by two enzyme types (total mass scaling with N) under mass-action kinetics. It defines 'stable subsets' of species whose counts remain O(1) on finite time intervals for suitable initial conditions, analyzes the role of the twelve catalytic rate constants in determining such subsets, establishes averaging principles for the underlying Markov process in multiple regimes, and shows that there exists a parameter regime in which the limiting dynamical system in R_+^4 possesses three equilibria, two of which are stable. Proofs rely on stochastic calculus for Poisson processes, couplings, results on M/M/∞ queues, and stability analysis of the limiting ODE.

Significance. If the central claims hold, the work supplies a rigorous stochastic-to-deterministic link for multistability in a canonical biological motif, with the explicit construction of a three-equilibrium regime providing a concrete, falsifiable example. The scaling (enzyme mass ~N) and use of queueing-theoretic tools are well-matched to the model and constitute a technical strength; the averaging principle under varying catalytic regimes could serve as a template for related CRNs.

major comments (1)
  1. [stability analysis of the limiting dynamical system] The section establishing the three-equilibrium regime (the stability analysis of the limiting system in R_+^4): the existence of a parameter regime yielding three equilibria with two stable is asserted via the ODE analysis, but the manuscript does not supply the explicit values or inequalities on the twelve catalytic constants that realize this regime, nor does it include a self-contained verification that the stability conclusions are not sensitive to post-hoc parameter tuning. This is load-bearing for the central claim.
minor comments (2)
  1. [Introduction] The definition of 'stable subset' (Introduction) should be stated as a formal definition with the precise initial-condition and O(1) requirements made explicit, rather than described narratively.
  2. Notation for the twelve catalytic constants is introduced but not tabulated; a table listing each constant with its associated reaction would improve readability.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the careful reading and constructive feedback. We address the single major comment below and will revise the manuscript to strengthen the presentation of the three-equilibrium regime.

read point-by-point responses

  1. Referee: [stability analysis of the limiting dynamical system] The section establishing the three-equilibrium regime (the stability analysis of the limiting system in R_+^4): the existence of a parameter regime yielding three equilibria with two stable is asserted via the ODE analysis, but the manuscript does not supply the explicit values or inequalities on the twelve catalytic constants that realize this regime, nor does it include a self-contained verification that the stability conclusions are not sensitive to post-hoc parameter tuning. This is load-bearing for the central claim.

    Authors: We agree that the current manuscript asserts existence of the regime without explicit parameter values or inequalities, which limits independent verification. In the revised version we will supply a concrete set of inequalities on the twelve catalytic constants that produce three equilibria in the limiting ODE (two stable). We will also add a self-contained stability verification, including explicit Jacobian computations at each equilibrium together with eigenvalue sign checks, to confirm robustness. These additions will appear in the section on the limiting dynamical system in R_+^4. revision: yes

Circularity Check

0 steps flagged

No significant circularity

full rationale

The derivation relies on external standard tools (Poisson stochastic calculus, couplings, M/M/∞ queue results, and ODE stability analysis in R_+^4) applied to the scaled CRN under the stated mass-action kinetics and enzyme scaling. These are independent mathematical results, not self-defined quantities, fitted parameters renamed as predictions, or load-bearing self-citations. The existence of regimes with three equilibria (two stable) follows directly from the limiting dynamical system without reduction to the paper's own inputs by construction.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 0 invented entities

The analysis rests on standard assumptions of mass-action stochastic kinetics and Poisson-process representations; no new entities are postulated and the twelve rate constants are treated as free parameters whose role is investigated rather than fitted to data.

free parameters (1)
  • twelve catalytic rate constants
    The abstract states that the role of these twelve constants is investigated to determine regimes admitting multiple stable equilibria.
axioms (2)
  • domain assumption Chemical reaction network obeys mass-action kinetics in its stochastic representation
    Invoked to define the Markov process whose large-N limit is studied.
  • domain assumption Total substrate mass fixed at N while enzyme mass scales linearly with N
    Central scaling assumption stated in the abstract that enables the thermodynamic limit analysis.

pith-pipeline@v0.9.1-grok · 5811 in / 1392 out tokens · 33959 ms · 2026-06-30T05:10:12.827274+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

51 extracted references

  1. [1]

    Cellular concentrations of en- zymes and their substrates

    K. R. Albe, M. H. Butler, and B. E. Wright. “Cellular concentrations of en- zymes and their substrates”. In: Journal of Theoretical Biology 143.2 (1990), pp. 163–195

    1990

  2. [2]

    Asymptotic analysis of multiscale approximations to reaction networks

    K. Ball, T. G. Kurtz, L. Popovic, and G. Rempala. “Asymptotic analysis of multiscale approximations to reaction networks”. In: The Annals of Applied Probability 16.4 (2006), pp. 1925–1961

    2006

  3. [3]

    Monostationarity and Multista- tionarity in Tree Networks of Goldbeter–Koshland Loops

    A. Barabanschikov and J. Gunawardena. “Monostationarity and Multista- tionarity in Tree Networks of Goldbeter–Koshland Loops”. In: Bulletin of Mathematical Biology 81 (2019), pp. 2463–2509

    2019

  4. [4]

    Bistability in the chemical master equation for dual phosphorylation cycles

    A. Bazzani, G. C. Castellani, E. Giampieri, D. Remondini, and L. N. Cooper. “Bistability in the chemical master equation for dual phosphorylation cycles”. In: The Journal of Chemical Physics 136.23 (June 2012), p. 235102

    2012

  5. [5]

    Billingsley

    P. Billingsley. Convergence of probability measures . Second Edition. Wiley Series in Probability and Statistics: Probability and Statistics. New York: John Wiley & Sons Inc., 1999, pp. x+277

    1999

  6. [6]

    The regulation of protein function by multisite phosphorylation–a 25 year update

    P. Cohen. “The regulation of protein function by multisite phosphorylation–a 25 year update”. In: Trends in biochemical sciences 25.12 (2000), pp. 596– 601

    2000

  7. [7]

    In distributive phosphorylation catalytic con- stants enable non-trivial dynamics

    C. Conradi and M. Mincheva. “In distributive phosphorylation catalytic con- stants enable non-trivial dynamics”. In: Journal of Mathematical Biology 89.20 (2024)

    2024

  8. [8]

    Emergence of Oscillations in a Mixed- Mechanism Phosphorylation System

    C. Conradi, M. Mincheva, and A. Shiu. “Emergence of Oscillations in a Mixed- Mechanism Phosphorylation System”. In: Bulletin of Mathematical Biology 81 (2019), pp. 1829–1852

    2019

  9. [9]

    Catalytic constants enable the emergence of bistability in dual phosphorylation

    C. Conradi and M. Mincheva. “Catalytic constants enable the emergence of bistability in dual phosphorylation”. In: Journal of The Royal Society Inter- face 11.95 (June 2014), p. 20140158

    2014

  10. [10]

    Dynamics of Posttranslational Modification Sys- tems: Recent Progress and Future Directions

    C. Conradi and A. Shiu. “Dynamics of Posttranslational Modification Sys- tems: Recent Progress and Future Directions”. In: Biophysical Journal 114.3 (2018), pp. 507–515

    2018

  11. [11]

    Measure-valued Markov processes

    D. A. Dawson. “Measure-valued Markov processes”. In: ´Ecole d’ ´Et´ e de Prob- abilit´ es de Saint-Flour XXI—1991. Vol. 1541. Lecture Notes in Math. Berlin: Springer, Nov. 1993, pp. 1–260

    1991

  12. [12]

    Chapter 4 - Stability of digital control systems

    M. S. Fadali and A. Visioli. “Chapter 4 - Stability of digital control systems”. In: Digital Control Engineering (Third Edition) . Ed. by M. S. Fadali and A. Visioli. Third Edition. Academic Press, 2020, pp. 103–140

    2020

  13. [13]

    The Kinetic Space of Mul- tistationarity in Dual Phosphorylation

    E. Feliu, N. Kaihnsa, T. de Wolff, and O. Y¨ ur¨ uk. “The Kinetic Space of Mul- tistationarity in Dual Phosphorylation”. In: Journal of Dynamic Differential Equations 35 (2022), pp. 825–852

    2022

  14. [14]

    Feliu and N

    E. Feliu and N. Kaihnsa. Network reduction and absence of Hopf Bifurcations in dual phosphorylation networks with three Intermediates. 2024. arXiv: 2405. 16179 [math.DS]

    2024

  15. [15]

    N-site Phosphorylation Systems with 2N-1 Steady States

    D. Flockerzi, K. Holstein, and C. Conradi. “N-site Phosphorylation Systems with 2N-1 Steady States”. In: Bulletin of Mathematical Biology 76 (2014), pp. 1892–1916

    2014

  16. [16]

    M. I. Freidlin and A. D. Wentzell. Random perturbations of dynamical sys- tems. Second. New York: Springer-Verlag, 1998, pp. xii+430. 32 REFERENCES

    1998

  17. [17]

    Stochastic Models of Resource Allocation in Chemical Reaction Networks

    V. Fromion, P. Robert, and J. Zaherddine. “Stochastic Models of Resource Allocation in Chemical Reaction Networks”. In: SIAM Journal on Applied Mathematics (2026)

    2026

  18. [18]

    Gantmacher

    F. Gantmacher. The Theory of Matrices . Chelsea, 1984

    1984

  19. [19]

    Studies concerning affinity

    C. M. Guldberg and P. Waage. “Studies concerning affinity”. In: CM Forhan- dlinger: Videnskabs-Selskabet i Christiana 35.1864 (1864), p. 1864

  20. [20]

    Multisite protein phosphorylation makes a good threshold but can be a poor switch

    J. Gunawardena. “Multisite protein phosphorylation makes a good threshold but can be a poor switch”. In: Proc Natl Acad Sci USA 102.41 (Oct. 2005), pp. 14617–14622

    2005

  21. [21]

    Time-scale separation–Michaelis and Menten’s old idea, still bearing fruit

    J. Gunawardena. “Time-scale separation–Michaelis and Menten’s old idea, still bearing fruit”. In: FEBS J. 281.2 (2014), pp. 473–488

    2014

  22. [22]

    Principle of Averaging for Parabolic and Elliptic Differ- ential Equations and for Markov Processes with Small Diffusion

    R. Z. Has’minskii. “Principle of Averaging for Parabolic and Elliptic Differ- ential Equations and for Markov Processes with Small Diffusion”. In: Theory of Probability & Its Applications 8.1 (1963), pp. 1–21

    1963

  23. [23]

    M. W. Hirsch and S. Smale. Differential equations, dynamical systems, and linear algebra. Academic Press, New York-London, 1974, pp. xi+358

    1974

  24. [24]

    Multistationarity in Sequential Distributed Multisite Phosphorylation Networks

    K. Holstein, D. Flockerzi, and C. Conradi. “Multistationarity in Sequential Distributed Multisite Phosphorylation Networks”. In: Bulletin of Mathemat- ical Biology 75 (2013), pp. 2028–2058

    2013

  25. [25]

    Separation of time-scales and model reduction for stochastic reaction networks

    H.-W. Kang and T. G. Kurtz. “Separation of time-scales and model reduction for stochastic reaction networks”. In: The Annals of Applied Probability 23.2 (2013), pp. 529–583

    2013

  26. [26]

    Reduction for Stochastic Bio- chemical Reaction Networks with Multiscale Conservations

    J. K. Kim, G. A. Rempala, and H.-W. Kang. “Reduction for Stochastic Bio- chemical Reaction Networks with Multiscale Conservations”. In: Multiscale Modeling & Simulation 15.4 (2017), pp. 1376–1403

    2017

  27. [27]

    Protein phosphorylation and cellular regulation I (Nobel lec- ture)

    E. G. Krebs. “Protein phosphorylation and cellular regulation I (Nobel lec- ture)”. In: Angewandte Chemie International Edition in English 32.8 (1993), pp. 1122–1129

    1993

  28. [28]

    Analysis of stochastic chemical reaction networks with a hierarchy of timescales

    L. Laurence and P. Robert. “Analysis of stochastic chemical reaction networks with a hierarchy of timescales”. In:Journal of Statistical Physics 192.3 (2025), Paper No. 39, 31

    2025

  29. [29]

    Scaling methods for stochastic chemical reaction networks

    L. Laurence and P. Robert. “Scaling methods for stochastic chemical reaction networks”. In: Stochastic Processes and their Applications 194 (Apr. 2026), p. 104855

    2026

  30. [30]

    Stochastic chemical reaction networks with dis- continuous limits and AIMD processes

    L. Laurence and P. Robert. “Stochastic chemical reaction networks with dis- continuous limits and AIMD processes”. In: Stochastic Processes and their Applications 186 (2025), Paper No. 104643, 32

    2025

  31. [31]

    Laurence and P

    L. Laurence and P. Robert. Thermodynamic Limits of Stochastic Chemical Reaction Networks with Phosphorylation. 2026. arXiv: XXXX.XXXXX [math.DS]

    2026

  32. [32]

    D. A. Levin, Y. Peres, and E. L. Wilmer. Markov chains and mixing times . American Mathematical Society, Providence, RI, 2009, pp. xviii+371

    2009

  33. [33]

    Averaging for martingale problems and stochastic approximation

    T. Kurtz. “Averaging for martingale problems and stochastic approximation”. In: Applied Stochastic Analysis, US-French Workshop. Vol. 177. Lecture notes in Control and Information sciences. Springer Verlag, 1992, pp. 186–209

    1992

  34. [34]

    Quantifying the phosphorylation timescales of recep- tor–ligand complexes: a Markovian matrix-analytic approach

    M. L´ opez-Garc´ ıa, M. Nowicka, C. Bendtsen, G. Lythe, S. Ponnambalam, and C. Molina-Par´ ıs. “Quantifying the phosphorylation timescales of recep- tor–ligand complexes: a Markovian matrix-analytic approach”. In: Open Bi- ology 8.9 (Sept. 2018), p. 180126. REFERENCES 33

    2018

  35. [35]

    Maplesoft, a division of Waterloo Maple Inc. Maple. Version 2025. Waterloo, Ontario, 2025

    2025

  36. [36]

    Signaling switches and bistability arising from multisite phosphorylation in protein kinase cascades

    N. I. Markevich, J. B. Hoek, and B. N. Kholodenko. “Signaling switches and bistability arising from multisite phosphorylation in protein kinase cascades”. In: The Journal of cell biology 164.3 (2004), pp. 353–359

    2004

  37. [37]

    A review of the deterministic and diffusion approximations for stochastic chemical reaction networks

    P. Mozgunov, M. Beccuti, A. Horvath, T. Jaki, R. Sirovich, and E. Bibbona. “A review of the deterministic and diffusion approximations for stochastic chemical reaction networks”. In: Reaction Kinetics, Mechanisms and Cataly- sis 123.2 (2018), pp. 289–312

    2018

  38. [38]

    Martingale ap- proach to some limit theorems

    G. Papanicolaou, D. W. Stroock, and S. R. S. Varadhan. “Martingale ap- proach to some limit theorems”. In: Proc. 1976. Duke Conf. On Turbulence . III. Duke Univ. Math, 1977

    1976

  39. [39]

    A new proof of the Routh-Hurwitz stability criterion using the second method of Liapunov

    P. C. Parks. “A new proof of the Routh-Hurwitz stability criterion using the second method of Liapunov”. In: Mathematical Proceedings of the Cambridge Philosophical Society 58.4 (1962), pp. 694–702

    1962

  40. [40]

    The stability of queues with non independent inter-arrival and service times

    R. Loynes. “The stability of queues with non independent inter-arrival and service times”. In: Proc. Cambridge Ph. Soc. 58 (1962), pp. 497–520

    1962

  41. [41]

    Mathematical models of gene expression

    P. Robert. “Mathematical models of gene expression”. In: Probability Surveys 16 (2019), pp. 277–332

    2019

  42. [42]

    P. Robert. Stochastic Networks and Queues . Vol. 52. Stochastic Modelling and Applied Probability Series. New-York: Springer, June 2003, pp. xix+398

    2003

  43. [43]

    L. C. G. Rogers and D. Williams. Diffusions, Markov processes, and martin- gales. Vol. 2. Cambridge Mathematical Library. Cambridge University Press, Cambridge, 2000, pp. xiv+480

    2000

  44. [44]

    W. Rudin. Real and complex analysis . Third. New York: McGraw-Hill Book Co., 1987, pp. xiv+416

    1987

  45. [45]

    Multisite protein phosphorylation–from molecular mechanisms to kinetic models

    C. Salazar and T. Hofer. “Multisite protein phosphorylation–from molecular mechanisms to kinetic models”. In: FEBS J. 276.12 (June 2007), pp. 3177– 3198

    2007

  46. [46]

    Kinetic models of phosphorylation cycles: A system- atic approach using the rapid-equilibrium approximation for protein–protein interactions

    C. Salazar and T. H¨ ofer. “Kinetic models of phosphorylation cycles: A system- atic approach using the rapid-equilibrium approximation for protein–protein interactions”. In: Biosystems 83.2 (2006), pp. 195–206

    2006

  47. [47]

    Com- puting the Stochastic Dynamics of Phosphorylation Networks

    M. Steijaert, J. V. D. Brink, A. Liekens, P. Hilbers, and H. T. Eikelder. “Com- puting the Stochastic Dynamics of Phosphorylation Networks”. In: Journal of Computational Biology 17.2 (2010), pp. 189–199

    2010

  48. [48]

    Exploring cyclic networks of multisite modification reveals origins of information processing characteris- tics

    T. Suwanmajo, V. Ramesh, and J. Krishnan. “Exploring cyclic networks of multisite modification reveals origins of information processing characteris- tics”. In: Scientific Reports 10 (2020), p. 16542

    2020

  49. [49]

    Unlimited multistability in multisite phosphorylation systems

    M. Thomson and J. Gunawardena. “Unlimited multistability in multisite phosphorylation systems”. In: Nature 460.7252 (July 2009), pp. 274–277

    2009

  50. [50]

    150 years of the mass action law

    E. O. Voit, H. A. Martens, and S. W. Omholt. “150 years of the mass action law”. In: PLoS Comput Biol 11.1 (2015), e1004012

    2015

  51. [51]

    Molecular simulations of enzymatic phosphorylation of disordered proteins and their condensates

    E. Zippo, D. Dormann, T. Speck, and L. S. Stelzl. “Molecular simulations of enzymatic phosphorylation of disordered proteins and their condensates”. In: Nature Communications 16 (2025), pp. 4649–4665. Appendix A. Appendix A.1. End of the Proof of Theorem 3.1. 34 REFERENCES Proof. The variables (X N 3 (t), AN 2 (t), BN 1 (t)) are slow in the sense that the...

    2025